application justifies it, such as in the case of chemotherapy). .... After being subjected to specific stimuli, the stem cell is induced to duplicate its genome.
Doctorate of Medicine Thesis Title The Basis for a Rational and Integrated Immunotherapy for Advanced Breast Cancer Author Gustavo Antonio Moviglia Physician Thesis Supervisor Professor Juan Carlos Cavicchia Doctor of Medicine, Emeritus Professor of Histology and Embryology Department of Medicine National University of Cuyo Place: Maimónides University, Buenos Aires, Argentina
1
Index Page Thesis Plan
3
Introduction
5
• Bioethical Basis
9
• Internationally Accepted Standards for the Use of Innovative Therapies
11
• The Biological Basis of Neoplastic Disease
15
• The Normal Development of the Mammary Gland and the Interaction of the Mammary Epithelium with the Immune System
26
Oncogenesis, Growth, Metastatic Dissemination and the Interaction between Breast Cancer and the Immune System
42
Current Oncological Therapies
59
The Basis of Current Immunotherapies
63
Clinical Application of Immunotherapies for Breast Cancer: Our Experience
79
• Materials and Methods
79
• Results: Preparation of the Different Cell Products
96
• Results: Safety of the Immunotherapies Utilized
105
• Results: Efficacy of the Immunotherapies Utilized
114
• Results: Efficiency of the Immunotherapies Utilized
125
Proposal of Principles for the Rational and Integrated Use of Immunotherapy with other Therapeutic Forms
128
Bibliography
132
Appendix 1
167
Appendix 2
197
Appendix 3
206
2
The Basis for a Rational and Integrated Immunotherapy for Advanced Breast Cancer
Thesis Plan Primary Objective To present the rational basis for the use of immunotherapies (monoimmunotherapies, poly-immunotherapies or combined therapies) for the treatment of Advanced Breast Cancer.
Secondary Objectives 1. To establish the feasibility of these innovative immunotherapies 2. To demonstrate the validity of the therapeutic principle of these immunotherapies 3. To establish safety criteria for these immunotherapies 4. To develop an appropriate method for studying the efficacy of these immunotherapies 5. To demonstrate the efficacy of these immunotherapies 6. To demonstrate the efficiency of these Immunotherapies
Activity Plan 1. Through a survey of the literature, we will summarize the functional development and physiological regulation of the healthy organ. 2. Through a survey of the literature, we will summarize the process of neoplastic
transformation
and
dissemination
resulting
from
the
dysregulation of cellular and microenvironmental relationships in the glandular tissue. 3. Through a survey of the literature, we will summarize the local and systemic immune response to the development and maintenance of neoplastic disease.
3
4. We will describe several current therapeutic strategies. 5. We will develop a framework for the immunotherapeutic strategies used according to the results cited in the international literature and obtained in this work. 6. We will discuss the results derived from our personal experience (regarding the feasibility, safety, efficacy and therapeutic window for several immunotherapies applied to humans). 7. We will postulate criteria for the integration of immunotherapies within the framework of a combined strategy, covering the various current therapeutic strategies, to find a cure for Advanced Breast Cancer.
4
Introduction Adenocarcinoma of the mammary gland (breast cancer) has the highest incidence and prevalence rates of all cancers among women. In Argentina and the majority of Western industrialized countries, breast cancer constitutes the second leading cause of cancer-related deaths. In 2002, more than one million new cases were reported, with a five-year prevalence of four million four hundred thousand cases [1]. Currently, special emphasis is being placed on maximizing healthcare efforts toward early diagnosis of the disease because the probability of completely curing the patient is especially high in the early stages of the disease. Nevertheless, despite the efforts made in this direction and the greater understanding of the molecular basis of breast cancer biology, approximately 30% of early stage patients who have received complete standard treatment develop recurrent disease. Furthermore, 5% of patients already have metastatic dissemination at the time of diagnosis [2]. Seventy-percent (70%) of the therapeutic successes observed in the treatment of early lesions is attributable to the correct use of local (surgery and radiotherapy) and systemic therapies (cytotoxic agents, hormonal drugs, drugs directed
at
specific
metabolic
targets
(targeted
therapies)
and
immunotherapies). In general, systematic agents act efficiently on 90% of primary tumors and 50% of metastatic tumors. However, after a lapse of variable length, the neoplasm either advances (by metastatic dissemination) or recurs (locally). When either of these two situations occurs, resistance to the previously utilized therapies is not only frequent but also expected. The five-year survival rate of patients with recurrent and/or metastatic disease is roughly 20%, while the average survival time varies between 12 and 24 months, according to different prognostic parameters. In this type of presentation, the adenocarcinoma has multiple clinical manifestations, and the current therapeutic focus on advanced disease is not for curative purposes, but rather palliative, focusing on extending the quality and
5
duration of patient survival. The therapeutic responses are of short duration, and subsequent treatments possess an increasingly lower percentage of effectiveness. Thus, the toxicity and adverse effects of the diverse therapies available play an essential role in therapeutic decision-making, with preservation of the quality of life being the main factor of consideration [2, 3]. It is therefore evident that the development of new efficacious and efficient therapies for Advanced Breast Cancer constitutes an unresolved medical challenge, one that is in great need of resolution. To face this challenge, it is essential to adopt a focus that integrates different aspects of cancer biology. This focus should not only encompass molecular biology (the genetic-metabolic perspective) but also the combined interactions that generate the tumor within the host organism during its development (epigenetic perspective). In 1960, while discussing cancer cell biology in his published work "Cell Biology for Doctors", Dr. H. Prieto Díaz proposed that disease is no more than the result of the exacerbation of some normal cell processes and the repression of others. According to this concept, which remains strongly held today, pathologies do not generate a new biology, but rather they partially and disharmoniously follow biological
processes
that,
in
other
circumstances
or
species,
occur
harmoniously to solve concrete problems. Studying the molecular biology that governs the behavior of oncogenes and their interactions with epigenetic factors and with other cells, tissues, organs and systems into which they integrate supports the idea that neoplasms copy the biology of the organ from which they originated. Among the interactions that affect the aggressiveness and progression of breast cancer, our current knowledge indicates that the endocrine and immune systems are critical to these processes. A great similarity has been found between how a tissue is repaired or, in the case of the breast, how the breast suffers changes depending on a woman's sexual life, and how a tumor grows, stimulates growth and escapes immunosurveillance.
6
As the mammary gland is a secondary female sexual organ, breast adenocarcinomas are particularly hormonally dependent for their genesis, growth and dissemination [4]. The immune system acts on the normal mammary tissue during the sexual cycle, pregnancy, lactation and post-lactation glandular size decrease. During these times, the immune system induces the differentiation of mammary tissue, controls its hormonal sensitivity and remission and allows passive immunization of the infant being nursed. These complex actions depend on an interaction of reciprocal signals, where the mammary epithelium normally controls its immune system tolerance [5, 6]. Although modern therapeutics focus on the regulation of endocrine influence on breast cancer [7], few if any advances have been made on the study and use of gland-immune system interactions for the immunological management of this pathology. This oversight is most likely due to the concept expressed by Allan and colleagues [8], who consider breast cancer poorly immunogenic and therefore an unsuitable candidate for immunotherapies. Various clinical and preclinical studies support the concept that the immune system, through the elaboration of specific effector reactions, performs an important role in the control and rejection of established mammary tumors. Nevertheless, this response could be reduced by multiple inhibitory factors. In fact, many identified tumor associated antigens (TAAs) have been linked to breast tumors. Some TAAs appear to play a critical role in tumorigenesis and therefore constitute attractive targets for the development of immunotherapies [9]. Evidence also exists for the recruitment and activation of dendritic cells inside the breast adenocarcinoma [10]. The presence of these cells favorably impacts patient survival. Moreover, a surprising inverse relationship between the presence of cancer cells in the sentinel lymph node and its dendritic cell content has been observed. The presence of tumor infiltrating leukocytes (TILs) has been documented. However, the function of TILs is diminished by the secretion of inhibitory 7
cytokines by the cancer cell, the alteration of MHC (major histocompatibility complex) expression in cancer cells, co-stimulatory molecules, the aberrant expression of the Fas ligand (FasL) and an increase in regulatory lymphocytes and macrophages among the cells that comprise the TIL infiltrate (Treg and myeloid-derived suppressor cells [MDSC], respectively), among other reasons [11]. These factors allow us to glimpse part of the complexity of developing new immune therapies. It should be recognized that multiple and different cellular antigens work at the same time to generate a specific effector immune response, which controls the different immune regulation factors generated directly and/or indirectly by the tumor. To
confront
these
problems
individually
and
to
generate
an
active
immunotherapy program, we have begun utilizing an autologous tumor cell vaccine generated from the patient's cancer cells and fused with activated autologous B lymphocytes (TBH, tumor cell B lymphocyte hybrid) [12]. This vaccine, developed in our laboratory, has been used in association with activated lymphocytes, the patient's naive dendritic cells [13-14] and with two immunomodulators, cyclophosphamide and thymosin α1 [15]. We have studied their effects as stand-alone treatments and examined their synergies when used in combination. Our results have led us to propose basic concepts for focused Advanced Breast Cancer immunotherapy. The TBH vaccine was developed in its preclinical phase by Guo and colleagues [16], and the results of these studies were published in Science in 1993. Afterward, in our laboratories, we corroborated the findings of Guo using a glioblastoma animal model produced by the F98 cell line (syngeneic Fisher 344 rats). We investigated the vaccine’s mechanism of action and showed that its association with other treatments could increase the likelihood of its therapeutic success. At the 1995 European Society of Hemapheresis conference, we presented clinical results achieved with the TBH vaccine. The work was selected for publication in Transfusion Medicine, which was the official journal of that association at the time [12].
8
The study of the variables related to the therapeutic use of the TBH vaccine was performed by analyzing the data obtained through treatment, on a compassionate basis, of a group of 275 patients suffering from different types of advanced cancers. No standard therapy was available for these patients, whether it was because none existed, the existing therapy had failed or the therapy was not tolerated, thus necessitating its cessation. Of the 275 patients, it is worth highlighting the development of 50 patients with Advanced Breast Cancer. With a strictly compassionate objective, we attempted to improve their therapeutic results over time by applying different innovative therapies. From the results of that study, we proposed a rationale for making advanced breast cancer immunotherapy an accepted clinical option.
Bioethical Basis The current work was performed according to the bioethical principles outlined in the Declaration of Helsinki [17], and in particular, those principles expressed in paragraph 32. These principles have been subsequently explained in the Belmont Report [18] and the International Ethical Guidelines for Biomedical Research involving Human Subjects prepared by the Council for International Organizations of Medical Sciences (CIOMS) in collaboration with the World Health Organization (WHO) [19]. Together with the Nüremberg Code, these documents constitute the ethical basis for all innovative medical activities applied to patients. They also form part of the international pacts to which the Republic of Argentina has adhered and which have been declared constitutional in the National Legal Order. These documents establish that all therapeutic procedures that are universally accepted because of their recognized efficiency and all innovative therapeutic procedures are medical acts. Concerning the use of innovative procedures, the Declaration of Helsinki states the following, in paragraph 32: "In the treatment of a patient, where proven prophylactic, diagnostic and therapeutic methods do not exist or have been ineffective, the physician, with
9
informed consent from the patient, must be free to use unproven or new prophylactic, diagnostic and therapeutic measures, if, in the physician’s judgment it offers hope of saving life, re-establishing health or alleviating suffering. Where possible, these measures should be made the object of research, designed to evaluate their safety and efficacy. In all cases, new information should be recorded and, where appropriate, published. The other relevant guidelines of this Declaration should be followed.” Thus, these innovative procedures not only can but should be used outside of the clinical trial context when the refractory pathological condition of a patient, according to the physician's best knowledge, can be improved or even reversed through its application. This concept has given way to both the Hippocratic tradition and the tradition of Claude Bernard (1813-1878), the founder of experimental medicine, who in 1865 affirmed in his Introduction to the Study of Experimental Medicine: “…Therefore, among the experiments that may be tried in man, those that can only harm are forbidden. Those that are innocent are permissible; and those that may do good are obligatory" [20]. The Belmont Report specifies that innovative therapies should be considered for "compassionate use" and that they should be governed by three principles. First, they should be used according to the physician's best knowledge and understanding, based on scientific reasons and for the good of the patient (Principle of Welfare). Second, they should be offered to everyone who needs it (Principle of Justice). Third, they should be freely accepted by the patient (Principle of Autonomy). This last point is embodied in informed consent [18]. From these three principles, at least two important ethical consequences are inferred. First, the patient should have the liberty to refuse to continue with the treatment at any time that he or she decides, and second, the physician should be ready to suspend the therapy if he detects that the procedure, far from being of benefit, is in any way harmful to the patient. Finally, paragraph 32 of the Declaration of Helsinki states: "…Where possible, these measures should be made the object of research, designed to evaluate their safety and efficacy. In all cases, new information should be recorded and,
10
where appropriate, published." Consequently, having rigorously applied these ethical principles to the previously mentioned 275 patients, we are obligated to study the safety and efficacy of the methods we used and then share these results for the benefit of medical knowledge.
Internationally Accepted Standards for the Use of Innovative Therapies According to CIOMS (founded by UNESCO in 1947), the acceptable use of an innovative therapy requires that three characteristics first be established: safety, efficacy and efficiency [19]. Understanding the safety of a medical action signifies knowing whether it produces any undesired adverse effects, what those undesired effects are and how the therapy should be performed so that it is not dangerous (if its application justifies it, such as in the case of chemotherapy). Understanding the efficacy of a medical action signifies knowing if it can achieve the effect that it is meant to achieve. For example, if a substance or product is administered to fight cancer, it should be capable of stopping tumor growth, diminishing tumor size or causing the tumor to disappear (what we call tumoral response). Understanding the efficiency of a medical action signifies knowing if the therapeutic effect achieved is capable of modifying the natural history of the disease on which it acts and of improving or matching the results of pre-existing treatments, but with greater benefit to the patient (lower complexity, better quality of life, decrease in associated toxicity and lower cost, among others). Moreover, the substance or product capable of stopping tumor growth should also be useful in extending the patient's survival, and/or generating a lower toxicity than pre-existing treatments or allowing a better quality of life. The natural history of disease refers to the development of a condition without the presence of medical interventions including forms of presentation, successive stages, survival time, sequelae, degree of disability and so forth. By extension, a patient correctly treated with the best standard treatment available can also be considered part of the "natural history".
11
How Safety, Efficacy and Efficiency Data are Obtained An understanding of the safety, efficacy and efficiency of a medical procedure is achieved through its controlled use in people who freely (with full knowledge of the novel character of the therapy, its potential risks and without any type of coercion) decide to participate. As defined by the Declaration of Helsinki, the Belmont Report and CIOMS [1719], the application of a new therapeutic remedy can be achieved in two distinct ways: as a medical action attempting to save the life or improve the pathological condition of those who have no other option (compassionate use), or within the framework of a clinical trial with the fundamental objective of obtaining knowledge (research). Almost all medical actions have been accepted into standard practice through the analysis of clinical results observed in patients treated for compassionate use. Recently, it has become acceptable for academic research activities called clinical trials to be conducted, particularly for the use of drugs. As opposed to compassionate use, clinical trials have the primary objective of obtaining scientific data on a drug rather than patient improvement, a usage that is fundamentally linked to the interests of the pharmaceutical industry. Classical examples of the acceptance of therapies based solely on the analysis of clinical results obtained include the use of blood transfusions, treatment of infections with penicillin, the development of laparoscopic surgery and bone marrow transplant, among others. During World War II, injured soldiers often bled significantly and acquired infections. The development of transfusion techniques and the use of penicillin for treating these patients were guided only by good medical judgment. When the military conflict ended, and after having saved thousands of lives through the application of these therapeutic measures to those who had no other option, the safety, efficacy and efficiency results of these therapies were published. Something similar occurred with the use of laparoscopic surgery. Its acceptance as common practice was not mediated by a clinical trial, but rather by the analysis of clinical results of those who had been treated on a compassionate basis with this methodology. Here, neither the urgency of war nor a fear for the
12
patient’s life was a factor. Laparoscopic surgery was attempted and achieved, with well-known techniques and indications, using an innovative technology to improve the associated toxicity and recovery time of classical surgical techniques. In contrast, and in compliance with the guidelines established by CIOMS, clinical trials aimed at considering the application of a drug or vaccine meant for bacterial and/or viral prophylaxis must undergo three phases of analysis following the preclinical phase (a study in laboratory animals and in vitro): phase I is designed to collect data related to the safety of the therapy, phase 2 tests for efficacy and phase 3 establishes efficiency. Phase 4 trials are considered when the efficiency of a medication or an antibacterial or antiviral vaccine is evaluated for an application different from the application originally described. According to international standards as suggested by governmental agencies in countries where strict health surveillance is practiced and by internationally recognized statisticians and epidemiologists [23], the necessary number of subjects from which data should be collected to assure a substantial and valid study is between 9 and 25 patients when reporting on safety data, between 20 and 200 patients for efficacy data and a variable number between 30 and thousands of patients for data relating to efficiency. The extensive range of variability in the number of patients required to establish these characteristics, particularly the efficiency of a drug or therapeutic method, rests on the degree of observed benefit. Statistically, a greater number of study cases are required when the resulting benefit is lower with respect to the natural history of the disease and/or of the pre-existing treatments for a particular condition. Thus, it was only necessary to compare three groups of 111 patients each to verify that interferon beta-1b lowered the progression rate of multiple sclerosis by 30% in its clinical relapse-remission phase in patients treated with high doses compared to those who were not treated [21]. However, 2800 patients were required to study the clinical results of advanced stage pancreatic cancer treated with gemcitabine (Gemzar) because the
13
difference in prolonged survival was only five weeks longer than that achieved with the previously existing therapeutic, 5-fluorouracil (5-FU). The average survival achieved by the 1400 patients treated with 5-FU was only four months and two weeks, while the average survival of the 1400 patients treated with Gemzar was five months and three weeks [22]. Once the health professional has collected safety, efficacy and efficiency data, these data should be adequately communicated to the scientific community according to international ethical agreement recommendations. The most appropriate method is to submit it for public discussion by publishing the results in journals with scientific editorial committees (peer review). An alternative to publishing is to present the data at international conferences, where the evaluation of research data is performed with the same rigorous selection methods. The acceptance of a study report into these types of journals or the presentation of the data at scientific conferences such as those previously mentioned confers scientific validity to the clinical observation. Regulating agencies such as the FDA, the health authorities of the European Union, the Swiss Confederation or Japan consider reports of both the clinical trial of cases treated on a compassionate basis and those obtained through the different phases that characterize clinical trials. In the case of clinical trials, a rigorous order should be followed. First, a phase I trial should be conducted to establish product safety and to obtain an approximate idea of the dose, or therapeutic range, of a drug or procedure. If this initial phase is approved, phase II is authorized to confirm the proposed mode of action, as well as to adjust the dose and the mode of application of the substance or therapy. Finally, phase III is authorized, in which the efficiency of the proposed drug or treatment is evaluated. An intermediate path rests at the point where the contributing data justifies commencement of a phase II or phase III trial. This is always assuming, of course, that the data observed are sufficiently in line with the data obtained in prior phases.
14
The Biological Basis of Neoplastic Disease Definition of Neoplasm Neoplasms originate from cells with reproductive potential that, through some genetic alteration, become dysregulated in terms of their growth and differentiation. When these cells undergo cell division, they generate malignant tumors that are composed of poorly differentiated cells, have an unregulated growth pattern and are capable of metastatic spread [24].
Figure 1: Synthesis of proteins – genetic regulation
Growth and Cell Differentiation For growth to occur, an increase in cell-specific protoplasm is required. Specific protoplasm is the accumulation of unique cellular structures and proteins that characterize the cell's function. The increased activity of specific genes, as indicated by an increase in corresponding RNA production, is the first sign of cell growth (see Figure 1). This increase in protein synthesis involves increasing the cell size without increasing the quantity of DNA. In cell biology, this situation is defined as hypertrophy [24]. In the presence of persistent stimuli promoting cell hypertrophy, the cell first responds by amplifying previously activated genes and second by duplicating
15
the genetic material (DNA synthesis). When the cell has duplicated its DNA, it may a) remain as it is without dividing its nucleus or cytoplasm, thus generating a tetraploid cell; b) divide only its nucleus, but without dividing its cytoplasm, thus generating a binucleated cell; or c) divide into two daughter cells. In these three cases, tissue hyperplasia occurs [24]. Although all of the nucleated cells in a tissue are capable of hypertrophy, only a small group of cells, usually the more immature ones, can duplicate their DNA and generate tissue hyperplasia. Among these cells are "tissue stem cells". Stem cells are cells that possess the ability to reproduce, generating two different cell types: one that conserves the characteristic of continuous (almost indefinite) division and the second that begins to mature (differentiates in a determined direction). This second type of cell can generally give rise to a limited number of generations of daughter cells [24].
Figure 2: Phases of the cell cycle
Stem cells are closely related to a second type of cell that acts by way of nursemaid cells. These cells, together with certain elements of the extracellular matrix, configure a peculiar microenvironment called a cell niche. This cell niche 16
determines when a stem cell should be reproduced and when it should remain in a quiescent state [25]. When a stem cell divides into two daughter cells, one permanently remains in the niche, while the other begins to differentiate into a cell type pertaining to the tissue that it forms. Because of this double descendant pathway in which one cell remains as a cell reserve and another begins its pathway toward cell differentiation, stem cell division is termed asymmetrical. Cell differentiation is understood as the process by which a stem cell daughter progressively acquires changes to its function and morphology to become a specialized cell. These changes correspond to the silencing of active stem cell genes and to the activation of other previously silent genes [24].
Figure 3: Crypts of Lieberkühn stem cell niches
When cells reproduce, they cyclically pass through highly specialized phases
17
that together form the "cell cycle" (see Figure 2) [24]. Under appropriate conditions, cells undergoing this cycle remain at rest (quiescent state), and the cells that are found in this phase are said to be in G0. After being subjected to specific stimuli, the stem cell is induced to duplicate its genome. The phase during which the cell is prepared to duplicate its DNA is called G1 (from gap: gap, interval), and the subsequent phase during which this genetic material is synthesized is called the S phase (synthesis). The cell later repairs DNA transcription errors that occurred during the S phase and prepares for the complex phenomena of mitosis (which first involves division of the nucleus and then of the cell cytoplasm). This phase is called G2. The final phase is called the M phase, in which mitotic division occurs [24]. After the process of differentiation, the cell gradually loses the ability to divide indefinitely and instead develops functions that allow the tissue that it forms to be physiologically efficient. Thus, the hematopoietic progenitor cells give rise to the different elemental forms of blood, intestinal epithelial progenitor cells situated in the crypts of Lieberkühn and to the highly specialized enterocytes of the intestinal villi [25-26].
…… Figure 4: Cellular Progeny of haematopyetic and intestine cells
Differentiating cells pass through four phases: proliferation, indeterminate phase, maturation and complete differentiation. During the proliferation phase, the first daughter cells that begin to differentiate conserve the possibility of dividing. Under appropriate stimuli, these cells can
18
re-acquire stem cell characteristics. In the indeterminate phase, the cells lose their ability to re-convert to stem cells, but they continue dividing and differentiating. During maturation, the cells lose their ability to divide, but they continue to differentiate. Finally, in the cell differentiation phase, the already differentiated cells remain in this state until their death by apoptosis.
Genetic and Epigenetic Regulation In the previous section, we mentioned that proteins are responsible for constituting the functional protoplasm of the cell. We should clarify that we can classify proteins by their function as structural, enzymatic or those that possess both qualities. These three types of proteins are therefore the genetic effectors that determine the morphology of the cell and the totality of cellular functions. Each one of an organism's somatic cells possesses all of the genes that regulate all of the possible proteins of that multicellular organism. The silencing of certain genes and the activation of others determines which types of proteins the cell will synthesize and is therefore the process by which cell diversity is generated. In other words, the same stem cell can produce neurons, fibroblasts, lymphocytes, epithelial cells and so forth by this process. What determines the cell differentiation pathway of stem cells? Currently, the accepted hypothesis is that during embryonic development and adult life, the cellular microenvironment influences the undifferentiated cell, causing the cell to take specific pathways of differentiation. These signals are emitted by the extracellular matrix in which the undifferentiated cell is supported, the cells with which it enters into contact, the humoral factors that flood the microenvironment in which it is immersed and its functional requirements [24]. These signals trigger a cascade of intracellular molecular reactions that end with the silencing of certain genes and the activation of others. The cascade that determines DNA differentiation is known as epigenetic regulation.
19
Epithelial-Mesenchymal Transition To understand how these cellular relationships are established, how they determine the architectural-functional organization for each specific organ, how changes related to postnatal functional maturation are produced, how tissue damage is repaired and how from the alteration of these relationships pathologies such as fibrosis and cancer can occur, we will refer to the process of Epithelial-Mesenchymal Transition (EMT) and its inverse, MesenchymalEpithelial Transition (MET). One of the most primitive phenotypic divergences occurring in early embryonic development is the distinction between epithelial cells and maesenchymal cells. Epithelial cells provide the cohesion in cell-cell unions that is essential for the maintenance of the integrity of a multicellular organism. These cells function as a necessary barrier to establish and maintain an internally controlled environment that is independent of the external environment [27]. In mammals, epithelialization occurs during the first stages of embryonic development, during blastula compaction [28, 29]. A short time later, gastrulation gives rise to mesenchymal formation [30]. The cells possessing the mesenchymal phenotype give structure to the epithelial cells through the production of the extracellular matrix. As opposed to epithelial cells, which possess very little movement, mesenchymal cells are very mobile, being displaced throughout the tissue [31]. The phenotypic interconversion between epithelial and mesenchymal cells is termed the EMT, and the inverse process is termed the MET. These processes provide
the
phenotypic
flexibility
that
is
essential
during
embryonic
development, tissue restructuring, healing of injuries and the regeneration of completely differentiated tissues [27, 32]. EMT is also observed in pathological processes such as fibrosis and cancer [33]. The majority of signaling pathways and transcription factors used in physiological EMT processes are activated in pathological EMTs. This is particularly true for tumor invasion and its metastatic dissemination [34, 35]. During the physiological process of EMT, a series of highly specialized events occurs in coordination (Figure 5) [27]. These events begin with the loss of 20
apicobasal polarization when the tight junctions (zonula occludens) that fix epithelial cells to each other are dissolved. In this way, the membrane proteins that are suspended in the bi-laminate lipidic fluid that forms this structure can freely mix [36]. Subsequently, the adherens junctions (zonula adherens) and the communicating junctions (gap junctions) begin to deteriorate. Finally, the underlying basement membrane begins to degrade [37]. Cell surface proteins such as E-cadherin and integrins mediate the connections between neighboring epithelial cells and with the basement membrane. However, they are replaced with N-cadherin and another group of integrins that provide the cell with transitory adhesion properties that the mesenchymal phenotype requires.
Figure 5: Cellular processes and the molecules implicated in the EMT process [332]
As part of this transition, the elements of the cytoskeleton reorganize. Peripheral actin molecules are replaced by stress proteins, while the cytokeratin intermediate filaments are replaced by vimentin filaments. These cell membrane and cytoskeletal changes are responsible for the transformation of cuboidal epithelium into fusiform epithelium. Lastly, the cell acquires invasive properties and the ability to move across the extracellular matrix without needing to contact any other cell. Presumably, during this process, the epithelial cell acquires anoikis resistance (they resist entering into apoptosis due to losing contact with the basement membrane) and the ability to respond to extracellular signals that precisely guide their migration until they arrive at their destination. Upon arrival, mesenchymal cells perform the inverse process (MET) [38]. Thus, during the course of development, the progenitor cells pass through different stages of EMT or MET one or more times until they produce the diversity of organs that constitute our body. In cancer, EMT characteristics have been observed in breast [39], ovarian [40], 21
colon [41] and esophageal [42] neoplasms. Oncogenic EMT is associated with the loss of apicobasal polarity [43], the disintegration of tight junctions [44] and of the zonula adherens, changes in the cytoskeleton, including the decrease in cytokeratin production and the increase of vimentin synthesis [45] and the acquisition of a mobile and invasive phenotype. These changes are parallel to those previously described for the physiological EMT process that occurs during development. Numerous initiators of EMT in neoplastic cell lines have been identified, including transforming growth factor-β (TGF-β) [46], Wnt [47], Snail/Slug [48, 49], Twist [50] and Six1 [51, 52]. These initiators are also critical during physiological EMT [53-56]. In the context of carcinomas, EMT provides a coherent and comprehensive explanation for the invasion of surrounding tissues and lymphatic blood vessels and of metastatic dissemination (Figure 6). Moreover, it has been clinically shown that the presence of EMT regulators in cancerous cells is directly related to poor prognosis and the aggressiveness of the tumor [57, 58].
Figure 6: Summary of the main EMT-MET processes occurring during oncogenesis, tumor invasion into adjacent tissues and metastatic dissemination [332]
22
Epigenetic Regulation and Cancer DNA methylation is the most well known epigenetic process, participating in the regulation of gene expression in two ways: directly, by impeding the union of transcription factors, and indirectly, by inducing the "closed" chromatin structure. In humans, DNA methylation occurs on the cytosine molecules that precede guanine molecules. These nucleotide pair sequences are called CpG dinucleotides, and DNA contains many 1000-1500 base pair (bp) CpG dinucleotide rich regions ("CpG islands") that are recognized by DNAmethyltransferases. During DNA replication, these enzymes methylate the number 5 carbon of cytosine in the recently synthesized chain; the "memory" of the methylation state is located in the DNA daughter chain (Figure 7).
Figure 7: DNA methylation
It is accepted that methylation is a unidirectional process; thus, when a CpG 23
sequence is methylated de novo, the modification is stable and is inherited as a clonal methylation pattern. However, the loss of genomic methylation (hypomethylation) as a primary event is frequently associated with the neoplastic transformation process, which is proportional to disease severity. DNA methylation occurs in the context of the chemical modification of nuclear proteins called histones. Histones are not simply DNA packaging proteins, but rather molecular structures that participate in the regulation of gene expression. They store epigenetic information through post-transductional modifications such as lysine acetylation, arginine and lysine methylation and serine phosphorylation. These modifications affect gene transcription and DNA repair. It has been proposed that different histone modifications give rise to a "histone code". The acetylation of histone lysines, for example, is generally associated with the activation of transcription. The functional consequences of histone methylation depend on the type of residue, lysine (K) or arginine (R), and the specific site that the methylation modifies (i.e., K4, K9 or K20). The methylation of H3 at K4 is connected to transcriptional activation, while methylation of H3 at K9 or K27 and of H4 at K20 is associated with transcriptional repression. What emerges from these findings is a flexible but precise pattern of DNA methylation and histone modification that is essential for the physiological activity of cells and tissues. The genomes of progenitor, pre-neoplastic and cancerous cells of patients share three important changes regarding their degree of methylation. These changes are considered early developmental events in some tumors. The first change is heterochromatin hypomethylation, which drives genomic instability and increases mitotic recombination events. The second change is the hypermethylation of individual genes. The third change is the hypermethylation of the CpG islands of constituent genes and tumor-suppressor genes. The processes of methylation and demethylation can occur individually or simultaneously. In general, hypermethylation is involved in gene silencing and hypomethylation in the over-expression of certain proteins involved in invasiveness and metastasis [59]. The loss of methylation is principally due to the hypomethylation of repetitive DNA sequences and to the demethylation of coding and intron regions, regions 24
of DNA that allow for the transcription of alternative versions of messenger RNA (mRNA). This lack of epigenetic control in tumors can be reverted if the tumor cells are cultivated in an adequate environment. Thus, various authors have shown that if human tumor lines are cultured in a young or even embryonic healthy stroma, the neoplastic cells acquire a normal phenotype [60]. A similar phenomenon has been observed in vivo through the transfer of tumor stem cells to the fetus during pregnancy. These cells seldom develop tumors in the descendant, although the opposite phenomenon, the transfer and persistence of mesenchymal cells from the fetus to the mother has shown to be a cause of autoimmune disease and neoplasms [61]. In light of these findings, tumorigenesis results not only from genetic alteration of a cell with proliferative capacity, but it may also be associated with epigenetic dysregulation of the elements that constitute the cell environment (extracellular substance, contact with other cells, humoral and microenvironmental factors and other functional requirements). Beginning with the aforementioned discussion, a new definition of neoplasm can be inferred, where a dynamic relationship of genetic and epigenetic dysregulation exists, originating primarily in the cellular environment that induces expression, over-expression and repression of various groups of genes. This gives rise to the formation of a tissue whose growth, morphology and function follow anarchic laws. This neo-organ, utilizing organic homeostasis methods, progresses at the expense of the rest of the organism, and gives rise to the emergence of neoplastic disease.
25
The Normal Development of the Mammary Gland and the Interaction of the Mammary Epithelium with the Immune System
Growth, Differentiation and Involution of the Mammary Gland during the Female Sexual Cycle Figure 8 summarizes the main characteristics of the adult mammary gland structure.
Figure 8: Structure of the mammary gland
In contrast to what occurs with the majority of the body’s organs, most of the organogenesis of this complex structure occurs after birth, particularly during pubertal development. For certain aspects of the secretory epithelium, organogenesis is only realized after the involutive period that occurs at the end of lactation, following the first pregnancy. From the moment that inherent development during the pubertal period is completed, the entire gland, similar to what occurs with the uterus and ovary, is continuously reconstructed because of the cyclic variations of reproductive hormones [62].
26
Epithelial Plasticity of the Mammary Gland The mammary gland develops in three stages. The first stage of embryonic development involves the development of the gland's primary rudiments. The second stage occurs during pubertal development and involves the elongation and branching of the glandular ducts. A third stage occurs during the first pregnancy, where third order branching of the glandular ducts associated with secretory differentiation of the glandular alveoli occurs (Figures 9 and 12) [63].
Figure 9: Concurrent processes involving EMT in the normal development of the mammary gland during oncogenesis and metastatic dissemination [332]
The development of the mammary gland begins in the sixth week of gestation with the emergence of the milk line formed by specific placodes that develop along that line (Figures 10 and 11). The placodes invaginate inside the underlying stroma to form the mammary bud. The mammary bud begins to invade the fatty tissue that surrounds it, later producing a glandular rudiment 27
that is formed by the first generation of glandular tubes (ducts); the rudiment remains in this state until puberty.
Figure 10: Embryonic development of the mammary gland
During puberty, the ducts elongate and branch into the mammary adipose panicle. This is how the gland remains until pregnancy when the alveoli develop so that milk is produced (Figures 9 and 11). During the second stage of glandular development, the terminal end buds (TEB) are formed (Figure 13) [63]. The TEB are formed by several layers of epithelial cells placed at the end of the alveolar ducts. From this stage, the alveolar ducts are composed of structures formed by two epithelial laminas: the luminal epithelia and the lamina composed of myoepithelial cells [64]. TEB bifurcation occurs under the regulation of numerous extracellular epithelial plasticity regulation inducers and EMT inducers, such as epidermal growth factor (EGF) and hepatocyte growth factor/scatter factor (HGF/SF), and the activation of proteases such as matrix metalloproteinase (MMP) [65]. Although the cells of the TEB do not reveal a complete mesenchymal phenotype nor lose their junctions with other cells, various signals confirm that this is an EMT process. The cells lose their apicobasal polarity, as shown by the uniform distribution of β-catenin on their membranes, their normal location on the basolateral surface and the protein-kinase C-ζ located on the apical surface
28
[66]. Thus, TEBs secrete proteases such as MMP3 to thin out the underlying basement membrane [64], activate the Msx2/Cripto-1 nuclear induction pathways [67, 68] and express their own group of integrin receptor proteins (α2, α3 and β1) in the extracellular matrix [64, 69]. TEB gene expression studies using “microarray" analyses, identify known EMT regulators: Snail and Twist [70]. During the embryonic formation of the breast, the mesodermal and ectodermal tissues of the milk line may interact. These mutually inductive processes, far from being simple and only confined to a single moment during development, are complex and continue to progress during all stages of development. Reciprocatively and sequentially, both embryonic tissues epigenetically modify their gene expression until arriving at the glandular epithelium conformation, which is highly specialized, hormonally dependent and capable of nourishing and of offering passive immunity to the child during the first six months of life. In the six-week-old embryo, a bilateral fattening of the mesoderm called the milk line appears (Figure 10, indicated as mammary ridges). Under the influence of this particular type of mesoderm, the ectoderm grows toward the interior of the mesenchyme, similar to the method of vegetative germination [71-72] (Figures 10 and 11). The influence of the mesenchyme on the mammary lines with respect to organ differentiation is homologous to that of the notochord and neural tube development. The signals emitted by the mesenchyme induce the proliferation and differentiation of the adjacent ectodermal epithelium. Figure 11 summarizes some of the changes in phenotypic expression that both epithelial and stromal tissues experience during glandular development [72-75]. In humans, the growth of the mammary gland begins in puberty. In this phase of development, the parenchymal cells branch from the blind ends of the primary and secondary ducts in an elaborate tree with conduits and multiple lobes. During each menstrual cycle, the number of mammary epithelial cells that proliferate varies, reaching its maximum proportion during the luteal phase, when the epithelial growth fraction is roughly 35% [76, 77] (Figures 9 and 12).
29
Figure 11: Embryonic development of the mammary gland (relationship with the menchymal according Gertraud W. Robinson Nature Reviews Genetics 2007; 8:963-972
Figure 12: Different Functional Moments of the Mammary Gland
During pregnancy, the number of alveoli per lobule is increased ten times and new lobules with similar characteristics are formed from the growth and branching of the existing terminal ducts. All this occurs at the expense of the connective tissue, which is reduced to its minimum expression in the interalveolar space [78].
30
Figure 13: Primitive Ducts and stem cells nich structure at branching
Sternlicht M et al
Differentiation. 2006; 74(7): 365–381
Based on cell kinetics studies, Taylor-Papadimitriou et al [79] proposed the existence of different stem cell populations in the adenomeres of the adult mammary gland. This was corroborated by Villadsen et al [80] using immunohistochemical staining techniques. These authors described three different types of epithelial stem cells: those that are located in the primary ducts (in their lateral ends, see black arrows in Figure 10), those of second and 31
third order ducts (in the terminals or lateral ends, Figures 13 and 14) and those of the alveoli.
Figure 14: Histochemistry characterization and topographic localization of stem cells in the primary and secondary of ducts as well as alveoli [80]
The reproductive capacity of these different cells progressively diminishes to the extent that, during their growth, they move away from the primary ducts. To return to the previous concepts regarding the life cycle of the cell, it is confirmed once more that the reproductive potential of each population of progenitor cells diminishes as they settle into regions of greater epithelial differentiation. This spatial disposition of the mammary gland epithelial cells offers a mirror image of what is observed in the intestinal epithelia. In the intestine, it is 32
observed that the reproductive capacity of the intestinal absorption epithelial cells diminishes and their degree of differentiation increases as they move away from the bottom of the crypt of Lieberkühn and approach the free surface of the intestine. In the mammary epithelium the inverse occurs, with the reproductive potential diminishing and the degree of cell differentiation increasing as cells move away from the nipple (compare Figures 3 and 4 with 13, 14 and 15).
Figure 15 represents the different stages of maturation corresponding to stem cell differentiation in the primary ducts [81].
These cells do not possess estrogen receptors (ER-negative). After a first asymmetric division, one of the daughter cells (which continue to be pluripotent) becomes ER-positive. These hormone dependent cells secrete paracrine factors that regulate the activity of ER-negative stem cells. Pluripotent epithelial cells can differentiate into three different cell lineages: the basal epithelial progenitor (green cells in Figure 15), the luminal duct epithelial progenitor (dark blue cells in Figure 15) and the alveolar epithelial progenitor (gray cells in Figure 15).
33
Stimulation with estrogens during the development of the ducts activates the GATA 3 gene. This could determine the destiny of the luminal epithelia (alveolar or ductal). On the other hand, the activation of the ΔN-p63 gene could determine differentiation into myoepithelial ductal cells. During pregnancy, prolactin activates, in combination, the GATA 3, Stat 5 and C/EBP genes, which stimulate the alveolar progenitor cells to differentiate into alveolar cells. These cells, through the activation of the Elf5 gene, begin to secrete milk. In contrast, activation of the ΔN-p63 gene converts the alveolar progenitors into myoepithelial alveolar cells. For years, it has been thought that only the direct action of sex hormones could explain all of the changes needed for the mammary epithelium to perform its fundamental function of secreting milk. However, this belief has been contradicted by very simple evidence that will be explained in the next section.
34
The Function of Myoepithelial Cells and Mammary Stroma as Inducers of the Development and Function of Epithelial Parenchyma From its embryonic origin, the mammary gland develops through the complex and reciprocal relationship between the ectoderm and the underlying mesenchyme. The processes of mutual, sequential and reciprocal differentiation continue throughout postnatal development so that the mesenchyme and ectoderm are transformed into the stroma and the glandular tissue, respectively. The cellular interactions described are essential for the proper changes to occur in the gland during pubertal development, every sexual cycle, pregnancy, lactation and post-lactation glandular involution. Disharmonies in this mutual induction allow malignant breast tumors to develop. Myoepithelial cells, stromal cells and the basement membrane proteins of adenomeres are the elements that form the stem cell niche that controls the proliferation and differentiation of the various progenitor cell populations of the mammary gland. Various experiences have demonstrated the veracity of these concepts. It has been observed that the first mammary epithelium cell cultures grown in plastic bottles without the complex hormonal mixtures and other humoral factors present to enrich the medium never managed to produce milk. However, after culturing these same cells in a three-dimensional reconstructed basement membrane matrix (Matrigel), they acquired a similar phenotype to the secretory mammary epithelia [82, 83]. Le Beyec et al [84] have shown that under these conditions, the three-dimensional relationship with other cells as well as contact with the basement membrane induce a rapid deacetylation of the H3 and H4 histones associated with an overall decrease in transcription. While the majority of the luminal epithelial cells in the normal mammary gland are not in direct contact with the basement membrane, they rest on myoepithelial cells (Figure 13). Luminal epithelial cells are those that produce the basement membrane (BM), which is the structure that physically divides the stromal compartment from the epithelial compartment.
35
In co-cultures of epithelial and myoepithelial cells, a spontaneous regrouping of both populations has been observed to form spheres composed of an internal layer of epithelial secretory cells surrounded by an external layer of myoepithelial cells. Within both cell populations, the BM's own extracellular elements appear (laminin 1 and 5) [85] and the epithelial cells become polarized after the spatial reorganizing described. This has been shown through immunohistochemistry, as the sialomucin (MUC1) and occludin molecules are only detected on the apical border, while epithelial specific antigen (ESA) and integrin b-4 can only be detected in the basolateral plasma membrane [83]. Regarding the relationship between the adenomere and the stroma, they work to signal four principle elements, the basement membrane (referenced in the previous two paragraphs), fibroblasts, adipocytes and immune system cells. During embryogenesis, the mesenchymal cells of the mesoderm in the milk line can determine the sexual phenotype of the gland, and the embryonic adipocytes [63, 86] specifically control the development of the mammary ducts (Figure 16).
Figure 16: Some of the interactions between mesenchyme and endoderm during the first stages of embryonic development. Note the sequential activation of different genes since day 11 to day 18. [63]
During pubertal development (Figure 17), the mammary stromal fat adipocytes support the expansion of the mammary duct tree (most likely influenced by the role they play in sexual hormone metabolism). In contrast, the fibroblasts of the
36
mammary stroma negatively regulate ductal growth through extracellular substance proteins, such as collagen I and chondroitin sulfate, induced by TGFβ from the environment. The direct contact of epithelial cells with these two proteins would activate them to begin apoptosis [86]. In adults, evidence obtained from tissue culture shows that the signals required for the induction of tissue-specific differentiation during pregnancy and the maintenance of function during lactation originate in the basement membrane. Beta-casein synthesis can be induced in cell cultures of pure mammary epithelial cells if they are absorbed inside a reconstituted basement membrane matrix. This induction occurs through a dependent integrin pathway. The degradation of the basement membrane matrix leads to the loss of casein production. The influence of the stroma on epithelial differentiation has been shown to induce spermatogonial cells [87] and neuroblasts [74] to transform into mammary epithelium. Moreover, the stroma is capable of altering the breast cancer cell phenotype to a more benign phenotype [73, 75].
Figure 17: Some of the interactions between mesenchyme and endoderm during the post natal development. Note the sequential activation of different genes during puberal development and pregnancy and secretory stages. [63]
In other words, it can be confirmed that, after normal stromal cells make contact with malignant cells, the extracellular substance and the soluble extracellular factors induce the trans-differentiation of cells whose genomes have already initiated another pathway of differentiation or have lost regulation. 37
Given the cellular interactions described, induction of the development and regression of the ducts still cannot occur without the assembly of a third element, the cells of the immune system.
Figure 18 Influence of the neighborhood relationship of to adjacent cells in the mammary epithelium differentiation [63]
In fact, Gouon-Evans et al [88] showed that the presence of macrophages and eosinophils was essential for the terminal end buds (TEB) of the epithelial ducts to grow and branch inside the stroma. This also occurs with the phenomena of regression, where TGF-β, a substance secreted by Th2 lymphocytes and Treg cells, plays an essential role in inducing fibroblasts to synthesize collagen I and chondroitin sulfates. Recently, Watson and his team [49, 90] have shown that the genes that 38
regulate the different states of epithelial cell differentiation are the same that regulate the differentiation of Th0 lymphocytes into Th1 or Th2 lymphocytes. The undifferentiated epithelium possesses activated Th1 genes and secretes its own cytokines from these lymphocytes (interferon gamma, interleukin 12 and TNF-α). The epithelium that differentiates to secrete milk under the action of active prolactin, on the contrary, possesses Th2 (Stat6) genes and secrete IL-4, IL-5 and IL-13 that, through paracrine signaling, activate neighboring cells. As observed in Figure 18, Watson [63] has summarized how contact with a properly activated cell, which according to Guon-Evans possesses macrophage M2 characteristics, and the secretion of estrogens and prolactin generates certain epigenetic factors. These factors act on undifferentiated epithelial cells, causing them to differentiate into secretory cells. These cells, together with the immune cells, then induce the rest of the ductal epithelial cells. Furthermore, during post-lactation involution, apoptotic epithelial cells activate two genes that induce an acute inflammatory response, the transcriptioninducing factor Stat3 and the nuclear factor NFkB. This inflammation would attract necessary macrophages and lymphocytes to the gland for restructuring and the prevention of opportunistic infections. The active immune system stimulates post-natal maturation of an organ, as well as its subsequent repair and the maintenance of its function, which seems to be a phenomenon occurring in many of the body’s organs. This phenomenon has been described through the joint effort of two teams, those of Cohen and Schwartz, from the Weizmann Institute in Israel. At the same time that Cohen developed the model of experimental allergic encephalitis and discovered an infinite amount of auto-aggressive T lymphocyte clones against nervous tissue, Schwartz established that the post-injury negative inflammatory reaction in mammals was what impeded tissue repair [91, 92]. Both teams had the suspicion that, beyond the apparent contradiction of their observations, a natural relationship existed between both systems. The defect or the partial exacerbation of the interaction that was established during an injury gave rise to a repair defect or to an autoimmune reaction, respectively. To test this presumption, both groups of investigators united. They generated
39
traumatic spinal cord injuries in mice and infused them with anti-myelin autoaggressive syngeneic lymphocytes that had been cloned in Cohen's laboratory. These lymphocytes, infused into intact animals, penetrate the brain and induce an autoimmune reaction similar to what is produced in the demyelinating plaques of multiple sclerosis. In the injured animals, however, the lymphocytes were selectively directed to the injured area where they increased local inflammatory reactions several times, but did not produce systemic symptoms. Within fifteen days, the animals walked again with their four legs, although they did so with difficulty. Hohlfeld’s group from the Max Planck Institute in Germany offered a crucial and illuminating contribution to this work. They showed that upon being exposed to specific antigen, the anti-myelin auto-aggressive lymphocytes secreted neurotrophins. Specifically, upon arriving at the damaged region of the CNS, the lymphocytes not only deployed their cytolytic activity but also produced growth and maturation factors from distinct elements of the nervous tissue [93]. With these elements, the concept of "protective autoimmunity" was forged and published in Nature Medicine in January 1999 [92]. At our Institute, we decided to study whether a direct relationship existed between
the
reparative
capacity
of
the
auto-aggressive
lymphocytes
responsible for the phenomena of protective autoimmunity and MSCs. To do this, we isolated auto-reactive lymphocytes against SNCs and MSCs from the same individual. Upon co-culturing, the MSCs were transformed into neural stem cells within 48 hours. Moreover, we verified that this culture could be utilized to repair CNS wounds. [94] We then decided to ascertain whether this phenomenon of specific induction of differentiation by organ-specific auto-reactive lymphocytes also applied to other organs. To do this, we used the same MSC sample to perform three different co-cultures with anti-myelin, anti-myocardium and anti-islets of Langerhans lymphocytes, respectively. We did not add any exogenous trophic factors, nor did we utilize any adhesion matrix. At 48 hours, the control culture with undifferentiated lymphocytes conserved its usual fusiform morphology, but the incubated culture with the anti-myelin 40
lymphocytes had acquired neuroblastic phenotypes. The culture corresponding to the anti-myocardium lymphocytes acquired a double phenotype of endothelial cells and of myocardial cells. Similarly, the culture corresponding to the islet of Langerhans lymphocytes also acquired a double phenotype, the first resembling neuroendocrine cells, and the second resembling secretory adenomeres (Figure 19) [95]. MSC Culture 48 hr control
MSC Culture + Anti-Islet Lymphocytes 48 hr co-culture
MSC Culture + Anti-Islet Lymphocytes 48 hr co-culture
Figura 19 Efecto de la co incubación de Linfocitos Autorreactivos anti Islotes de Langerhans sobre la diferenciación de las MSC a las 48 horas de co-cultivo. [95]
After observing these last results, we believe that we could extend the aforementioned concepts to affirm that the immune system possesses a coordinating role in these diverse physiological phenomena, which most likely involves acting on stromal cells as well as mammary epithelial cell function.
41
Oncogenesis, Growth, Metastatic Dissemination and the Interaction between Breast Cancer and the Immune System Oncogenesis, Growth and Metastatic Dissemination With respect to breast cancer tumor biology and based on the concepts that we have covered to present, we should make special reference to a few aspects of particular importance. First, these tumors originate through the genetic disharmony of the progenitor cell, which is characterized by the loss of reproductive control and adequate development of contact with the organism's normal structures. After this occurs, the progenitor cell is considered the tumor stem cell. Second, the tumor stem cell is contained inside a cell niche, which is what causes the stem cell to be activated or to remain in a quiescent state. Third, it is likely that the EMT-MET transformation is the principle change in cell programming that the tumor stem cells pass through to cross the basal membrane, invade neighboring tissues, penetrate blood and lymphatic vessels and finally establish tumor metastasis. Fourth, stroma tumor stem cell should be permissive, and even proactive in the promotion of tumor growth for the tumor cells . Fifth, the immune system cells that comprise the stroma appear to be one of the most important factors controlling the emergence of mammary tumors or stimulating their growth and propagation.
Tumor stem cell There exists strong evidence that breast cancer originates and maintains itself from a small population of undifferentiated cells with self-renewal capacity [96]. CD44+/CD24-/low markers characterize the majority of these cells [97]. The small population of progenitor cells that express this phenotype generates an assembly of more differentiated cells representing the main tumor mass. This tissue behavior copies the hierarchical organization of the cells that form the normal mammary adenomere [96-98]. These cancer stem cells (CSCs) appear to share a similar phenotype with their normal counterparts. Nevertheless, they possess dysfunctional patterns of proliferation and cell differentiation compared to normal cells, and they do not
42
respond to the physiological controls that regulate cell renewal. As opposed to normal stem cells, CSCs promote the formation of blood vessels, promote cell motility as well as resistance to various therapies [97] and are mainly responsible for metastatic dissemination [98-100]. Since the discovery of CSCs, a paradigm shift regarding oncogenesis has occurred [101]. Previously, it was thought that any cell could randomly suffer mutations on the way to becoming a tumor, but now it is known that the neoplasm originates from normal tissue stem cells that suffer a dysregulation of the genetic and epigenetic pathways that regulate their auto-renewal. The dysregulated cells are expanded, as are other cell populations. They can be sensitive to different genetic and epigenetic signals and can differentiate in aberrant and various forms, thus giving rise to the cellular heterogeneity of neoplasms [102]. From this, we can deduce that the resultant cancer can be influenced by the variety of mammary progenitor cells in which these changes occur, the carcinogen that induced the mutation and/or genetic dysregulation and, especially, by the cell niche that contains the progenitor cell. Of all of the characteristics that they share with normal stem cells, however, their particular insensitivity to toxic environmental and pharmacological agents such as radiation stands out. CSCs over-express diverse transport molecules such as ABC, and they possess a greater capacity for DNA repair and apoptosis resistance enzymes [97, 103-105]. For these reasons, tumors include a stem cell population that can survive radiotherapy as well as chemotherapy. The survival of this population causes local relapse and metastasis.
Stem Cell Niche The stem cell niche is formed by the extracellular substance and the cells that surround it. These surrounding cells emit signals preventing the stem cells from differentiating and dividing. The role of the niche in the physiology of the mammary gland is obvious; under hormonal influence, it induces the complex transformations that the secretory epithelial gland goes through during puberty, pregnancy, lactation and subsequent restructuring [106].
43
The normal mammary cell niche, as previously mentioned, is capable of reprogramming the differentiation of previously differentiated cells (cell transdifferentiation), such as spermatocytes or neuroblasts, leading them to occur in normal mammary epithelia [107]. These facts show the importance of the cell microenvironment to the intrinsic nature of cell differentiation in alternative adult tissues. The breast tumor stem cell niche induces its growth, metastatic invasion and tumor dissemination [108, 109]. When tumor cells have been cultured inside the normal embryonic or adult adipose panicle, they exchange their malignant phenotype for a benign one, integrating into the normal glandular structures [73, 75]. These three situations show that the microenvironment in which the tumor stem cell is found is essential to determining its mature destiny with respect to normalcy or neoplastic transformation. To dissect this phenomenon and achieve a better comprehension of it, we will divide the niche-tumor stem cell problem into three aspects: 1) the changes incurred by a stem cell to grow, form the tumor mass and metastasize (EMTMET transformation); 2) what occurs in the stroma to facilitate these changes and 3) the importance of the immune system, which appears to be the main actor in the transformation of normal stroma into pro-tumor stroma (and vice versa).
EMT-MET Transformations of Mammary Oncocytes The medical prognosis of a breast cancer patient is directly related to the degree of dissemination of the disease [110]. Although the general cure rate has increased, the cure rates for advanced forms have not changed significantly for several decades. However, to develop a new therapeutic strategy that changes this situation, we first need to understand how and why the oncocytes can escape from the tissue in which it originates to invade neighboring tissues and colonize distant ones. The EMT-MET transformations that occur in tumor cells [111] are one of the avenues of study that has explained this problem with greater clarity. As mentioned, malignant epithelial cells make use of the transformation to dissolve
44
the basement membrane that contains them, invade adjacent tissues, penetrate blood and lymphatic vessels and invade distant organs. In malignant cells, MET signaling occurs, allowing it to grow as a metastasis (Figure 20). The morphological study of these phenomena has been complicated by the fact that it is difficult to distinguish the mesenchymal epithelial cells from the connective tissue of fibroblasts. In recent decades, immunohistochemistry and florescent in situ hybridization DNA staining techniques (FISH) have been used to describe these cells better. Thus, with a few exceptions [111-113], it has been ascertained that this phenomenon occurs in so-called active tumor margins [114, 115]. In the margins, EMT is initiated through epigenetic modifications and changes in genetic expression. Then, through extracellular stimuli such as those producing TGF-β and Wnt they acquire the mesenchymal phenotype and redirect their pathway. The mesenchymal cells respond to these stimuli by activating Snail/Slug/Twist, Cripto-1 and the Six1 regulatory genes, they acquire the mesenchymal phenotype, orient themselves and produce local tissue invasion and even undergo metastasis (Figure 9).
Figure 20: Cell culture of Adenocarcinoma cells from pleural effusion. In this picture is able to observe both epithelial and msenchymal phenotypes
The number of genes involved in EMT is directly related to the aggressiveness of the tumor. Furthermore, the expression of these genes can revert according
45
to the microenvironmental influences into which the cells arrive [116-121]. In addition to the type of metastatic dissemination induced by EMT, which results in tumor dissemination by isolated tumor cells, the collective migration of groups of oncocytes has also been described [122]. This has been observed in the study of human biopsies [123], in three-dimensional in vitro tumor cultures [124], in experimental animals through intravital staining and utilizing noninvasive methods [125, 126]. Having described these two forms, a third possibility arises that has also been observed in wound healing: the combination of epithelial cells that, without losing their contact with neighboring cells, develop mesenchymal characteristics allowing them to invade tissues and migrate in groups. Interestingly, it has been observed that the preferred place of metastasis is determined according to whether the epithelial or mesenchymal characteristic predominates. The cells that move alone, and particularly with the cooperation of macrophages, migrate preferentially toward blood vessels [127, 128], while those that migrate collectively do so toward the lymphatic vessels [126]. In summary, it has been observed that genetic induction caused by TGF-β, which induces EMT and consequently the migration of remote cells, is necessary to allow these remote cells to enter blood vessels, leave them and established themselves as hematogenous metastases [126,129]. Although inhibiting TGF-β signaling can inhibit the migration of isolated cells, it cannot inhibit the migration of all cells, resulting in the tumor metastasizing through the lymphatic pathway [126]. Another important effect of EMT is inducing oncogenesis after the increase in tumor stem cells [130]. The expression of multiple genes responsible for EMT in breast milk lines increases the number of tumor initiating cells, as determined by the increase in stem cell markers detected through flow cytometry, mammosphere-forming cells and the dilution limit test for the formation of tumors [130, 131]. The activation of the EMT regulating genes also influences cell survival, especially after chemotherapy treatment [132, 133], and prevents oncogene-induced senescence [134-136]. Both of these properties are associated with disease progression. Finally, it has been observed that the tumor cells that activate the genes responsible for EMT, such as the Snail gene, 46
generate immunosuppressant cytokines that block immunosurveillance action [137]. In conclusion, we can summarize that the induction of EMT actives the migratory properties of the tumor, induces its dispersion, selects its dissemination pathway, induces the proliferation of tumor stem cells, induces the emergence of chemoresistant clones, increases average tumor cell survival, prolongs senescence induced by the activation of oncogenes and induces cytokine secretion that blocks immunosurveillance. All of these characteristics induce tumor progression.
Influence of the Stroma on Oncogenesis, Development and Metastatic Dissemination of the Breast Tumor Various studies using experimental models have shown that proliferation, survival, polarity, the degree of differentiation and the invasive capacity of breast cancer cells can be modulated by myoepithelial and other stromal cells [138-150]. The anatomopathological diagnosis criteria utilized by pathologists to determine whether the carcinoma is classified as in situ or invasive are the presence or absence, respectively, of an intact layer of myoepithelial cells. This is established through an analysis of the specific markers found on these cells: actin related to smooth muscle, calponin and CD10 [151]. To ascertain why this very important cell layer disappears and allows, for example, a ductal carcinoma in situ (DCIS) to become invasive, Hu and colleagues used an experimental cellular xenotransplantation model where they injected human MCF10 cells into immunodeficient mice [152, 153]. This cell line has the particular quality of differentiating into myoepithelial cells as well as luminal
epithelial
cells
and
of
forming
tumors
when
injected
into
immunodeficient mice. These tumors possess significant molecular and histological resemblance to a comedocarcinoma in situ (ductal carcinoma in situ) that progresses to an invasive tumor such as invasive ductal carcinoma (IDC). If these cells are injected and accompanied by healthy human myoepithelial and fibroblast cells, the development of the tumor is prevented and/or impeded from progressing to IDC. However, if the stromal cells come
47
from a mammary tumor or the fibroblasts have originated in regions affected by rheumatoid arthritis, tumor emergence as well as progression to IDC is advanced [153]. In another group of experiments, it was shown that the loss or attenuation of TGF-β production (and of its corresponding signaling) accelerated tumor progression through the increase of cytokine Cxcl1 and Cxcl5 secretion as well as the overall expression of other genes associated with this cytokine from the tumor cell [154]. Other authors confirmed these findings after studying tumors with high cytokine production. These oncocytes recruit a peculiar type of myeloid cell (MDSC, myeloid derived suppressor cells) on the invasive tumor margin, which increases their ability to invade and metastasize. However, when the Cxcl1 and Cxcl5 chemokine receptors are blocked, Cxcr2 along with small paralyzing molecules decrease MDSC infiltration and generate pulmonary metastasis [155]. Various authors have found these fibroblast and macrophage pro-tumor associations in the abnormal response of these cells when intervening in the injury/scar formation process, finding groups of genes that are always expressed in stromal cells when a tumor is generated and progresses [156167]. The FSP1 gene, also known as S100A4, or metastasis promoting protein [157-161], is most likely the most important gene currently described. In turn, it has been observed that the inhibition of the PTEN gene in stromal fibroblasts stimulates the malignant transformation of the mammary epithelium, as well as tumor initiation and progression, transformation of the extracellular substance by increasing angiogenesis and recruitment of innate immune system cells, particularly M2 macrophages [168]. M2 macrophages have been associated with the tolerance phenomena and have been found to be infiltrated in different breast cancer metastases. Animal experiments have shown that when the action of these cells is inhibited, the aggressiveness and metastatic capacity of breast tumors also decrease [169]. This occurs when the M2s induce the epithelial cells to express and activate a greater quantity of epidermal growth factor receptors [170]. Coussens and colleagues have shown that IL-4-secreting CD4(þ) lymphocytes regulate the activity of these macrophages and the activity of MDSC [145] to favor their protumor action. 48
The pathological state of the stroma most associated with the risk of oncogenesis is most likely that of chronic inflammation. Recent epidemiological studies regarding the use of non-steroidal anti-inflammatory drugs and the risk of suffering from different types of cancer support this hypothesis [171].
Breast Cancer and the Immune System From the previous discussion, it can be deduced that the genetic and epigenetic alterations of the genes regulating cell proliferation, polarity, survival and cell differentiation of the mammary epithelia are the probable "oncogenic initiators" of breast cancer. In turn, the interactions of the mammary epithelia with their stroma, and particularly with the cells of the immune system that compose it, are what determine their future evolution [172-178]. Subsequently, we will analyze the apparently ambiguous role of protection-promotion that each cellular component of the immune system plays.
Leukocytes and Tumor Development The cells of the innate immune system include macrophages, granulocytes, dendritic cells (DCs), mast cells and natural killer cells (NK). They represent the first line of defense against pathogens and foreign agents. Because of their capacity to destroy, absorb and produce antigens to generate a specific immune response, the first three cell types are called antigen-presenting cells (APCs). When tissue homeostasis is altered, the resident APCs and the mast cells
produce
cytokines,
chemokines,
bioactive
mediators
and
matrix
remodeling proteins with the purpose of recruiting greater quantities of circulating leukocytes to concentrate at the lesion site. This is the first phase of the inflammatory process [173, 179, 180]. The action of APCs in situ can eliminate the pathogen. Once absorbed, the APCs migrate toward the regional lymph nodes where the pathogen is treated as an antigen and is presented to the T and B lymphocytes, which initiate the specific (adaptive) immune response [181, 182] (see Figure 21). Nevertheless, as we saw in the previous section, this immune response can be responsible for tumor promotion before oncocyte rejection. This concept was
49
developed by Virchow in 1863 when he proposed that cancer originated in sites of chronic inflammation [183]. Memory B Activated B Memory T reg
Activated T reg
CoCo-stimulation
Differentiation Plasmocito
B Naive Stimulation CD4 Naive Activated CD4 TH2 Memory CD4 Th2
Ag Presentation
Stimulation Activated CD4 TH1 Activated Macrophage
Memory CD4 Th1 Naï Naïve CD8
Ag Process
Effector Response
Resting Macrophage Phagocytosis Virus
Differentiation
NKT
Activated CD8
Memory CD8
Figure 21: Anti viral Effector Immune response associated to a regulatory respponse induced by T reg Lmphocytes. The MDSC intervention is not graphictated
Historically, it has been thought that the leukocytes that infiltrated and surrounded tumors represented the host's attempt to eradicate the transformed cells. This is particularly true for natural killer T (NKT) and NK cells [184]. There exists epidemiological data supporting these hypotheses, especially in reference to virus-associated tumors [185-189]. The most frequent viruses are those of the human papillomaviruses (HPV) associated with uterine, cervical and epidermoid carcinomas, herpes 8 associated with Kaposi's sarcoma and Epstein-Barr virus associated with Non-Hodgkin's Lymphoma [185-189]. Similarly, it has been shown that in the immunocompromised population, melanomas as well as lung adenocarcinomas are also more frequent [187-190]. Contrary to these observations, the arrival of AIDS has shown epidemiologically that immunocompromised women possess a lower risk of contracting breast adenocarcinoma [187,190-193]. Currently, a growing body of literature exists that indicates that, particularly regarding breast cancer, stromal leukocytes 50
appear to play a developmental role in neoplasia, as their numbers increase when tumor aggressiveness increases (Figures 22 and 23).
Figure 22: The development of human breast cancer is characterized by an increasing infiltration of immune cells (detected by immunmarction of CD45) as the severity is grater - Breast Cancer Res. 2007; 9(4): 212.
Figura 23: The development of human breast cancer is also characterized by an increasing infiltration of immune cells (detected by immunmarction of CD4, CD8 and CD20) as the severity is grater - Breast Cancer Res. 2007; 9(4): 212.
B Lymphocytes as the Promoters of Breast Cancer The secretion of antigen-specific immunoglobulins by differentiated B lymphocytes (plasma cells) has been associated with the anti-cancer effect of these cells (Figures 21 and 24). Although it may be true that acute inflammatory responses eradicate incipient neoplasms [194], new data indicate that chronic 51
activation of these lymphocytes plays an important role in tumor promotion (Figure 24). In fact, the activation of B lymphocytes in acute wounds can lead to the release of cytokines (such as IL-4, IL-6 and IL-13), other soluble mediators, immunoglobulin subtypes and to activation of the complement cascade, doubling their effect. Together, these factors trigger the beginning of an effector immune response, but when it is prolonged over time, they produce a chronic inflammation that generates deleterious effects, also mediated by the cells of the innate immune system, such as is observed in several pathological situations (rheumatoid arthritis and other autoimmune illnesses) [195].
Figure 24: Dual relationship of the immune system with respect to transformed breast cells. An acute reaction is characterized by the effector action of T and B lymphocytes supported by M1 macrophages. This collective action leads to the destruction of the fledgling tumor. However, when the inflammation becomes chronic, these same cells develop immune tolerance and tumor growth promotion through epigenetic stimulation of the tumor, which copies an organ being repaired. Breast Cancer Res. 2007; 9(4): 212.
During the carcinogenesis of breast adenocarcinoma, a growing infiltration of B lymphocytes is observed, increasing in number in accordance with disease progression (Figure 23). Compared with biopsies from healthy individuals, the patients' sentinel lymph nodes are enriched with mature B lymphocytes (IgG+) 52
[196]. The number of cells is directly correlated with the increase in tumor staging and aggressiveness [197-199]. Similar findings have been obtained from primary tumor biopsies with the observation that B lymphocytes surpass T lymphocytes in number and their quantity is proportional to disease progression [200, 201] (Figure 23). Approximately 20% of invasive carcinomas contain a high number of B lymphocytes. In these patients, the percentage of lymphocytes can be 60% of the total number of mononuclear cell infiltrates [202]. This observation is not exclusive to breast cancer, as it has been reported that 70% of solid tumors are infiltrated with these cells [203]. In DCIS and IDC, B lymphocytes are found in the perivascular space in groups associated with T lymphocytes, forming follicular structures [202-205]. These follicles have lymphocytes that are related by interdigitation with DCs characterized by CD21+ markers. Because of this, they can be identified as authentic ectopic follicles. These follicles contain plasma cells, an index of the chronicity of the process. These structures have also been described in autoimmune diseases such as rheumatoid arthritis, multiple sclerosis, Sjögren’s syndrome and Graves’ disease, where it is thought that they are part of the pathogenesis of that particular condition [206-209]. The auto-antibodies produced by these plasma cells have been found in the serum and/or interstitial deposits of tumor tissues [210]. The early emergence of auto-antibodies in the serum (particularly smooth anti-muscle) is considered a sign of poor prognosis for the patient [211]. Approximately 50% of patients with breast cancer carry circulating immunoglobulins against tumor antigens. AntiErbB2/HER2/neu antibodies are present in 20% of ErbB2-positive patients [212]. Paradoxically, these are the patients with the worst prognosis [210-213] (Figure 24), indicating the inefficiency of these antibodies and their probable role in tumor promotion. Confirming this last point can explain how more than 50% of patients with breast cancers possess antibodies against tumors even though the registered cases of spontaneous tumor remission are anecdotal [214, 215]. Various factors can be the cause of this biological inefficiency, including the concentration of specific immunoglobulins, the expression of determined HLAs, the generation of immune tolerance, altered cytotoxic T cell activity and others.
53
Adaptive Immunity and the Development of Breast Cancer: T Lymphocytes T lymphocytes associated with tumor development have been detected through immunohistochemistry (Figure 23), but their prognostic role is under debate. While B Lymphocytes predominate during initial tumor stages [204], both CD4+ and CD8+ T cells predominant in high grade DCIS and IDC [216]. The proportion of these cells with respect to the total cell mass covers an extensive range, between 1% and 45% [217]. In rapidly proliferating tumors, their presence is an index of good prognosis and they are correlated in patients with negative sentinel lymph nodes and small tumor size with a lower degree of histological malignancy and greater disease-free survival [218]. All these data support their role in immunosurveillance.
Figura 25: Relación dual del sistema inmune respecto de la evolución del tumor. Papel de la respuesta inflamatoria aguda dominada por la actividad de los Th1 y de la respuesta inflamatoria crónica dominada por la actividad Th2 para producir rechazo o promoción tumoral respectivamente Breast Cancer Res. 2007; 9(4): 212.
The most important indicator of tumor progression and patient prognosis is the presence of malignant epithelial cells in the sentinel lymph node [219-221]. Contrary to what was previously mentioned, while the role of CD8+ cells is unclear, it is accepted that the greater the quantity of CD4+ cells infiltrating the
54
primary tumor, the worse the patient prognosis. This includes a greater number of metastatic ganglions, greater increase in primary tumor mass and greater tendency toward metastatic dissemination [217, 222]. Possibly more significant than the individual proportion of lymphocytes is the CD4+/CD8+ proportion in the infiltrated tumor. Higher CD4+/CD8+ proportions are related to metastatic ganglions and a lower patient survival [217, 222]. Similar results have been reported for carcinomas of other origins [223, 224]. A possible explanation regarding these contradictory results might be based on the different polarities that the CD4+ phenotype can develop, specifically Th1 and Th2 (Figure 25) [225]. Faced with an antigenic stimulus, Th1s secrete IFNγ, TNF-α and IL-2 [226]. These cytokines act together with the cytotoxic function of CD8+ cells [227] to increase the production of antigens through proteasomes and the expression of class I and class II MHC molecules and co-stimulatory molecules in tumor cells. The secretion of IFNγ also induces the activation of M1 macrophages, which become cytotoxic [228]. In contrast, Th2s express IL-4, IL-5, IL-6, IL-10 and IL-13, which induce anergia of the cytotoxic CD8+ cells and increase humoral immunity (B lymphocytes) [229]. Taken together, these facts prompt us to think that the immune response governed by Th1s produces anti-tumor effects that are beneficial for the patient [230-233], while the reaction governed by Th2s diminishes anti-tumor immune responses [233-237] and increases the tumor-promoter response [238, 239].
The Role of Regulatory T Lymphocytes (Treg) In addition to Th1-Th2 polarity, CD4+ cells can have a third phenotype, Treg. Through cell-cell contact, Treg cells inhibit the action of Th1 CD4s, CD8s and dendritic cells. Normally, they protect tissues from autoimmune aggressions. They can be identified because they express CD4, CD25 and FOXP3 proteins on their surface. In this way, it has been established that they are present in a variable proportion from 5-10% of all T lymphocytes present in a healthy organ. In breast cancer, DCIS possesses a normal proportion of Treg, which increases as they transform into invasive cells and the disease progresses [240]. This increase in Treg is an indicator of poor prognosis. They inhibit effector immune
55
responses through cell-to-cell contact with all of the acting effectors in a dosedependent manner [241-245]. It is believed that tumors are the main instigators of Treg through the secretion of various factors such as prostaglandin E2 and CCL22 by M2 (pro-tumor) macrophages [246-249]. Innate Immunity and the Development of Breast Adenocarcinoma As has been mentioned on several occasions, when innate immune cells act in chronic inflammation, they induce the development of the breast tumor [173, 178, 250]. This is attributed to the enormous plasticity of these cells and to the fact that they can secrete a myriad of cytokines, chemokines, metallo-serine and metallo-cysteine proteases, free radicals, histamine and other bioactive mediators [173, 178, 250]. Several of the physiological processes that induce these substances to begin injury healing are also beneficial for tumor progression, such as an increase in cell life, cell matrix reconstruction, angiogenesis and suppression of the adaptive immune response. This regulation has been illustrated by the positive correlation between the number of innate immune response cells (macrophages, mast cells and neutrophils) and the degree of tumor infiltration and the number of blood vessels [251, 252]. Studies conducted using animal models have also shown that reducing the innate
immune
response
reduces
angiogenesis,
thus
limiting
tumor
development [253-260]. In the mid-1950s, pioneering published studies showed that this pro-tumor effect of adaptive immunity with respect to the passive transfer of tumor-specific antibodies accelerated the growth of transplanted tumors in vivo [261-263]. In the last decade, various studies have reinforced these findings that the protumor effect of B lymphocytes is related to the regulation of innate immune system activity and not through direct action on the tumor [264, 265]. Studies by Barber-Guillem and colleagues found that humoral anti-tumor immune responses facilitate tumor growth and invasion in vivo in murine and human transplanted tumors. This effect is mediated by the recruitment and activation of pro-tumor granulocytes and macrophages [264, 266, 267]. The deposit of immunoglobulins in the neoplastic microenvironment can mediate the recruitment of macrophages through the complement activation cascade or through Fc receptor expression in resident APCs [268, 269]. During breast 56
tumor development, phagocytosis of stromal tumor immunoglobulins increases the production of macrophages associated with vascular endothelial growth factor (VEGF) and therefore tumor angiogenesis [266]. Other experiments directly correlate the degree of tissue infiltration by macrophages with virusinduced tumor progression [256, 270]. It also induces the development of pulmonary metastasis [256]. In contrast to what has been described, NK cells protect the host from tumor progression by non-specifically detecting and eliminating transformed cells and by inhibiting tumor angiogenesis [273, 274]. What is certain is that the microenvironment in which the tumor grows is also modified by this very tumor. It produces immunologically active substances that promote pro-neoplastic chronic immune responses opposite to that of the acute immune response [178, 275], where lymphocytic cytotoxic activity is inhibited by T effectors [276]. Additionally, these tumor secretions attract a special subpopulation of MDSC innate immunosuppressant cells toward them. These are identified by CD11b(+)/Gr-1(+) surface markers. They accumulate around the tumor and invade the blood and other tumor-bearing lymphatic organs [275279]. It is accepted that effector immune activity is inhibited by cell-to-cell contact with effectors. It is also known that they promote tumor growth through the secretion of angiogenic substances [280]. The cytokines derived from the activation of humoral immunity and/or Th2s act directly on macrophages. The chronic activation of B lymphocytes (typical of ectopic germinal centers) produces GMCSF, IL-6 and IL-10 [281]. These cytokines in combination with those produced by Th2 (IL-4, IL-13 and IL-10) produce a polarization of the innate immune system, generating the pro-tumor M2
macrophages
while
simultaneously
repressing
pro-cytotoxic
M1
macrophages [282] and the differentiation of DC monocytes [283]. Taking all of these facts together, we can conclude that all of the factors generated through chronic inflammation construct a tumor microenvironment through the development of immune tolerance and the promotion of tumor progression. In summary, it becomes apparent that the development of an effective immunotherapy against breast cancer should primarily be designed to change the myriad conditions that convert the primary tumor stroma and its metastasis 57
into a pro-tumor environment. The immune system is modulated by the tumor that develops the same actions as a tissue being repaired, and consequently converts the immune system into its ally. After summarizing the current nonimmune therapeutic strategies, we will develop, through the analysis of our clinical trial, the principles of a rational immunotherapy.
58
Current Oncological Therapies According to the official information supplied by the United States National Cancer Institute (NCI) in its therapeutic manual the Physician Data Query (PDQ), the treatment of different cancers is regulated by the size and local advancement of the primary tumor, the location and number of lymph nodes affected and whether metastatic dissemination of the disease exists. The determination of these characteristics in the patient is called the staging of malignant disease. This allows therapeutic strategies to be established for each group of patients according to their respective prognosis. Therapy decision-making for breast cancer patients considers the stage of disease, the presence or absence of estrogen, progesterone, or human epidermal growth factor 2 (HER/2neu) receptors in the tumor tissue, whether the patient is in menopause and her clinical condition (ECOG or Karnovsky index). According to current medical knowledge, therapeutic possibilities with the intention of healing are considered in patients with a tumor less than 50 mm at its maximum diameter and/or less than 9 metastases in the ipsilateral regional lymph nodes. The presence of extra regional lymph nodes or of any sign of metastatic disease, according to modern criteria, classifies the patient as having advanced disease. Therefore, her therapy is not performed with the intent to cure, but rather with the purpose of extending her survival expectancy and quality of life. This group of patients, according to the United States NCI, the American Society of Clinical Oncology (ASCO) and the European Society for Medical Oncology (ESMO), is considered qualified to receive innovative therapies with a therapeutic purpose or with the purpose of establishing their safety and efficacy [284]. In Appendix 1, different disease-staging criteria as well as the official therapeutic summary established by the United States NCI PDQ database is transcribed. Currently, the treatment of advanced stages is combined. Hormone therapy and chemotherapy associated with trastuzumab is utilized. The use of radiotherapy and/or surgery is limited to patients with specific types of symptomatic metastases. According to the United States NCI, all patients with advanced breast carcinoma should be considered candidates to receive
59
innovative therapeutics [285]. The criteria for surgical indication include patients who have impetiginized or painful thoracic wall lesions, metastases causing pain due to their location. This also includes metastases that have migrated to the brain, those that compress the spinal cord, and those whose removal can impede or prevent a bony fracture, pulmonary or remote liver metastases or those where a pleural or pericardial effusion exists. Radiotherapy is indicated when local symptoms caused by the metastasis should be alleviated including painful bony metastases, cerebral metastases or spinal medulla or meningeal unresectable metastases. Radiotherapy is also indicated for impetiginized or painful thoracic wall lesions, following CNS decompression surgeries or following the setting of pathological fractures. Currently, the possibility of systemic administration of strontium 89 is being studied when multiple bony metastases exist. Regarding systemic therapies, bisphosphonates, various forms of hormone therapies, monoclonal antibodies that block epidermal growth factor receptors and chemotherapy are currently being utilized. It has been verified that bisphosphonates reduce the morbidity of patients with bony metastases. They act by modulating the activity of osteoclasts, which are macrophages that are essential for the growth of bony metastases associated with breast cancer. The relationship between bony metastasis and osteoclasts copies the interaction of TEB and macrophage growth in the mammary tissue. Phosphonates have indirectly verified the influence of macrophages in angiogenesis, as patients treated with this drug can present the toxic effect of avascular necrosis of the jawbone [286, 287]. Hormonotherapy is particularly indicated as a first line of treatment for patients with a diagnosis of estrogen receptor positive (ER+) and/or progesterone receptor positive (PR+) or ER/PR unknown advanced breast cancer. The first medication used was testosterone. Afterward, hormone receptor competing inhibitors were used (tamoxifen and similar) with effects similar to that of testosterone, but without the secondary virilizing effects. LHRH agonists are a third generation of drugs that inhibit the production of gonadotropins by saturating the pituitary gonadotropin cell receptor. Finally, aromatase inhibitors 60
inhibit the synthesis of all steroid hormones (including corticosteroids). It has been observed that hormonal medications can promote pharmacological drugs if they are combined [288-293].
Figura 26 Isoformas de las dos subunidades del receptor del factor de crecimiento epitelial humano (hEGF) del que sobresale la subunidad HER2. Este antígeno se ha denominado HER2/neu (también conocido como ErbB-2, del ingles Human Epidermal growth factor Receptor 2)
Approximately 25% of patients with breast cancer possess cells that overexpress a subunit of the epidermal growth factor receptor HER2/neu (see Figure 26). Based on this observation, two different types of humanized antibodies are used as a second line medication, Trastuzumab and Lapatinib. Both have shown a synergistic effect when combined with chemotherapy, although they have also been shown to promote several of the side effects of the respective chemotherapeutic drugs [294-299]. Patients who worsen during hormonotherapy treatment are candidates to receive chemotherapy. Similar criteria have been adopted for those who have been diagnosed as ER/PR negative [300]. The use of more than one chemotherapeutic agent seems to be more effective than the use of a single agent. The multiplicity of cytostatics appears to increase the proportion of
61
patients with tumor response and disease-free time, but overall survival has not improved with any combination. The same can be said of high dose chemotherapy associated with bone marrow transplants [301-306]. Overall, the decision to use chemotherapy on these patients, or the decision of which chemotherapeutic agents should be used, is presently based on the speed with which patients relapse and, in each case, their general state at the time of deciding on the therapy (for more details, see Appendix 1) [301].
62
The Basis of Current Immunotherapies In the middle of the 20th century, Burnet and Thomas suggested that the immune system would be able to defend us from the growth of tumor cell neoformations. In the 1960s, it was proposed that the changes in growth control and cell differentiation of cancer occurs mostly from genetic alterations produced by secondary, spontaneous mutations in response to environmental factors. It was calculated that every day between 104 and 106 cells potentially capable of generating a tumor were produced. The role of destroying these potential tumor sources was given to the immune system cells. This gave rise in the 1970s to the development of various therapeutic techniques capable, at least in experimental models, of preventing the development of experimental tumors. Later, some technical immunotherapies managed to show clinical utility, but at the beginning of the 21st century, new studies revealed that under certain conditions, the immune system could be become an ally of tumor development. This process was called tumor immunoediting [307-309]. The immune system prevents the emergence of tumors through different mechanisms. First, it protects the organism from virus-induced tumors by eliminating or suppressing viral infections. Second, through the rapid elimination of pathogens, and consequently of associated inflammation, it prevents the establishment of a chronic inflammatory state that induces tumorigenesis. Third, it can specifically identify and eliminate new tumor cells, recognizing the mutated proteins that the oncocyte presents and that do not belong to the adult individual; this is termed immunosurveillance [308]. Tumor immunoediting is divided into three phases: elimination, equilibrium and escape
(Figure
27).
The
first
of
these
phases
coincides
with
the
immunosurveillance concept described in the previous paragraph, where the immune system detects and destroys cells that have been transformed because of the genome's intrinsic failure to eliminate mutations and oncogene expression. The elimination phase is considered complete if all of the transformed cells are destroyed or incomplete if only a portion of these cells is eliminated. In this last case, a second phase of temporary equilibrium between the immune system and tumor growth is established. It is assumed that during
63
the equilibrium period the tumor cells remain in a quiescent state, or continue evolving and accumulating more changes in their DNA. In either case, they produce certain changes to their phenotype, which include non-expression of the antigens previously recognized by the immune system, the decrease or suppression of the genetic expression of antigen presenting and co-stimulatory molecules and the over-expression of the genes related to TGF-β and IL-10 production. When the destruction of all of the immunosusceptible cells is surpassed by that of resistant cells, the third phase, disease progression, begins [309].
Figure 27: Tumor Immune editing phases - Smyth, M.J y col. [309]
The phenotypic changes of the oncocyte that occur during the equilibrium phase generate changes in the immune system and in the rest of the stromal tumor elements. Effector T lymphocytes recognize the oncocyte antigens when they are presented by the MHC molecules that are associated with costimulatory molecules. The lack of co-stimulatory molecules produces tumor anergia. In the meantime, the tumor produces IL-10 and TGF-β, both of which induce the conversion of Th1 into the Th2 effector response (from chronic
64
inflammation), Tregs are generated, fibroblasts are induced to develop an antiinflammatory matrix rich in collagen I, IFNγ production is inhibited, MDSCs are attracted toward the lesion site and M1 macrophages are transformed into M2 macrophages (Figure 28). Memory B
Memory T reg Activated B
Activated T reg CoCo-stimulation
Differentiation Plasmocito
B Naive Stimulation CD4 Naive Activated CD4 TH2 Memory CD4 Th2
MDSC Ag Presentation
Stimulation Activated CD4 TH1 Activated Macrophage
NK
Memory CD4 Th1
Naï Naïve CD8
Ag Process Resting Macrophage
Activated CD8 NK
Phagocytosis
Anti-tumor Response
Differentiation
NKT
Memory CD8
Figure 28: Antitumor Immune response (blue lines) and the corresponding inibitory immune reactions (red lines). The later Reactions induce immune tolerance to the tumor.
The consequence of these changes is that a pro-tumor chronic inflammatory state is established. This state is very difficult to reverse because the preexisting anti-tumor clones have disappeared. In addition, the lack of adequate cell signals from the oncocyte, which now finds itself in an environment that protects and promotes it, prevents the immune system from developing a new effector cell response. Various therapeutic strategies have been developed to overcome these difficulties and to restore the conditions necessary to return to the first phase, tumor cell elimination. Figure 29 lists the different treatments that have been tested in our center; the treatments are placed in the figure where it is assumed that they exercise their action.
65
Chemoth. MAB
Tα1 MLC
Treg V ALT
DC-TBH
MLC
CD3+ CD25-
Tα1 LAK
TBH
Anti-tumor Immunotherapy
NK
TIL
ALT
Figure 29: Different anti-tumor immunotherapies utilized at our center
We have used two types of autologous cell vaccines, the TBH vaccine [12-16] and a vaccine composed of autologous dendritic cells challenged with autologous TBH [310, 311]. Some patients received a cytoimplant (MLC) to induce vaccine therapy [312-314] consisting of a mixed culture of lymphocytes produced by culturing peripheral mononuclear donor cells that were unrelated to the patient's irradiated cells. When it has been possible to obtain TIL lymphocytes, we have isolated, cultured and re-infused them into the patient [315]. On other occasions, we have given the pre-immunized patient T effector lymphocytes (CD3+CD25-), which behave as TILs [316, 317]. To regulate the activity or the emergence of regulatory T lymphocytes, we have administered low dose chemotherapy prior to each vaccine [315]. When the patient presented intolerance to this medication, we administered thymosin α1 (Tα1) [318]. We also tested the use of an auto-vaccine using tumor-protecting Treg lymphocytes. The object of this immunization was to stimulate anti-idiotypic T lymphocytes that eliminate these lymphocytes [319]. Finally, on one occasion, we utilized ALT therapy [320-323] to promote the reactivation of memory lymphocytes. Because of their historical importance, LAK therapy and 66
monoclonal antibody therapy are also mentioned in Figure 29; neither of these two therapies was applied to our patients. We subsequently developed a TBH vaccine preclinical history because we have been pioneers in its clinical use and it is the base of most of our cell therapies.
The TBH Vaccine as a Means of Overcoming the Obstacles of Tumor Cell Recognition Activated B lymphocytes are one of the most robust APCs for CD4+ lymphocytes. They express MHC type II and the rest of the necessary costimulatory molecules on their membrane. They also produce several leukines, which positively promote immune response. The antagonistic similarity between activated B lymphocytes and tumor cells inspired the creation of a tumor vaccine composed of a cell hybrid between the two. This "new cell" is cultured in media containing special supplements to increase the number of cells and maintain the state of lymphocytic activation, particularly for the B lymphocyte. Once the patient is injected, this allows the hybridoma to "present" its own Ag (originating from the lymphocyte) and those from the cancer cell on its cell membrane (Figure 30). These Ags are important to the highly specialized communication with CD4, which possesses MHC II molecules. Thus, an unexpected phenomenon is that the hybridoma also presents its "own" Ag on MHC I molecules. This is the vehicle of communication between specific cytotoxic lymphocytes (CD8), which perform the real antitumor tasks under the instruction of sensitized CD4s. Finally, this hybridoma secretes more activating leucines than anti-inflammatory substances, which increases the organism's immune response. Independently, Guo and Moviglia [12-16] developed and studied this model in laboratory animals, a mouse model of hepatocellular carcinoma and in patients with different types of advanced neoplasms, respectively. They demonstrated the immunological basis, low toxicity and high clinical effectiveness of this therapy. In studies using immunostaining of neoplastic cells where the neoplasms originated from different human and murine tumors with low antigenic capacity, it was verified that all of the cells expressed low levels of, or failed to express,
67
proteins belonging to the major histocompatibility complex (MHC) type II molecule. They did, however present proteins belonging to MHC I molecules. Thus, they were lacking in the expression of co-stimulatory adhesion molecules such as B7. MHC II molecules, along with co-stimulatory molecules, allow the recognition of an antigen on a cell transformed by CD4 tumor specific clones. These cells initiate the rejection reaction, instructing CD8s to target the cells that need to be destroyed. However, even when obtaining professional CD8 tumor antigen clones, they were not effective in vitro in the destruction of tumor specific cells even though they became specialized through MHC I. The aggregation of co-stimulatory molecules, such as B7, through genetic engineering allowed the CD8s to comply with their assignment. Thus, the blockade of co-stimulatory molecules not only generates a lack of reaction, but it also induces peripheral tolerance. Guo utilized a syngeneic Wistar rat tumor line. The cell membrane was generated from a Wistar rat hepatoma cell line, BERH-2, obtained by treating these animals with methylcholanthrene, which presents very little or null antigenicity, low expression of MHC type I and ICAM 1 and does not express MHC type II, B7 or LFA-1 (Figure 31) [16]. Activated B lymphocytes and CD4 APCs are almost as powerful as dendritic cells, and these cells present the molecules missing from BERH-2 on their surface. Considering these facts, Guo and colleagues [16] generated a hybrid line (TBH) resulting from the polyethylene glycol-mediated fusion of BERH-2 with activated B lymphocytes, thus generating BERH-2-B (Figure 30). This new cell line expresses all of the surface molecules that are separately present in both progenitor lines. When syngeneic animals are injected, they produce an important inflammatory reaction with lymphomonocytic infiltrates. When the infiltrated lymphocyte phenotype is studied, a 70% CD8 and 30% CD4 phenotype is presented. The TILs originating from this reaction have cytotoxic action against BERH-2-B cells and the original BERH-2 cells, according to what is shown by cytotoxicity with Cr51 liberation trials. However, this cytotoxic action is tumor specific, as it cannot destroy syngeneic tumor lines originating from transitional cell or lung tumors.
68
TBH
Activated B-Lymphocyte
CD8 (T(T-Lymphocyte)
APC Hybrid cell Tumor cell and B-lymphocyte fusion “Wild” Wild” Tumor Cell
Tumor Cell Protein
CD4 (T Lymphocyte)
“Wild” Wild” Tumor Cell Figure 30: Schema of TBH Vaccine formation
Figure 31: Comatogrphic análisis of different cell membran proteins on activated B cells, Tumor Cells and Fusions cells (TBH).
The animal model described, which is based on the production of a cell line formed from the hybridization of the hepatoma line, BERH-2s, with autologous activated B lymphocytes to generate BERH-2-B cells, was utilized to study the actions and properties of this type of immunization. 69
In the first experiment, 10 rats were injected through the portal vein with 2x106 BERH-2 cells, which died inside the injected rats within 60 days. Another 10 rats, injected through the same pathway with 2x106 BERH-2-B cells, did not develop tumors during the 180 days of observation of the experiment. These results were repeated by comparing four additional tumor cell lines with hybridomas obtained in the same way as the BERH-2-Bs. In this way, it was concluded that the hybridomas lost their ability to develop into syngeneic rat tumors. However, after injecting nude mice, animals that lack a thymus and thus permit the development of xenografts and tumors grafts, each one of the hybridomas developed a tumor. It was therefore concluded that the difference in tumor development was associated with the inflammatory reaction that generated each of the hybridomas. In a second group of experiments, we attempted to determine whether animals could be immunized with a TBH pre-treatment and with the subsequent intrahepatic tumor graft that gave rise to TBH. For this experiment, three sets of eight
animals
were
utilized
(Figure
32).
Each
group
was
injected
subcutaneously with 2x106 cells; the first group received BERH-2, the second BERH-2-B and the third irradiated BERH-2-B. At the end of two weeks, the three groups received an intrahepatic implant of 2x106 BERH-2 cells. Only the second group, which was immunized with viable BERH-2-B cells, survived for the 120 days that the rats were maintained; none developed tumors. The other two groups developed tumors and died within 60 days of having received the second injection. This experiment was repeated two times with identical results (Figure 32). To corroborate this immunogenic effect of BERH-2-B, fourteen rats were injected intrahepatically with 2x107 BERH-2 cells. Ten days later, eight of these rats were immunized with a subcutaneous injection of 5x106 BERH-2-B hybrid cells, while the remaining six received a subcutaneous injection of 5x106 BERH2 cells. The eight TBH-immunized rats survived for the remaining 100 days of observation; they did not develop intra-liver tumors. However, the six rats immunized with parental tumor cells developed intra-liver tumors and died within 60 days of injection. This experiment was repeated two times with identical results (Figure 33).
70
Surviva l 10 0
Survival probability (%)
80
60 T reatmen t BERH-2 BERH-2 i rra d. BERH-2 -B 40
20
0 0
20
40
60
80
10 0
12 0
T i me in d ays
Figure 32: Kaplan Meier estimator plot showing the percent of surviving animals after intrahepatic injection of a deadly dose of BERH-2, pre-immunized to varying degrees (see legend in the box). Time (x-axis) is expressed in days.
To study whether the degree of tumor development could influence this BERH2-B immunogenic effect, small fragments of BERH-2 tumor masses (0.3 mm x 0.5 mm) were implanted in the liver. Ten days later, eight of these rats were immunized with a subcutaneous injection of 5x106 BERH-2-B hybrid cells. The remaining six rats received a subcutaneous injection of 5x106 BERH-2 cells. The eight rats immunized with TBH remained alive for 60 days; between 60 and 70 days, two animals died from intrahepatic tumor development. The six survivors did not present tumor masses on autopsy after completing 120 days of observation. Similar to the previous experiment, the six rats immunized with parental tumor cells developed intrahepatic tumors and died within 60 days of injection. This experiment was repeated two times with similar results (Figure 34).
71
Survival 100
Survival probability (%)
80
60 Treatment BERH-2 BERH-2-B 40
20
0 0
20
40
60
80
100
120
Time in days
Figure 33: Kaplan-Meier plot showing the percent of surviving animals after intrahepatic injection of a deadly dose of BERH-2, immunized 15 days later to varying degrees (see legend in the box). Time (x-axis) is expressed in days. Survival 100
Survival probability (%)
80
60 Treatment BERH-2 BERH-2-B 40
20
0 0
20
40
60
80
100
120
Time in days
Figure 34: Kaplan Meier plot showing the percent of surviving animals after intrahepatic tumor implant of small fragments of BERH-2, immunized 15 days later to varying degrees (see legend in the box). Time (x-axis) is expressed in days.
72
To corroborate that this was the result of the hybridoma and not the sum of both cells, three groups of eight rats were selected. The first group was injected with 5x106 BERH-2 cells, the second was injected with an equal quantity of BERH-2B cells and the third was injected with 5x106 BERH-2 cells plus 5x106 B lymphocytes in the presence of polyethylene glycol; all three groups were injected subcutaneously. At the end of two weeks, all animals in the three groups received an intrahepatic implant of 5x106 BERH-2 cells. Only the second group, which was immunized with BERH-2-B cells, survived and did not develop tumors during the 150-day period over which the rats were maintained. The other two groups developed tumors and died within 60 days of having received the second injection. This experiment was repeated three times with identical results (Figure 35). In this last group of experiments, the TBH cells were unpurified, but nonetheless they conserved their effectiveness. Survival 100
Survival probability (%)
80
60
Treatment BERH-2 BERH-2 + B BERH-2-B
40
20
0 0
20
40
60
80
100
120
Time in days
Figure 35: Kaplan Meier plot showing the percent of surviving animals after intrahepatic injection of a deadly dose of BERH-2, immunized 15 days later to varying degrees (see legend in the box). Time (x-axis) is expressed in days.
These four experiments corroborate what was proposed. The hybridization of these two cells, acting as live cells in the animal, induces an immunogenic effect. The tumor cells alone, when injected in an unfavorable place such as the
73
subcutaneous tissues, do not proliferate and do not produce any antigenic reaction. They therefore lack the membrane signals necessary for recognition by CD4s. Not even the presence of B lymphocytes with tumor cells is enough. B lymphocytes must receive the antigen presented by another APC. Only the hybrid has this property because in the formation of the hybridoma membrane, protein fractions of the tumor cell antigens are processed and presented by the cellular elements that utilize B lymphocytes to physiologically communicate with T cells. This is also the reason why irradiated TBHs did not generate any effector response and allowed tumor development. T lymphocytes that have this strange recognition property are responsible for initiating the rejection response that, through cross-reaction, prevents the development of or destroys hybridoma parental tumor cells. Once the immunogenic effectiveness of treatment with TBH was confirmed, a third group of experiments was developed to ascertain which T lymphocyte subpopulation or subpopulations were responsible for this phenomenon. Three groups of six rats each were treated with purified mouse monoclonal antibodies against CD4 (OX38) for the first group, against CD8 (OX8) for the second group and against dimethylamine pentaacetic acid (DTPA) for the third group (control). Each animal received an intravenous injection of 500 μg of monoclonal antibody, two times per week, for three weeks. Two days prior to the TBH injection, the grade of inhibition reached in peripheral blood was controlled by immunostaining and fluorescence microscopy. Those rats treated with anti-CD4 reduced the number of CD4s by more than 95%, those treated with anti-CD8 reduced the number of CD8s by approximately 95%, and those treated with the control antibody showed no changes in the number of CD4s or CD8s. A fourth group of six rats, also considered controls, received no monoclonal antibody treatment. Each rat in the four groups received an intrahepatic injection of 5x106 BERH-2-B hybrid cells. Four of the rats with low levels of CD4 and five of the rats with low levels of CD8 developed intrahepatic tumors. However, none of the animals in the control group developed tumors. What has been explained thus far appears to establish that in the induction phase, CD4 cells as well as CD8 cells are essential for the immunogenic effect of TBH (see Table 1). This result maintains a certain degree of logic with what is
74
known concerning the role that these subpopulations play in the effector mechanism. CD4s, taught by TBHs, recognize and promote the cytotoxic action of CD8 cells. Both are necessary. To corroborate this hypothesis, three additional groups of five rats each were used. All animals were immunized with a subcutaneous injection of 2x106 hybrid BERH-2-B cells. Two weeks after immunization, the animals were treated with monoclonal antibodies in a similar manner as in the previous experiment. Again, the degree of cell reduction in each population, according to each treatment, was verified by peripheral blood immunostaining. Three days after the last injection of antibodies, each rat was injected in the liver with 5x106 BERH-2 cells. Neither the group with low levels of CD4s nor those treated with non-specific antibodies developed tumors. However, all of the rats treated with anti-CD8 antibodies developed tumors. This result is coherent and corroborates the overall hypothesis, as CD4s remain as memory cells and are stimulated by tumor cell injection. However, the lack of CD8s impedes CD4s from inducing an immune response, and this allows cell proliferation through effector anergia.
Table 1 Immunization with BERH2-B, treatment with Antibody and Intrahepatic Injection of BERH-2
Specificity of the Antibody
Treatment with Antibody followed by Intrahepatic Injection of BERH-2-B
No
-
0/6
CD4
+
4/6
CD8
+
5/6
Control Antibody
+
0/6
Number of animals with tumors over the total animals treated
CD4
+
0/5
CD8
+
5/5
Antibody control
+
0/5
Finally, to corroborate the degree of specificity of the immune response generated by the TBH cells, the following experiment was designed. Two groups of eight rats each were subcutaneously immunized with 2x106 hybrid BERH-2-B cells. Two weeks later, the first group received an intrahepatic injection with 5x106 BERH-2 cells.
75
Figure 36: Development of the experimental GBM model using an intracerebral injection of 105 F98 cells in the brain of Fisher rats.implántela the animals are dead by day 32 post injection..
76
The second group was injected with an equal quantity, but subcutaneously with a transitional carcinoma cell line called NBT-II obtained from the American Type Culture Collection. As expected, none of the rats from the first group developed a tumor, but all of the rats in the second group developed a tumor from the NTB-II line and died within 45 days (Table 2).
Table 2 Number of animals with
Immunization cells
Challenge cells
BERH-2-B
BERH-2
0/8
BERH-2-B
NBT-II
8/8
tumors
100
Survival probability (%)
80
treatment Control TBH Thy+TBH Thymosine
60
40
20
0 0
10
20
30
40
50
60
Time in days
Survival time Endpoint
: Survival : Death
Factor codes : treatment Factor: Control Sample size: 12 Median surv.: 25.5
TBH 12 -
Thy+TBH 12 -
Thymosine 12 -
0.0 3.4
0.0 3.4
0.0 3.4
Comparison of survival curves Logrank test Endpoint-Observed n Expected n Chi-square DF Significance
12.0 1.9 = 65.1371 = 3 P < 0.0001
Figure 37: Survival study using Kaplan Meier analysis in Fisher rats with intracerebral implant of 105 F98 cells with respect to the treatment that was received (unpublished results).
77
To corroborate these findings in another animal model, we utilized a glioblastoma multiforme (GBM) model, which is the most aggressive and least immunogenic tumor known. After injecting Fisher rats in the brain, 100% of the animals died between days 28 and 32 because of complete occupation of the injected hemisphere (Figure 36). If the animals are treated with a TBH vaccine one week post-injection, they did not develop tumors within the next 60 days of observation (Figure 37).
78
The Clinical Application of Immunotherapies for Breast Cancer: Our Experience Materials and Methods Synopsis of the Clinically Utilized Protocol Therapeutic Indication Criteria 1. Patients of both sexes with any type of advanced tumor that had worsened after receiving therapies of verified effectiveness, and/or who had been intolerant of the immunotherapy and/or who did not qualify for any other therapy with an established efficiency. 2. Patients with a reliable diagnosis of neoplastic disease, as established by 2.1. A positive image detected by ultrasound (ultra), computed tomography (CT), nuclear magnetic resonance (NMR) and/or positron emission tomography (PET); 2.2. Positive images corroborated by a histopathological study. 3. Patients were accepted without limitation as to their clinical state. In all cases, performance was established by the ECOG (Eastern Oncology Cooperative Group) index. 4. Patients had to be of legal age or have written authorization of their legal representative. 5. Patients older than 18 years were required to weigh more than 40 kg and children were to be within normal percentiles. 6. The hematocrit was to be equal to or greater than 25%. 7. The peripheral leukocyte count was to be equal to or greater than 3000 /mm3. 8. The mean corpuscular volume (MCV) was to be equal to or greater than 70 fl. 9. The mean corpuscular hemoglobin (MCH) was to be equal to or greater than 21 pg.
79
10. The platelet count was to be lower than 500000 /mm3. 11. The patient was required to have addressed any renal, respiratory, liver, heart function or pre-existing metabolic condition. 12. The patient, or his/her legal representative, must have been appropriately informed and have accepted the receipt of treatment through the signed informed consent form. Therapeutic Non-indication Criteria 1. Patients who had received radiotherapy or chemotherapy during the 30 days prior to the start of the immunotherapy. 2. Patients with a second active neoplasm (excluding basal cell carcinoma). 3. Patients with a systemic viral or bacterial infection or visceral mycosis at the time that treatment with the proposed immunotherapy regimen was to begin. 4. Patients presenting with jaundice or liver failure. 5. Women who were pregnant or nursing. 6. Women of reproductive age who did not follow a non-hormonal contraceptive regimen during treatment. 7. Patients with uncontrolled addictions or psychiatric illnesses. 8. The presence of any other associated illness that in medical opinion could interfere with treatment or patient adherence to the therapy. 9. Refusal to sign the informed consent. Evaluation Prior to the Initial Phase of Treatment During the week prior to the beginning of the indicated treatment, patient checks were conducted. These checks included the compilation of a complete medical history, blood analysis, serological blood tests to detect hidden infections, heart function analysis studied by a specialist and including an enrollment electrocardiograph, respiratory function analysis using spirometry and an analysis of anti-tumor immune response through the lymphocyte proliferation test. The clinician, utilizing the ECOG scale, monitored the patient’s clinical performance at the beginning of treatment.
80
Therapeutic Procedures Elaboration and Application of the TBH and TIL Vaccines 1. Obtaining tumor cells According to the clinical condition of the patient, one of two different methods were used to obtain autologous tumor cells: surgical cell dissociation or cell culture following thin needle biopsy. The second procedure was usually performed under a tomography guide and less frequently under an ultrasound-scanning guide. Both types of image-guided biopsy were always performed under local anesthesia, general sedation and in sterile conditions. The cytology of the sample obtained was always checked to verify that the cells collected were tumor cells. The procedure was performed by an intervention radiologist, aided by an anesthesiologist, and as a group with the pathologist and a team clinician. After performing the biopsies, the patients were medicated with analgesic and gastroprotective drugs and observed while hospitalized (only if necessary) or as an outpatient for a total of 72 hours. Regardless of collection method, the material was always divided into two samples. One sample was processed for histological and histochemical tests, and the other for isolation and culture of tumor cells. Tumor cell cultures were produced in our maximum-security GMP quality biological laboratory, which was specialized for these needs. 2. Tumor cell and tumor infiltrating lymphocyte (TIL) isolation The biopsy tissue destined to produce TBHs was mechanically dissociated and passed through a 40-μ pore metallic netting until a suspension of individual cells was obtained. The cell suspension was separated on a Ficoll-Hypaque (clinical use CG) gradient and the mixture was centrifuged at 1200 rpm for 30 minutes. The precipitated mononuclear cells were the tumor cells, while those that remained in the interface ring were white mononuclear cells, among which the TILs were found.
81
The tumor cells were cultured in DMEM enriched with recombinant human insulin and recombinant human "epidermal growth factor". The TILs were cultured in DMEM + IL-2 + ranitidine + indomethacin to obtain another therapeutic product. 3. Obtaining the patient's mononuclear leukocytes The patient's mononuclear leukocytes were obtained through volume apheresis. Continuous flow cell separators (Fenwal CS3000 Plus or Cobe Spectra) were used along with a "buffy coat" retrieval program. In all cases, a blood volume equal to 1 patient volume was processed. To facilitate the procedure, if the patient did not possess veins of an adequate resistance and caliber, a 12 French double lumen catheter was placed in the central line. 4. Expansion, activation and selection of the B lymphocytes The buffy coat collection bag was processed in the GMP maximum biosafety laboratory. The buffy coat cells were washed with balanced salt solution (BSS) minus Ca2+ and Mg2+. The mononuclear cells were then separated by a FicollHypaque (CG) gradient and centrifuged at 1200 rpm for 30 minutes. The interface ring cells were washed with BSS minus Ca2+ and Mg2+. Subsequently, the cells were seeded in DMEM enriched with IL-4 and IL-6. On the third day, the cells were harvested and incubated in BSS with an antibody cocktail with affinity for all of the cells that could possibly be present in the buffy coat, except for B lymphocytes. Each antibody was united by its Fc portion with an anti-Teflon antibody, which was joined to a Teflon microsphere having magnetic iron dust in its structure. When the cells pre-incubated in this antibody cocktail were passed through a magnetic immunoselection column, all of the cells except the B lymphocytes were retained. The purity of the B lymphocyte suspension eluted from the column was between 92 and 98%. 5. TBH generation The B lymphocytes as well as the isolated tumor cells were harvested, and the cell suspensions were mixed in a 10:1 proportion, respectively. This cell suspension was treated with polyethylene glycol (PEG) for three minutes. Then, the cells were washed and cultured in DMEM with insulin, EGF and IL-6.
82
6. Quality control and assurance of the elaboration and cell production process Microbiological studies were performed to evaluate the identity, viability and biological power of samples from the different cells used in the elaboration process and the different cell products (apheresis bag, isolated cells, product of each cell culture and so forth). The microbiological tests investigated the presence of aerobic bacteria, anaerobic bacteria, mycobacteria, fungi and viruses. For positive tests, a decontamination protocol was applied, if possible. Otherwise, if it was not feasible to decontaminate the sample, or if the decontamination process failed, the biological product was eliminated. When possible, the entire process was repeated as a final safety measure. The biological tests included total cell count, investigation of cell viability through the trypan blue dye inclusion test, identification and quantification of the various cell populations through their morphological microscopic immunohistochemical analysis, supported by flow cytometry analysis, as appropriate. For example, to corroborate whether the TBH hybrid had formed, May Grünwald Giemsa and immunohistochemical staining were performed to determine whether a cell (binucleated or not) possessed both cytokeratin microfilaments in its cytoplasm and CD19 proteins on its cell membrane (see Appendix 1). 7. Intravenous infusion of cyclophosphamide Forty-eight hours prior to vaccination, the patient was intravenously infused with a dose of cyclophosphamide (350 mg/m2 of corporal surface) in an outpatient setting. 8. Intraganglionic immunization The immunizations were performed by intraganglionic injection of the vaccine. The lymph node should not be irradiated or show signs of neoplastic invasion (on palpitation). The immunizations were performed every three weeks, alternating which lymph node was injected for each immunization. The procedure can be summarized as follows: A non-irradiated lymphatic region was chosen that showed no indication of
83
neoplastic invasion upon palpating the zone and the local lymph nodes. The lymph nodes in the groin were preferred. By palpitation, a lymph node was located that was mobile and had a soft consistency. It was fixed utilizing the index and middle finger of the hand not in use. With the hand in use, gauze with 70% alcohol was passed over the skin that covered the chosen lymph node, as well as over the two fingers that held it. Then, the syringe containing the vaccine was taken. The syringe was required to be 1 cm3 or have a sufficient size so that it could contain the total volume of the vaccine suspension. The syringe was loaded and mounted with a 25 gauge or smaller caliber needle. Placing the needle parallel to the surface of the skin, it was depressed and entered the lymph node at one of the ends. Slight suction was applied to verify that we had not entered a blood vein. Then, the piston was pressed down gently to avoid any extraganglionic reflux of the vaccine. The dose of vaccine to be administered was required to be over 1x107 total cells. Preparation and Application of the TBH AAL Vaccine 9. Obtaining mononuclear leukocyte cells (MNCs) from the patient As soon as the patient had received the first two TBH immunizations, leukapheresis was performed. The cell product of this procedure is called the "buffy coat". It consists of an MNC-enriched suspension and a small proportion of granulocytes, red blood cells and platelets. The cell product was checked (see Appendix 1) to determine the quality, quantity and proportion of existing MNCs. The MNCs were washed with BSS without Ca2+ or Mg2+, separated on a FicollHypaque (CG) gradient and centrifuged at 1500 rpm for 30 minutes. The cells that formed the Ficoll-BSS interface ring were collected, washed, counted and checked. 10. MNC and TBH co-culture
84
The TBHs and MNCs (1:10) were co-cultured for three days at 37ºC, 5% CO2 in DMEM culture media enriched with rh-insulin, rh-EGF, rh-SCF, rh-TNF-α and rh-GMCSF. 11. Quality control and assurance of the cell product elaboration process The procedure was similar to that described in point 6. 12. Division into aliquots and preservation of the TBH AAL dose After performing the pertinent controls, the cells were divided into 5x108 cell aliquots. The aliquots were suspended in DMEM with 2% human albumin and 10% DMSO and frozen at -84ºC until use. 13. Intravenous infusion with cyclophosphamide The procedure was similar to that described in point 7. 14. Intraganglionic immunization The corresponding dose was thawed and washed to remove the DMSO solution. The product quality was checked and the procedure was similar to that described in point 8. Elaboration and Cytoimplant of the Mixed Leukocyte Culture (MLC) 15. Donor selection A donor of mononuclear cells was selected from among three possible donors. Each donor was required to meet the conditions of a blood donor, have good vein quality in the upper extremities, not be related, and in the case that the patient was a woman, not be her sexual partner. A mixed culture of unidirectional lymphocytes, with lymphocytes from each of the possible donors added to the patient's irradiated lymphocytes, was performed. The donor who met all of the selection criteria and who possessed the greatest incompatibility indices in the mixed lymphocyte culture was chosen. 16. Obtaining MNCs A leukapheresis was performed on the patient. The cell product of this procedure is called the "buffy coat". It consists of an MNC-enriched suspension and a small proportion of granulocytes, red blood cells and platelets. The cell product was checked (see Appendix 1) to determine the quality, 85
quantity and proportion of existing MNCs. The MNCs were washed with BSS without Ca2+ or Mg2+, separated on a FicollHypaque (CG) gradient and centrifuged at 1500 rpm for 30 minutes. The cells that formed the Ficoll-BSS interface ring were collected, washed, counted and checked. The patient's purified MNCs were irradiated with 2500 rad. 17. Mixed lymphocyte culture (MLC) The MNCs from the donor were mixed with the irradiated MNCs from the patient (10:1) and cultured in DMEM at 37ºC, 5% CO2 for 72 hours. At the end of this time, the culture was harvested and quantified, and the pertinent controls were performed. A minimum of 1x109 MLC cells was required. 18. MLC implant The MLCs were implanted inside the tumor during a surgical procedure or through a tomography- or ultrasound-guided puncture. In all cases, a pathologist's cytological report corroborated whether the site chosen for the puncture was adequate. After the implant, the puncture site or sites
were
sealed with a fibrin clot, using a commercial preparation (Tissucol®, Baxter– Immuno, Chicago, USA). Elaboration and Administration of the DC-TBH Autologous Vaccine 19. A TBH for each of the different types of metastasis and for the original tumor was developed according to the methodology described in points 1 to 6. 20. Mobilization of naive dendritic cells (DC) from bone marrow Over five consecutive days (days 1 to 5), the patient received 150 μg GM-CSF subcutaneously. The injection was performed after 19 hours. 21. Obtaining the patient's MNCs On day six, a leukapheresis was performed on the patient. The cell product of this procedure is called the "buffy coat". It consists of an MNC-enriched suspension and a small proportion of granulocytes, red blood cells and platelets. The cell product was checked (see Appendix 1) to determine the quality, 86
quantity and proportion of existing MNCs. The MNCs were washed with BSS without Ca2+ or Mg2+, separated on a FicollHypaque (CG) gradient and centrifuged at 1500 rpm for 30 minutes. The cells that formed the Ficoll-SSB interface ring were collected, washed, counted and checked. 22. DC isolation The MNCs were incubated with an antibody cocktail (StemSep® human DC enrichment cocktail, Stem Cell Technology, Vancouver). Each antibody recognized one of the diverse cell populations that composed the MNC suspension, except for the DCs. Each antibody was united by its Fc fraction to the Fc fraction of a second antibody that reacted with Teflon. After 30 minutes of incubation, the MNCs with the antibody cocktail were added to a Teflon micro-sphere suspension that contained a particular magnet in its structure. The marked MNCs were allowed to drain through a magnetic immunoselection column; all of the marked cells were retained while the DCs were drained. The DCs were then washed with BSS. 23. Activation of the DCs challenged with TBH The isolated DCs were co-cultured for three days with TBH (4:1) in DMEM enriched with TNF-α and GM-CSF. The cells were then harvested, quantified and appropriately controlled. 24. Quality control and assurance of the elaboration process and cell product The procedure was similar to that described in point 6. 25. Intravenous infusion of cyclophosphamide The procedure was similar to that described in point 7. 26. Intraganglionic immunization The procedure was similar to that described in point 8. Elaboration and Administration of the Specific Anti-tumor Autologous CD3+CD2527. When the patients were treated with anti-tumor cell vaccines, specific activated CD3+CD25- lymphocytes appeared in the blood, usually one week
87
after the first immunization, but the number of specific effector lymphocytes increased several-fold after the second immunization. For this reason, two volumes of CMNs were extracted by apheresis two weeks after the second immunization. 28. CD3+ lymphocytes were isolated through negative selection. They were cocultured for 48 hours with TBH in DMEM, without any other aggregate, in a 10:1 proportion. On the third day, through negative selection, the CD3+CD25lymphocytes were isolated and added to the IL-2 culture. On the fifth day, the lymphocytes processed in this way were harvested, counted and checked. 29. Quality control and assurance of the elaboration process and cell product The procedure was similar to that described in point 6. 30. Intravenous infusion of cyclophosphamide The procedure was similar to that described in point 7. 31. Intratumor implant of the CD3+CD25- lymphocyte suspension The procedure was similar to that described in point 18.
Treatment with Thymosin α1 (Tα1) Tα1 is a thymic hormone that selectively acts on cell immunity to modulate the effector response, primarily of Th1. It is used as a second line medication for patients with chronic hepatitis B and C who have not responded to treatment with ribavirin and interferon. Tα1 has shown to be highly effective at elevating cell immunity in oncology patients who have received or who are receiving aggressive chemotherapeutic treatment. Thus, it has been shown to be a powerful adjuvant for prophylactic antiviral vaccines. For this reason, it was decided to administer this immunomodulator to patients who showed signs of depressed cell immunity (mycosis, herpes zoster, having received very aggressive chemotherapy treatment and so forth). Consequently, those patients who failed to increase the absolute value of their lymphocyte proliferation index (LPA scores) by more than 10x103 cells/μl after the first immunization were treated with Tα1 (commercial name Zadaxin®, produced by SciClone
Pharmaceuticals).
One
vial
of
1.6
μg
was
administered
subcutaneously, two times per week, at 19 hours. 88
Therapeutic Safety Evaluation The safety evaluation of the selected immunotherapy was conducted according to what is described in the FDA Federal Regulation for the IND 21CFR 312.32 document [324]. The pertinent control checks were made on the patient first thing in the morning on the day in which she was to be immunized, and then at 24 and 96 hours after this first check. Establishing this system of control checks allowed study personnel to monitor patients and to take the necessary clinical measures for any early or late adverse event that could have occurred. The clinical exam included documentation of weight, core temperature, heart rate,
blood
pressure,
breathing
rate
and
digital
oximetry,
resting
electrocardiogram and laboratory analyses of the first morning urine and of a sample of 20 cm3 of blood. The clinical physician administered a structured questionnaire complemented by a general physical exam and a neurological examination. The clinical status of the patient was scored according to the ECOG scale. Thus, any treatment producing a change in cardiac, respiratory, hepatic renal, neurologic or internal indications could be eliminated. If the patient developed any signs compatible with a severe adverse event (SAE) such as acute hypersensitivity, anaphylaxis or systemic inflammatory response syndrome (SIRS), she would be treated with the appropriate measures and, if necessary, taken to the intensive care unit (ICU). SAE was defined as any adverse event that reached grade 3 or 4, as well as 2nd, 3rd, or 4th degree hypersensitivity reactions using the toxicity scale transcribed in Appendix 2. SAEs were reported to the biosafety committee within seven days of occurrence, and the biosafety committee would decide if the patient was able to continue with treatment. Clinical and laboratory data were recorded on a structured medical history form. All data were scored according to the pre-established scale described in Appendix 2. The absolute values of each item were added and the total resultant value was called the index of toxicity (IT). Using this index, a statistical evaluation could be conducted to ascertain whether there was any correlation between the scale and the clinical status of the patient (ECOG) prior to each 89
immunization, to the quantity of cells (dose) that comprised each immunization and to the number of doses already received. The comparisons were made using the Spearman’s rank correlation test.
Treatment Efficacy Evaluation As a measure of therapeutic efficacy, two parameters were considered: the development of an anti-tumor effector immune response and tumoral response. Immune Response To study the functional tumor-patient immune response, the patient's TBH cells were co-cultured with his or her peripheral mononuclear cells. The studies involving the patient's systemic immune response were performed on the day each new therapeutic cycle was initiated. The procedure, described previously, can be summarized as follows. Mononuclear leukocytes, isolated from a 20 cm3 sample of the patient's peripheral blood and concentrated to a volume of 1500 μl were seeded on a 96well plate. Each well received 100 μl of the mononuclear cell suspension in DMEM, 100 μl of DMEM with 5% albumin or autologous serum (one well for each condition; the first well represents a serum-free co-culture and the second well represents a serum co-culture) and 100 μl of a suspension of TBH in DMEM at a concentration 10 times lower than the mononuclear suspension. TBH cells were not added to the first two wells (one with serum and the other without serum), which were treated as controls. In the second two wells (one with serum and the other without serum), TBH cells (irradiated with 2500 rad of gamma radiation), which originated from another patient carrying a similar (homologous) tumor, were added. In the last two wells (one with serum and the other without serum), a suspension of autologous TBH cells (irradiated with 2500 rad of gamma radiation) were added. Each culture was made in duplicate. The approximate number of mononuclear leukocytes aliquotted into each well was 5x105 cells and 5x104 TBH cells. The study plate was incubated at 37ºC, in a filtered air environment, humidified and with 5% CO2 for 96 hours. The final number of cells was quantified using an automatic cell counter (Autocounter AC 9000 Swelab TM). The lymphocyte proliferation index (LPI) used to evaluate the immune response 90
was calculated by dividing the absolute value of the number of cells present in the cultures already challenged with homologous TBH or with autologous TBH without serum and with serum, divided by the absolute value of cells present in the TBH unchallenged cultures without serum and with serum, respectively. If the LPI was lower than 0.7, it was interpreted as a tolerant immune response. If the LPI was between 0.7 and 1.0, it was interpreted as a stable immune response and if the values were over 1.0, it was interpreted as an effector immune response. Tumoral Response Tumoral
response
is
defined
as
a
decrease
in
neoplastic
mass.
In general, to evaluate tumoral response after chemotherapy, radiotherapy or hormonotherapy, the RECIST criteria are utilized (a decrease of two diameters of the tumor mass taken in two comparable images).
A
C
B
D
Figure 38: A and B Pictures belong to the adenopaties of a NHL Patient previous to any treatment. C and D shows the therapeutic response after 3 DC-TBH immunizations.
A complete response (CR) exists if the tumor mass has diminished by 100% in both diameters. If the average reduction in both diameters is over 50%, but less than 100%, then the response is said to be partial (PR). If the reduction in
91
diameters is less than 50% but over 20%, it is considered minor remission (MR). If the variation in diameters is roughly lower than 20%, it is said that the patient presents stable disease (SD). If the diameters have increased more than 20%, it is said that the patient possesses progressive disease (PD). Figure 38 shows an example where a reduction in the tumor mass observed is easily evaluated by the application of the RECIST criteria using pre- and posttreatment images. Immunotherapy treatment first involves tumor inflammation, and then the interaction of the immune cells with the tumor, which can imply metabolic paralysis. Finally, there is a reduction in tumor mass. From this, it can be observed that the RECIST criteria cannot always be applied. Tumor inflammation can be confused with an increase in tumor tissue size. The arrest of tumor metabolism is one of the first steps preceding tumor destruction, but this cannot be detected by a TAC or an RMN. A second aspect that must be considered is the behavior observed regarding the plasma concentration of tumor markers. As illustrated in Figure 39, in a successful treatment, the monthly evaluation of the first immunization is characterized by a dramatic increase in tumor marker serum concentration. To understand this paradoxical phenomenon, we should analyze what occurs for other methods of tumor removal or reduction.
3000
2500
2000
1500
1000
500
0 Start
1st month
3rd month
6th month
12 month
18 month
Figure 39: CA 19-9 serum level variation of a patient with Pancreatic Cancer during its succesful immune therapy.
During the two to three weeks following successful surgery in which the entire
92
tumor has been removed, a serum increase in tumor markers is usually observed. These diminish by the fourth week post-surgery. It is hypothesized that a great number of oncocytes are destroyed during the tumor abscission, and the markers are liberated into the bloodstream. Their average lifespan would explain their serum persistence over three weeks. This destruction only occurs during the surgical procedure.
Figure 40: Partial remission of Advanced Breast Cancer hepatic metastasis treated with two cycles of DC TBH, as detected by PET.
Figure 41: Images A, B and C illustrate the appearance of a retroperitoneal leiomyosarcoma evaluated using TAC, PET and histology. Image C shows a solid tumor mass with no presence of immune cells. After three weeks of immunotherapy, the tomographic image (D) and that obtained by PET (E) have enlarged slightly. However, the two biopsies obtained from the periphery and the center (F and G, respectively) show that most of the tumor is composed of an
93
activated lymphomonocytic infiltrate presenting scarce tumor cells in the periphery of the histological image.
The tumor cytolysis generated by immune treatment is a prolonged event, usually lasting for several weeks. Thus, it is logical to deduce that the plasma washing of these substances can be represented by a graphic similar to Figure 39. Using positron emission tomography (PET) associated (or not) with the simultaneous execution of a TAC (PET-TAC) allows us to evaluate the metabolic response of tumors to immunotherapy action (Figure 40). Nevertheless, as illustrated in Figure 41, when tumor destruction is mediated by a strong infiltration of activated lymphomonocytes, the interpretation of the PET results are confusing. In fact, similar to what occurs in tuberculomas, the cellular immune response possesses SUV similar to what is produced by tumor activity. To evaluate the tumoral response against immunotherapy better, we have adopted the procedural algorithm illustrated in Figure 42. Axial Computed Tomography
Reduction in Tumor Size
No Reduction in Tumor Size
Tumor Markers
Reduction in Value
No Reduction in Value
Positron Emission Tomography
Reduction in SUV
No Reduction in SUV
Tumor Biopsy
Positive Cytology without Inflammation
Negative Cytology or Positive Cytology with Inflammation
Figure 42: Algorithm for studying the tumor response
94
Evaluation of Treatment Efficiency The efficiency of immunotherapy treatment was evaluated by disease-free survival and by overall patient survival. 1. Disease-Free Survival Time: This parameter was checked by monthly clinical exams and serological studies (including tumor markers). If any abnormality was detected during these checks, a medical imaging study was conducted (TAC, RMN, PET and so forth) to investigate the possibility of relapse or metastatic dissemination (progression of the disease). 2. Disease-free survival time was studied (for patients with the same type of neoplasm and treatment) using a Kaplan-Meier analysis and a two side log rank test with a Chi-squared test (when comparative studies were performed). 3. Overall survival: To evaluate overall survival, a specially assigned nurse recorded the survival of each patient every week during the first two years of monitoring. This check was performed monthly during the second year of processing.
Using the same criteria described for disease-free survival, the values collected from all patients, which were grouped by pathology and by therapy, were analyzed by Kaplan-Meier analysis, and the curves were compared using a twoside log rank test.
95
Results: Preparation of the Different Cell Products Autologous tumor cell culture: The patient's tumor cells can be obtained using surgical sections, percutaneous puncture biopsy and/or collection of pleural, pericardial or ascites effusions and cerebrospinal fluid (CSF). A sample of each metastasized tissue is taken as it has been shown that, most likely because of the different adhesion molecules that each one expresses, they possess independent immune responses (Figure 43).
Figure 43: Schema of TBH Elaboration
Except in the third case (effusions) the tissue should be dissociated to obtain a single cell suspension. To obtain this suspencion there are only utilized mechanical means to avoid the enzimatic action on the cell membrane receptors and adhesion molecules. The cells are seeded into tissue culture-grade plastic bottles with the appropriate optical quality and without any type of cover.
96
Figure 44: Mamosphere
Figure 45: Breast adenocarcinoma stem cells growing as mammospheres (yellow circles) and with a mesenchymal shape.
The medium used is chemically defined and contains pharmacological-grade recombinant human insulin aggregate, recombinant human epithelial growth factor, is free of pathogens and is of GMP quality. Through gradients, dilution cultures and successive purifications, it is possible to obtain a pure tumor epithelium culture, likely composed of stem cells. The 97
cells can grow in two different patterns, mammospheres (Figures 44 and 45) and mesenchymal cells. Both are positive for cytokeratin, but the second cell type is also positive for vimentin. TBH Preparation: Figure 43 illustrates the different procedures that should be performed to obtain activated B lymphocytes from the patient [12, 13, 325-327]. Briefly: apheresis of a patient blood sample should lead to the obtention of approximately 5x109 to 1x1010 total mononuclear cells. After passing through the Ficoll-Hypaque (CG) gradient, approximately 60% of this initial number (3x109 to 6x109 total cells) is retained. These cells are cultivated in DMEM + IL-4 + IL-6 to stimulate the growth and activation of B lymphocytes, and ranitidine and indomethacin to complicate Treg growth. At the end of three to five days, the cells are harvested and the activated B lymphocytes are purified by negative immunoselection using a commercial kit (see Figure 43). Approximately 3x108 to 6x108 activated B lymphocytes are obtained using this method. A mixed suspension with a proportion of four activated lymphocytes for each isolated tumor cell is prepared, and the suspension is treated with polyethylene glycol to obtain cell hybrids. Approximately 10 to 15% of all incubated cells are transformed into a TBH hybrid. This hybrid can be duplicated in vitro if it is cultured with IL-6, insulin and EGF. Each dose of vaccine is composed of 1x107 to 5x107 total cells. Preparation of the MLCs: The procedure is illustrated in Figure 46 [13, 326]. A healthy donor is selected who is not histocompatible with the patient. Leukapheresis is practiced on both the patient and the donor to obtain approximately 5x109 and 1x1010 total mononuclear cells from each. Both cell suspensions are separated on a Ficoll-Hypaque gradient. After centrifuging, the ring of cells is harvested from the Ficoll gradient-saline solution interface, from which approximately 3x109 to 6x109 total cells are collected, irradiated at 2500 rad and co-cultured in DMEM (with donor cells), without aggregates, for three days. The remaining aliquot is incubated in media such as the one previously described to obtain activated B lymphocytes in case the patient receives combined MLC + TBH or MLC + DC-TBH treatment (Figure 47).
98
Figure 46: MLC Elaboration and Treatment
Figure 47: MLC + TBH combined elaboration
When the MLCs are harvested, they are injected into the main tumor or its metastasis or metastases. The procedure can be performed by open puncture during a surgical procedure or by tomography- or ultrasound-guided puncture.
99
In either case, after having reached the tumor, a sample is extracted so that the pathologist, who is present, can corroborate the positivity of the site. A sample is also taken to develop a TBH together with the previously obtained B lymphocytes. Finally, the MLCs are injected into the tumor and the point of penetration is sealed with a surgical stitch or Tissucol. Preparation of the DC-TBH Vaccine: The procedure is illustrated in Figure 48 [14, 15, 328]. It is assumed that one or various TBHs with the patient's tumor or tumors have already been developed. The patient is then given 150 μg of GMCSF subcutaneously each day for five consecutive days at 19 hours.
Figure 48: DC-TBH elaboration and therapy
On day six, a leukapheresis is practiced using a buffy coat program to collect approximately 1x1010 to 3x1010 mononuclear cells total. The cells are passed through a Ficoll-Hypaque gradient (CG) and are then selected by negative immunoselection. Approximately 5% of the mononuclear cells cross the column (between 5x108 and 1.5x109 mononuclear cells, MNC). From these cells, only approximately 40% are DC, according to flow cytometry (Figure 49).
100
Figura 49: FACS DC concentration after negative selection
The TBH cells are then co-incubated in a 1:1 proportion with the MNCs enriched in Dendritic cells. The cells are cultured for three days in DMEM enriched with rh-TNF-α and rh-GM-CSF. On the fourth day, the cells are harvested and injected into a healthy lymph node. Each step described has been carefully studied. Figure 50 shows the results when five patients were treated for five days with G-CSF and the other five patients with GM-CSF. With G-CSF, the total number of white cells was elevated, at the expense of the granulocytes, from 5000 to 40000 cells/μl. The DCs increased the basal value by 2 to 3 times. However, the GM-CSFs raised the basal value of DC by 35 times, while the white cell total barely surpassed 10000 cells/μl. To estimate the time required to administer GM-CSF, the cytokine was administered to the patient for two, four, five and six consecutive days. According to Figure 51, the maximum number was achieved on the fifth day. On the sixth day, most likely because of leukocyte activation, the sample of MNCs showed greater cell-to-cell adhesion, which complicated negative selection.
101
Figure 50: Leukin pretreatment effect on yield of the DC-TBH vaccine
Finally, Figure 52 illustrates how the TNF-α + GM-CSF combination is more effective in obtaining greater performance and DC-TBH activity than the use of GM-CSF + IL-4 as other authors have proposed.
Figura 51: Tiempo óptimo de administración de GMCSF para la movilización de las DC.
102
Figura 52: Leukin Mixing influence on the yield DC-TBH elaboration
Preparation of CD3+CD25- autologous anti-tumor effector lymphocytes: The procedure is illustrated in Figure 53 [317]. It is assumed that one or various TBHs have been developed with one or more of the patient's tumors and that he or she has been immunized, whether with the TBH or DC-TBH vaccine, having received at least two consecutive immunizations with an interval of three weeks between each immunization. A leukapheresis is performed on the patient, from whom a product with 1x1010 to 2x1010 MNCs is obtained. After the cells are purified and concentrated, the buffy coat is incubated for four days in the presence of TBH. At the end of that time, the CD3+CD25- cells, usually 10% of all MNCs in culture, are isolated by immunomagnetic selection. They are concentrated and administered by slow endovenous infusion over eight hours, or they are injected into a remaining tumor mass. The intra-tumor injection site is studied by a pathologist. If the biopsy shows an abundant infiltration of activated lymphocytes, local injection should be avoided and the vaccine should be administered systemically.
103
Figure 53: Elaboration of the Treg vaccine (TregV) and CD3+CD25- cell suspension
Preparation of the Treg V vaccine: The procedure is illustrated in Figure 53 [318] and is similar to the procedure described for obtaining CD3+CD25lymphocytes. The main difference is that a separation of CD4+ cells is performed followed by a second separation, this time with anti-CD25. The lymphocytes retained in the second column are cultured for three days in DMEM with IL-2 and are then harvested, irradiated with 2500 rad and administered by endovenous infusion at a dose of 1x108 cells per infusion.
104
Results: Safety of the Immunotherapies Utilized Study of the safety of TBH immunization: Twenty-one patients were treated between 1992 and 1994 that had advanced tumors of different origin that were no longer responsive to any therapy. Each patient, under compassionate use therapy, received between 3 and 10 immunizations with approximately 5x107 TBH cells. Any adverse events observed were evaluated according to the scale established by the WHO for biological response modifiers and were noted in each patient's medical record. These results were published in 1995; the data were extracted and are summarized in Table 3. Initial
Cutaneous Reaction
Renal Toxicity
Neurol. Toxicity
Sepsis
CV&R Toxicity
G.I. Toxicity
Hemato. Toxicity
Defect. Coagul.
NG HC LG NB CB LR NV LP AB NT JV JN RC SL CC JP SH SC FA PM GA
+ + + + + + + +
+ + + + + + + + -
+ + + + + + + + + + + +
-
+ + + + -
+ + + + + + + -
-
-
Table 3: Adverse events observed following TBH immunization. References: - No or without changes; + Mild; ++ Moderate; +++ Severe
The immunization plan was not completed in three patients because of the fast evolution of their disease during the two first months of treatment. The collateral effects were limited to pseudo-flu symptoms with exhaustion and fever (grade 2). Patients seldom had a fever reaching 38.5ºC. The effects described generally ceased within the first eight weeks after initiating treatment. Cases of neutropenia, anemia or thrombocytopenia were not observed. The normal lifestyle of the patient was not altered, except for occasional pain due to a local reaction at the injection site that generated an aseptic abscess that was treated 105
with cold, anti-inflammatory drugs and, occasionally, drainage. The resolution of these abscesses was occasionally prolonged for two or three months. Each reaction was biopsied and the presence of tumor cells was never found. This local reaction appears to be related to a better therapeutic response. A few patients presented symptoms of grade 2 hypotension (the systolic pressure was not lower than 70 mmHg) that resolved with rest and reinstatement of fluids and without pharmacological treatment. In only one case was hospitalization for 48 hours necessary. Later, after the use of LPS to activate B lymphocytes was suppressed, the emergence of local abscesses was no longer observed; however, a transitory local inflammation was observed between the second and fourth immunization. This treatment has been applied to approximately 375 patients. Of this total, a fever between 37 and 38ºC (grade 2) was observed in less than 20% of the patients. In approximately 50% of patients, however, a pseudo-flu syndrome with spontaneous resolution was observed after the second and third immunization. In five patients who had different autoimmune conditions, grave worsening of these pre-existing conditions was observed. They resolved after spacing out the interim between immunizations. In a diabetic patient affected by an osteosarcoma with lymph node and pulmonary metastasis, a loss of control of glucose levels (grade 3) and anemia (grade 2) resulting from spleen hemocatheresis was observed. After suspending his immunization plan, the patient experienced a total remission of his osteosarcoma without receiving any other medication. Safety study of MLC and MLC-TBH immunization: The safety of immunization with the administration of a single MLC was reported by Chang and colleagues; Tables 4 and 5 are taken from that original study [312]. The author injected vaccine by transgastric puncture into the primary tumor of eight patients with unresectable pancreatic adenocarcinoma, four with stage 2, three with stage 3 and only one in stage 4. From an initial view of the adverse events table, it can be inferred that the treatment did not generate adverse systematic events, but did generate local events through tumor inflammation. All events were resolved when the tumor mass was reduced.
106
Table 4: Adverse events and MLC dose received by the patients. Table taken from the work of Chang and colleagues [312].
Table 5: Serious adverse events and manner of resolution experienced by each patient. Table taken from the work of Chang and colleagues [312].
Our experience with an identical treatment applied alone or in TBH immunization in patients with pancreatic tumors, post GBM resection, patients with advanced breast tumors and other advanced tumors, was similar [13, 324327]. Above grade 2, hyperpyrexia was observed in less than 10% of treated patients, and the symptoms were circumscribed to those produced by local inflammation. This local inflammation is dangerous when observed in patients with GBM because it generates grade 2 and grade 3 encephalitis that must be treated with steroids. These events caused us not to use MLC for patients with GBM. However, in grade 2 breast mastitis, patients did not require any other treatment beyond symptomatic treatment. Two patients with advanced breast adenocarcinoma and bony metastasis as the only place of disease establishment, presented grade 2 pancytopenia at the time of being treated. Their symptoms were temporarily increased after 107
treatment with the cytoimplant, but both experienced a PR and CR, respectively. No one required specific treatment and the symptoms reversed within the following two weeks. Within six hours of receiving the cytoimplant, a patient with pancreatic adenocarcinoma with liver metastasis and previous irradiation surgery of the zone presented with a severe gastrointestinal bleed (grade 3) in the lesion. This condition was reversed after the administration of plasma, platelets, two units of blood via transfusion and inhibition of the proton pump. Grade 3 respiratory distress syndrome was observed in two patients treated with MLC + TBH. Both patients had previously had an interstitial metastatic lung infiltration and had to be admitted to the ICU where they recovered. The administration of two MLCs to the same patient was performed in six patients with stage 4 pancreatic adenocarcinoma. All patients had an important systemic inflammatory reaction after the second immunization (four grade 2 and two grade 3). This was attributed to the change in tumor stroma that, from the prior anti-inflammatory stage to the first cytoimplant, later acquired proinflammatory characteristics. The two patients with grade 3 SIRS required hospitalization, and although their life expectation was doubled, the gravity of the adverse events observed led to a contraindication of the repetition of MLCs. DC-TBH immunization safety study: Twenty patients, treated on a compassionate basis and affected by diverse terminal neoplasms for which there was no standard therapeutic available, were treated with tumor autovaccines composed of autologous dendritic cells challenged in vitro with autologous TBH (DC-TBH). The 20 patients received 51 vaccines total and the collateral events observed are summarized in Table 6. The adverse events that could occur were evaluated using 127 clinical and laboratory parameters (Appendix 3) on the day of the immunization and 24 to 96 hours after immunization ("Common terminology criteria for adverse events, v3.0" (CTCAE) (http://ctep.cancer.gov/forms/CTCAEv3.pdf) developed by the United States NCI to evaluate biological therapies).
108
Total Local Injection Site Reaction Induration Ulceration Limb Oedema Gait-Walking Alteration Systemic Pain Geni-Urinary Obstruct. Anorexia Fever Chills Flu Like Syndrome Nauseas Vomiting Dehidrat. Liver Dysf. Dyspepsia Constipation Blood WBC Platelets Haemoglob Granulcytes Lymphocyte Clotting Fibrinogen RIN KPTT Liver Function Bilirubin ASOT ASPT ALP GGT LDH Renal Function Creatinine Proteinurie Metabolism Glucose Amylase Calcium Magnesium Albumin
Grade 1
Grade 2
Grade 3
AP/TP (1)
PE/TE (2)
AP/TP (1)
PE/TE (2)
AP/TP (1)
PE/TE (2)
AP/TP (1)
PE/TE (2)
10/20
19/51
5/20
9/51
4/20
9/51
1/20
1/51
13/20 1/20
25/51 1/51
12/20
24/51
1/20 1/20
1/51 1/51
3/20
5/51
2/20
3/51
1/20
2/51
3/20
3/51
2/20
2/51
1/20
1/51
12/20
15/51
4/20
6/51
1/20
1/51
5/20 3/20 2/20
8/51 5/51 4/51
4/20 3/20 1/20
6/51 5/51 2/51
1/20
2/51
1/20
2/51
1/20
2/51
1/20
1/51
1/20
1/51
5/20 4/20 3/20 1/20 1/20 4/20
8/51 6/51 4/51 1/51 1/51 4/51
3/20 2/20 1/20
4/51 2/51 1/51
1/20 1/20 1/20 1/20 1/20
1/20 4/20 14/20 1/20 15/20
1/51 6/51 16/51 1/51 20/51
1/20 3/20 6/20 1/20 3/20
1/51 4/51 8/51 1/51 7/51
2/20 12/20 4/20
2/51 14/51 5/51
2/20 12/20 3/20
6/20 8/20 5/20 9/20 6/20 10/20
6/51 12/51 7/51 12/51 8/51 12/51
3/20 2/20 8/20 6/20 11/20 9/20 9/20
7/20
8/51 1/20
1/51
2/51 2/51 2/51 1/51 1/51
1/20 1/20 1/20
2/51 2/51 1/51
5/20
4/51
3/20
4/51
7/20
7/51
5/20
6/5
2/51 14/51 4/51
1/20
1/51
2/20 7/20 4/20 5/20 4/20 8/20
2/51 10/51 6/51 7/51 5/51 10/51
3/20 1/20 1/20 2/20 1/20 1/20
3/51 2/51 1/51 3/51 1/51 1/51
1/20
1/51
2/20 1/20 1/20
2/51 2/51 1/51
3/51 2/51
3/20 2/20
3/51 2/51
10/51 6/51 15/51 11/51 9/51
6/20 4/20 10/20 8/20 7/20
7/51 4/51 14/51 10/51 8/51
1/20
2/51
1/20 1/20 1/20 1/20 1/20
1/51 1/51 1/51 1/51 1/51
1/20
1/51
Grade 4 AP/TP (1)
PE/TE (2)
1/20
1/51
5/20 4/20 3/20 1/20 1/20 4/20
8/51 6/51 4/51 1/51 1/51 4/51
1/20
1/51
Table 6: Adverse events due to DC-TBH. (1) AP/TP: number of patients affected/total number of patients treated. (2) PE/TE: Number of positive events/total number of feasible events. The initial ECOG index and the dose of cells administered were considered as factors that could influence adverse events that could occur after each immunization.
109
The total number of adverse events observed appeared to be linked to the local inflammatory reaction in the different neoplastic masses. To evaluate if the quantity of cells utilized in each immunization and/or the clinical status of the patient prior to each immunization had an influence on the adverse events observed, the sum of the partial scores of each item was evaluated. Thus, an index of toxicity for each immunization was established. It was not observed that the toxicity of any immunization was associated with the number of DCs administered by immunization (Figure 54). 70 60
Toxicity
50 40 30 20 10 0 1000000
100000000
100000000
Doses Sample size = 51 Correlation coefficient r = -0,1562 P=0,2737 95% Confidence interval for r = -0,4140 to 0,1248 Figure 54: Different cell doses (expressed as logarithms) applied to each immunization and the index of adverse events for each immunization.
However, the initial clinical state of the patient (measured by the ECOG index) did have a direct relationship on the adverse events observed. The statistical evaluation shows a highly significant correlation between the ECOG prior to each immunization and the index of adverse events later obtained for each event (Figure 55).
110
70 60
Toxicity
50 40 30 20 10 0 1.0
1.5
2.0
2.5
3.0
ECOG Sample size = 51 Spearman's coefficient of rank correlation (rho) = 0.609 P1 mm but ≤5 mm in greatest dimension.
T1b
Tumor >5 mm but ≤10 mm in greatest dimension.
T1c
Tumor >10 mm but ≤20 mm in greatest dimension.
167
T2
Tumor >20 mm but ≤50 mm in greatest dimension.
T3
Tumor >50 mm in greatest dimension.
T4
Tumor of any size with direct extension to the chest wall and/or to the skin (ulceration or skin nodules).c
T4a
Extension to the chest wall, not including only pectoralis muscle adherence/invasion.
T4b
Ulceration and/or ipsilateral satellite nodules and/or edema (including peau d'orange) of the skin, which do not meet the criteria for inflammatory carcinoma.
T4c
Both T4a and T4b.
T4d
Inflammatory carcinoma.
DCIS = ductal carcinoma in situ; LCIS = lobular carcinoma in situ. a
Reprinted with permission from AJCC: Breast. In: Edge SB, Byrd DR, Compton
CC, et al., eds.: AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer, 2010, pp 347–76 b
The T classification of the primary tumor is the same regardless of whether it is
based on clinical or pathologic criteria, or both. Size should be measured to the nearest millimeter. If the tumor size is slightly less than or greater than a cutoff for a given T classification, it is recommended that the size be rounded to the millimeter reading that is closest to the cutoff. For example, a reported size of 1.1 mm is reported as 1 mm, or a size of 2.01 cm is reported as 2.0 cm. Designation should be made with the subscript "c" or "p" modifier to indicate whether the T classification was determined by clinical (physical examination or radiologic) or pathologic measurements, respectively. In general, pathologic determination should take precedence over clinical determination of T size. c
Invasion of the dermis alone does not qualify as T4.
Table 2. Regional Lymph Nodes (N)a Clasificación por examen clínico NX
Regional lymph nodes cannot be assessed (e.g., previously removed).
N0
No regional lymph node metastases.
N1
Metastases to movable ipsilateral level I, II axillary lymph node(s).
N2
Metastases in ipsilateral level I, II axillary lymph nodes that are clinically fixed or matted.
168
OR Metastases in clinically detectedbipsilateral internal mammary nodes in the absence of clinically evident axillary lymph node metastases. N2a
Metastases in ipsilateral level I, II axillary lymph nodes fixed to one another (matted) or to other structures.
Metastases only in clinically detectedb ipsilateral internal mammary nodes N2b and in the absence of clinically evident level I, II axillary lymph node metastases. Metastases in ipsilateral infraclavicular (level III axillary) lymph node(s) with or without level I, II axillary lymph node involvement. OR N3
Metastases in clinically detectedb ipsilateral internal mammary lymph node(s) with clinically evident level I, II axillary lymph node metastases. OR Metastases in ipsilateral supraclavicular lymph node(s) with or without axillary or internal mammary lymph node involvement.
N3a Metastases in ipsilateral infraclavicular lymph node(s). N3b
Metastases in ipsilateral internal mammary lymph node(s) and axillary lymph node(s).
N3c Metastases in ipsilateral supraclavicular lymph node(s). a
Reprinted with permission from AJCC: Breast. In: Edge SB, Byrd DR, Compton
CC, et al., eds.: AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer, 2010, pp 347–76. b
Clinically detected is defined as detected by imaging studies (excluding
lymphoscintigraphy) or by clinical examination and having characteristics highly suspicious for malignancy or a presumed pathologic macrometastasis based on fine needle aspiration biopsy with cytologic examination. Confirmation of clinically detected metastatic disease by fine needle aspiration without excision biopsy is designated with an (f) suffix, for example, cN3a(f). Excisional biopsy of a lymph node or biopsy of a sentinel node, in the absence of assignment of a pT, is classified as a clinical N, for example, cN1. Information regarding the confirmation of the nodal status will be designated in site-specific factors as clinical, fine needle aspiration, core biopsy, or sentinel lymph node biopsy.
169
Pathologic classification (pN) is used for excision or sentinel lymph node biopsy only in conjunction with a pathologic T assignment. Table 3. Pathologic (pN)a,b Clasificación por examen anatomo patológico pNX
Regional lymph nodes cannot be assessed (e.g., previously removed or not removed for pathologic study).
pN0
No regional lymph node metastasis identified histologically.
Note: ITCs are defined as small clusters of cells ≤0.2 mm, or single tumor cells, or a cluster of 0.2 mm and/or >200 cells but none >2.0 mm).
pN1a
Metastases in 1–3 axillary lymph nodes, at least one metastasis >2.0 mm.
pN1b
Metastases in internal mammary nodes with micrometastases or macrometastases detected by sentinel lymph node biopsy but not clinically detected.c
pN1c
Metastases in 1–3 axillary lymph nodes and in internal mammary lymph nodes with micrometastases or macrometastases detected by sentinel lymph node biopsy but not clinically detected.
pN2
Metastases in 4–9 axillary lymph nodes. OR
170
Metastases in clinically detectedd internal mammary lymph nodes in the absence of axillary lymph node metastases. pN2a
Metastases in 4–9 axillary lymph nodes (at least 1 tumor deposit >2 mm).
pN2b
Metastases in clinically detectedd internal mammary lymph nodes in the absence of axillary lymph node metastases.
pN3
Metastases in ≥10 axillary lymph nodes. OR Metastases in infraclavicular (level III axillary) lymph nodes. OR Metastases in clinically detectedc ipsilateral internal mammary lymph nodes in the presence of one or more positive level I, II axillary lymph nodes. OR Metastases in >3 axillary lymph nodes and in internal mammary lymph nodes with micrometastases or macrometastases detected by sentinel lymph node biopsy but not clinically detected.c OR Metastases in ipsilateral supraclavicular lymph nodes.
pN3a
Metastases in ≥10 axillary lymph nodes (at least 1 tumor deposit >2.0 mm). OR Metastases to the infraclavicular (level III axillary lymph) nodes.
pN3b
Metastases in clinically detectedd ipsilateral internal mammary lymph nodes in the presence of one or more positive axillary lymph nodes; OR Metastases in >3 axillary lymph nodes and in internal mammary lymph nodes with micrometastases or macrometastases detected by sentinel lymph node biopsy but not clinically detected.c
pN3c
Metastases in ipsilateral suprclavicular lymph nodes.
Posttreatment ypN –Posttreatment yp "N" should be evaluated as for clinical (pretreatment) "N" methods above. The modifier "sn" is used only if a sentinel node evaluation was
171
performed after treatment. If no subscript is attached, it is assumed that the axillary nodal evaluation was by AND. –The X classification will be used (ypNX) if no yp posttreatment SN or AND was performed. –N categories are the same as those used for pN. AND = axillary node dissection; H&E = hematoxylin and eosin stain; IHC = immunohistochemical; ITC = isolated tumor cells; RT-PCR = reverse transcriptase/polymerase chain reaction. aReprinted with permission from AJCC: Breast. In: Edge SB, Byrd DR, Compton CC, et al., eds.: AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer, 2010, pp 347–76. bClassification is based on axillary lymph node dissection with or without sentinel lymph node biopsy. Classification based solely on sentinel lymph node biopsy without subsequent axillary lymph node dissection is designated (sn) for "sentinel node," for example, pN0(sn). c"Not clinically detected" is defined as not detected by imaging studies (excluding lymphoscintigraphy) or not detected by clinical examination. d"Clinically detected" is defined as detected by imaging studies (excluding lymphoscintigraphy) or by clinical examination and having characteristics highly suspicious for malignancy or a presumed pathologic macrometastasis based on fine-needle aspiration biopsy with cytologic examination. Table 4. Distant Metastases (M)a a
Reprinted with permission from AJCC: Breast. In: Edge SB, Byrd DR, Compton
CC, et al., eds.: AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer, 2010, pp 347–76. M0
No clinical or radiographic evidence of distant metastases. No clinical or radiographic evidence of distant metastases, but
cM0(i+)
deposits of molecularly or microscopically detected tumor cells in circulating blood, bone marrow, or other nonregional nodal tissue that are ≤0.2 mm in a patient without symptoms or signs of metastases.
172
M1
Distant detectable metastases as determined by classic clinical and radiographic means and/or histologically proven >0.2 mm.
Posttreatment yp M classification. The M category for patients treated with neoadjuvant therapy is the category assigned in the clinical stage, prior to initiation of neoadjuvant therapy. Identification of distant metastases after the start of therapy in cases where pretherapy evaluation showed no metastases is considered progression of disease. If a patient was designated to have detectable distant metastases (M1) before chemotherapy, the patient will be designated as M1 throughout.[1] Table 5. Anatomic Stage/Prognostic Groupsa, b Stage
T
N
M
0
Tis
N0
M0
IA
T1b
N0
M0
T0
N1mi
M0
T1b
N1mi
M0
T0
N1c
M0
T1b
N1c
M0
T2
N0
M0
T2
N1
M0
T3
N0
M0
T0
N2
M0
T1b
N2
M0
T2
N2
M0
T3
N1
M0
T3
N2
M0
T4
N0
M0
T4
N1
M0
T4
N2
M0
IIIC
Any T
N3
M0
IV
Any T
Any N
M1
IB
IIA
IIB
IIIA
IIIB
173
a
Reprinted with permission from AJCC: Breast. In: Edge SB, Byrd DR,
Compton CC, et al., eds.: AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer, 2010, pp 347–76. b c
T1 includes T1mi.
T0 and T1 tumors with nodal micrometastases only are excluded from Stage
IIA and are classified Stage IB. –M0 includes M0(i+). –The designation pM0 is not valid; any M0 should be clinical. –If a patient presents with M1 prior to neoadjuvant systemic therapy, the stage is considered Stage IV and remains Stage IV regardless of response to neoadjuvant therapy. –Stage designation may be changed if postsurgical imaging studies reveal the presence of distant metastases, provided that the studies are carried out within 4 months of diagnosis in the absence of disease progression and provided that the patient has not received neoadjuvant therapy. –Postneoadjuvant therapy is designated with "yc" or "yp" prefix. Of note, no stage group is assigned if there is a complete pathologic response (CR) to neoadjuvant therapy, for example, ypT0ypN0cM0. References 1. Breast. In: Edge SB, Byrd DR, Compton CC, et al., eds.: AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer, 2010, pp 347-76. 2. Singletary SE, Allred C, Ashley P, et al.: Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 20 (17): 3628-36, 2002. 3. Woodward WA, Strom EA, Tucker SL, et al.: Changes in the 2003 American Joint Committee on Cancer staging for breast cancer dramatically affect stage-specific survival. J Clin Oncol 21 (17): 3244-8, 2003. 4.
174
Breast Cancer Treatment (PDQ®) Health Professional Version Stage IIIB, Inoperable IIIC, IV, Recurrent, and Metastatic Breast Cancer Inoperable Stage IIIB or IIIC or Inflammatory Breast Cancer Multimodality therapy delivered with curative intent is the standard of care for patients with clinical stage IIIB disease. In a retrospective series, approximately 32% of patients with ipsilateral supraclavicular node involvement and no evidence of distant metastases (pN3c) had prolonged disease-free survival (DFS) at 10 years with combined modality therapy.[1] Although these results have not been replicated in another series, this result suggests such patients should be treated with the same intent. Initial surgery is generally limited to biopsy to permit the determination of histology, estrogen-receptor (ER) and progesterone-receptor (PR) levels, and human epidermal growth factor receptor 2 (HER2/neu) overexpression. Initial treatment with anthracycline-based chemotherapy and/or taxane-based therapy is standard.[2,3] In one series of 178 patients with inflammatory breast cancer, DFS was 28% at 15 years with a combined-modality approach.[2] [Level of evidence: 3iiiDii] For patients who respond to neoadjuvant chemotherapy, local therapy may consist of total mastectomy with axillary lymph node dissection followed by postoperative radiation therapy to the chest wall and regional lymphatics. Breast-conserving therapy can be considered in patients with a good
partial
or
complete
response
to
neoadjuvant
chemotherapy.[3]
Subsequent systemic therapy may consist of further chemotherapy. Hormone therapy should be administered to patients whose tumors are ER-positive or unknown. All patients should be considered candidates for clinical trials to evaluate the most appropriate fashion in which to administer the various components of multimodality regimens. Stage IV, Recurrent, and Metastatic Breast Cancer Recurrent breast cancer is often responsive to therapy, though treatment is rarely curative at this stage of disease. Patients with localized breast or chest wall recurrences, however, may be long-term survivors with appropriate
175
therapy. Prior to treatment for recurrent or metastatic cancer, restaging to evaluate extent of disease is indicated. Cytologic or histologic documentation of recurrent or metastatic disease should be obtained whenever possible. The ER and PR levels, HER2/neu positivity at the time of recurrence, and previous treatment should be considered, if known, when selecting therapy. ER status may change at the time of recurrence. In a single small study by the Cancer and Leukemia Group B (MDA-MBDT-8081 6), 36% of hormone receptor–positive tumors were found to be receptor negative in biopsy specimens isolated at the time of recurrence.[4] Patients in this study had no interval treatment. If ER and PR status is unknown, then the site(s) of recurrence, disease-free interval, response to previous treatment, and menopausal status are useful in selecting chemotherapy or hormone therapy.[5] Recurrent local-regional breast cancer Patients with local-regional breast cancer recurrence may become long-term survivors with appropriate therapy. A clinical trial indicated that between 10% and 20% of patients will have locally recurrent disease in the breast between 1 and 9 years after breast-conservation surgery plus radiation therapy.[6] Nine percent to 25% of these patients will have distant metastases or locally extensive disease at the time of recurrence.[7-9] Patients with local-regional recurrence should be considered for further local treatment (e.g., mastectomy). In one series, the 5-year actuarial rate of relapse for patients treated for invasive recurrence after initial breast conservation and radiation therapy was 52%.[8] A phase III randomized study showed that local control of cutaneous metastases could be achieved with the application of topical miltefosine; however, the drug is not currently available in the United States.[10][Level of evidence: 1iiDiii] Local chest wall recurrence following mastectomy is usually the harbinger of widespread disease, but, in a subset of patients, it may be the only site of recurrence. For patients in this subset, surgery and/or radiation therapy may be curative.[11,12] Patients with chest wall recurrences of less than 3 cm, axillary and internal mammary node recurrence (not supraclavicular, which has a poorer survival), and a greater than 2-year disease-free interval prior to recurrence have the best chance for prolonged survival.[12] The 5-year DFS
176
rate in one series of such patients was 25%, with a 10-year rate of 15%.[13] The local-regional control rate was 57% at 10 years. Systemic therapy should be considered in patients with local regional recurrence caused by the high risk of subsequent metastases.[14] No randomized controlled studies are available to guide patient care in this situation. Stage IV and metastatic disease - Systemic disease Treatment for systemic disease is palliative in intent. Goals of treatment include improving quality of life and prolongation of life. Although median survival has been reported to be 18 to 24 months,[15] some patients experience long-term survival. Among patients treated with systemic chemotherapy at a single institution between 1973 and 1982, 263 patients (16.6%) achieved complete responses. Of those, 49 patients (3.1% of the total group) remained in complete remission for more than 5 years, and 26 patients (1.5%) were still in complete remission at 16 years.[16][Level of evidence: 3iiDiii] Treatment of metastatic breast cancer will usually involve hormone therapy and/or chemotherapy with or without trastuzumab. Radiation therapy and/or surgery may be indicated for patients with limited symptomatic metastases. All patients with metastatic or recurrent breast cancer should be considered candidates for ongoing clinical trials. Surgery Surgery may be indicated for selected patients. Examples include patients who need mastectomies for fungating/painful breast lesions, parenchymal brain or vertebral metastases with spinal cord compression, isolated lung metastases, pathologic (or impending) fractures, or pleural or pericardial effusions. (Refer to the PDQ summary on Pain 7 for more information; for information on pleural and pericardial effusions, refer to the PDQ summary on Cardiopulmonary Syndromes 8.) Radiation therapy Radiation therapy has a major role in the palliation of localized symptomatic metastases. Indications include painful bony metastases, unresectable central nervous system metastases (i.e., brain, meningeal, and spinal cord), bronchial obstruction, and fungating/painful breast or chest wall lesions. Radiation therapy 177
should also be given following surgery for decompression of intracranial or spinal cord metastases and following fixation of pathologic fractures. Clinical trials (including the completed Radiation Therapy Oncology Group's trial [RTOG-9714 9]) are exploring the optimal radiation fractionation schedule. Strontium 89, a systemically administered radionuclide, can be administered for palliation of diffuse bony metastases.[17,18] Systemic Therapy Bisphosphonates The use of bisphosphonates to reduce skeletal morbidity in patients with bone metastases should be considered.[19] Results of randomized trials of pamidronate and clodronate in patients with bony metastatic disease show decreased skeletal morbidity.[20-22][Level of evidence: 1iC] Zoledronate has been at least as effective as pamidronate.[23] (Refer to the PDQ summary on Pain 12 for more information on bisphosphonates.) Hormone therapy Hormone therapy should generally be considered as initial treatment for a postmenopausal patient with newly diagnosed metastatic disease if the patient’s tumor is ER-positive, PR-positive, or ER/PR-unknown. Hormone therapy is especially indicated if the patient’s disease involves only bone and soft tissue and the patient has either not received adjuvant antiestrogen therapy or has been off such therapy for more than 1 year. While tamoxifen has been used in this setting for many years, several randomized trials suggest equivalent or superior response rates and progression-free survival for the aromatase inhibitors compared to tamoxifen.[24-26][Level of evidence: 1iiDiii] In a metaanalysis that included randomized trials in patients who were receiving an aromatase inhibitor as either their first or second hormonal therapy for metastatic disease, those who were randomized to a third-generation drug (anastrozole, letrozole, exemestane, or vorozole) lived longer (HR for death = 0.87; 95% CI, 0.82–0.93) than those who received standard therapy (tamoxifen or a progestational agent).[27][Level of evidence: 1iA] Several randomized but underpowered trials have tried to determine if combined hormone therapy (luteinizing hormone-releasing hormone [LHRH]
178
agonists + tamoxifen) is superior to either approach alone in premenopausal women. Results have been inconsistent.[28-30] The best study design compared buserelin (an LHRH agonist) versus tamoxifen versus the combination in 161 premenopausal women with hormone receptor–positive tumors.[31] Patients receiving buserelin and tamoxifen had a significantly improved median survival of 3.7 years compared with those receiving tamoxifen or buserelin who survived 2.9 and 2.5 years, respectively (P = .01).[31][Level of evidence: 1iiA] Very few women in this trial received adjuvant tamoxifen, which makes it difficult to assess whether these results are applicable to women who relapse after adjuvant tamoxifen. Women whose tumors are ER-positive or unknown, with bone or soft tissue metastases only, who have received an antiestrogen within the past year, should be given second-line hormone therapy. Examples of second-line hormone therapy in postmenopausal women include selective aromatase inhibitors, such as anastrozole, letrozole, or exemestane; megestrol acetate; estrogens; androgens;[32-40] and the ER down-regulator, fulvestrant.[41,42] In comparison to megestrol acetate, all three currently available aromatase inhibitors have demonstrated, in prospective randomized trials, at least equal efficacy and better tolerability.[32-38,43] In a meta-analysis that included randomized trials of patients who were receiving an aromatase inhibitor as either their first or second hormonal therapy for metastatic disease, those who were randomly assigned to a third-generation drug (e.g., anastrozole, letrozole, exemestane, or vorozole) lived longer (HR for death 0.87; 95% CI, 0.82–0.93) than those who received standard therapy (tamoxifen or a progestational agent).[27][Level of evidence: 1iA] Two randomized trials that enrolled 400 and 451 patients who had progressed after receiving tamoxifen demonstrated that fulvestrant yielded similar results to anastrozole in terms of its impact on PFS.[44,45] The proper sequence of these therapies is currently not known.[43,46] Premenopausal women should undergo oophorectomy (surgically, with external-beam radiation therapy or with an LHRH agonist).[47] Patients with lymphangitic pulmonary metastases, major liver involvement, and/or central nervous system involvement should not receive hormone therapy as a single
179
modality. Patients with structural compromise of weight-bearing bones should be considered for surgical intervention and/or radiation in addition to systemic therapy. Patients with vertebral body involvement should be evaluated for impending cord compression even in the absence of neurologic symptoms. Increasing bone pain and increasing alkaline phosphatase within the first several weeks of hormone therapy does not necessarily imply disease progression.[48] Patients with extensive bony disease are at risk for the development of symptomatic hypercalcemia early in the course of hormone therapy.[48] Early failure (e.g.,
