Myeloablative chemotherapy with autologous peripheral blood stem ...

Bone Marrow Transplantation, (1999) 24, 837–843  1999 Stockton Press All rights reserved 0268–3369/99 $15.00 http://www.stockton-press.co.uk/bmt

Myeloablative chemotherapy with autologous peripheral blood stem cell transplantation for metastatic breast cancer: immunologic consequences affecting clinical outcome NG Chakraborty, S Bilgrami, LJ Maness, C Guo, A Perez-Diez, B Mukherji and P Tutschka Department of Medicine, University of Connecticut School of Medicine, Farmington, CT, USA

Summary: Autologous peripheral blood stem cell transplantation following myeloablative chemotherapy is being increasingly utilized in the treatment of a variety of malignancies. We administered busulfan 16 mg/kg orally, thiotepa 500– 700 mg/m2 i.v., and carboplatin 800–1000 mg/m2 i.v. to 56 women with metastatic carcinoma of the breast. Autologous peripheral blood stem cells, which had been collected after a combination of chemotherapy and granulocyte colony-stimulating factor, were infused on day 0. The major toxicities of the conditioning regimen included severe pancytopenia, stomatitis, nausea, emesis, diarrhea, fever, and infection. Transplant-related mortality was 1.8%. The incidence of opportunistic viral infections was 42.9%. Fourteen individuals achieved a complete response. The actuarial survival at 1223 days was 13.7% for the entire group of patients; the actuarial survival at 1009 days was 39.3% among complete responders. The functional status of the immune system was determined following transplantation in a subset of patients. Peripheral blood mononuclear cells were obtained before and after stem cell infusion, and were analyzed phenotypically and functionally. Proliferative and interleukin-2 synthetic ability of these cells was assessed following stimulation with phytohemagglutinin and anti-CD3 antibody. The response to influenza peptides was also ascertained. Proliferative and interleukin-2 synthetic capacity was markedly impaired for over a year. Memory response was virtually absent for up to 2 years following transplantation. The prolonged and marked immunosuppression following this myeloablative regimen was associated with a high incidence of opportunistic viral infections, and may have contributed to disease relapse and progression especially in patients who failed to achieve a complete response following transplantation. Keywords: myeloablative chemotherapy; breast cancer; stem cell transplant; immunosuppression

Autologous peripheral blood stem cell transplantation (PBSCT) has emerged as a credible therapeutic modality in Correspondence: Dr S Bilgrami, MC-1315, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, CT 06030, USA Received 4 January 1999; accepted 26 May 1999

the treatment of recurrent and high-risk malignancies.1–4 Although this form of treatment has been most successful in hematologic malignancies, some therapeutic gains have also been achieved in solid tumors, including carcinoma of the breast.5–7 Myeloablative doses of chemotherapy usually result in a prolonged period of immunologic compromise. In prior studies, it has been demonstrated that although profound and prolonged immunosuppression is a feature of all types of bone marrow transplantation (BMT), the depth and duration of immunosuppression is greatest following allogeneic BMT, intermediate after syngeneic BMT, and least following autologous BMT.8,9 It also appears that the use of autologous PBSCT rather than autologous BMT leads to faster hematopoietic and immunologic reconstitution.10 The implications of these findings are important and would favor a lower incidence of opportunistic infections and, conceivably, relapse of the underlying malignancy following autologous PBSCT.10 We utilized a novel, highly myeloablative conditioning chemotherapy regimen followed by autologous PBSCT in 56 women with metastatic carcinoma of the breast. Response rates and survival status were determined as was the incidence of opportunistic viral infections following transplantation. The degree of immunosuppression and the kinetics of immunologic reconstitution after transplantation were studied in an attempt to correlate immune recovery with clinical outcome parameters.

Materials and methods Patients Fifty-six women with metastatic breast carcinoma underwent autologous PBSCT between June, 1994 and April 1996 at the University of Connecticut Health Center. The initial phase of treatment consisted of two (20 patients) or three (36 patients) courses of dose-intense chemotherapy (etoposide 60 mg/kg intravenously; cyclophosphamide 3 g/m2 intravenously; and paclitaxel 200 mg/m2 intravenously) administered for the dual purpose of stem cell procurement and tumor cytoreduction. The goal was to obtain 5 × 106 CD34-positive cells/kg body weight following the second such course of chemotherapy. One month after the final course of this chemotherapy, a myeloablative conditioning regimen consisting of busulfan 4 mg/kg/day (day −10 to day −7) orally, thiotepa 125–175 mg/m2/day

Immunosuppression after autologous stem cell transplant NG Chakraborty et al

838

(day −6 to day −3) intravenously, and carboplatin 200– 250 mg/m2/day (day −6 to day −3) intravenously was administered followed by autologous stem cell infusion (day 0). Toxicity was evaluated in accordance with the World Health Organization criteria.11 All individuals who were sero-positive for herpes simplex virus (HSV) received acyclovir, 400 mg orally twice daily, from the initiation of conditioning chemotherapy until 50 days following stem cell infusion. Disease status was reassessed 60 days following stem cell infusion by computerized axial tomography of the chest, abdomen, and pelvis as well as by total body bone scan. An immunocytochemical staining assay (BIS Laboratories, Reseda, CA, USA) was utilized to detect micrometastases in the bone marrow and in the peripheral blood stem cell (PBSC) product obtained after the first and second cycles of stem cell mobilizing chemotherapy. Only PBSC products with a negative immunocytochemical assay were utilized for stem cell rescue. Complete response (CR) was defined as the absence of clinical and radiologic evidence of disease. Partial response (PR) and non-response (NR) were described, respectively, as greater than or less than 50% reduction in the size of measurable disease. Progression was indicated by worsening disease. Restaging studies were repeated every 3 to 6 months. The incidence of opportunistic viral infections within the first year following autologous PBSCT was ascertained by reviewing the charts of all patients. A diagnosis of varicella-zoster virus (VZV) infection was confirmed by direct fluorescent antibody testing (Merifluor VZV; Meridian Diagnostics, Cincinnati, OH, USA). Cytomegalovirus (CMV) infection or excretion was detected by the identification of immediate early and early antigens in various body fluids (Chemicon International, Temecula, CA, USA). In addition, VZV, CMV, HSV and adenovirus infections were confirmed by culture. Ten patients agreed to participate in the immunologic study. A total of 5 to 7 ml of heparinized peripheral blood was collected after obtaining informed consent from these 10 patients prior to HDC, and subsequently, at various time intervals up to 2 years after stem cell infusion. Peripheral blood was also obtained from six normal individuals, and from five patients with melanoma and five patients with breast cancer treated with other therapeutic modalities. Culture medium Iscove’s modified Dulbecco’s medium (IMDM; Life Technologies, Grand Island, NY, USA) containing 5 mm each of l-asparagine, l-glutamine and l-arginine supplemented with 10% fetal bovine serum (Life Technologies, Grand Island, NY, USA) and 10 mg/ml of gentamicin (ElkinsSinn, Cherry Hill, NJ, USA) was utilized for all in vitro experiments. Henceforth, this culture medium is referred to as complete medium (CM). Isolation of peripheral blood mononuclear cells (PBMC) Peripheral blood mononuclear cells (PBMC) were isolated by gradient centrifugation using Ficoll–Hypaque (Sigma Chemicals, St Louis, MO, USA). The total number of

viable cells was ascertained. Freshly isolated cells were used for phenotypic and functional analyses. Monoclonal and polyclonal antibodies Monoclonal antibodies against CD1a, CD3, CD4, CD8, CD34, MO1, B1, and polyclonal antibodies against mouse IgG and goat anti-mouse IgG-FITC (fluorescein isothiocyanate) were purchased from Coulter-Immunotech (Miami, FL, USA). Antibodies against T cell receptor (TCR) were also purchased from Coulter-Immunotech. Phenotypic analysis Mononuclear cells were isolated and washed three times with phosphate-buffered saline (PBS). A total of 2–4 × 105 cells were labeled with FITC-conjugated monoclonal antibodies and with control antibody at 4°C for 30 min. Subsequently, the cells were washed twice with cold PBS and resuspended in 0.5 ml of fixative buffer. Fluorescence-positive cells were counted in a fluorescence-activated cell sorter (FACS; Becton Dickinson, Franklin Lakes, NJ, USA) using Cell Quest software. Proliferation of PBMC Ficoll–Hypaque-isolated PBMC were washed four times with PBS and resuspended in CM. These cells (2 × 105 per well) were cultured in triplicate in 96-well U-bottomed plates (Costar, Cambridge, MA, USA) with phytohemagglutinin (PHA; Sigma Chemical Company) 2 ␮g/ml or with anti-CD3 antibody (Coulter) 40 ␮g/well for 5 days in the presence or absence of 50 U/ml of interleukin-2 (IL-2; Gibco-BRL, Gaithersburg, MD, USA). Radioactive thymidine 3HTdR (Nen, Boston, MA, USA) 1 ␮Ci/well was added to the wells for the last 16 h of culture. Cells were harvested in a Titertek semi-automatic cell harvester (Skatron, Sterling, VA, USA) and thymidine incorporation was measured in a Beckman liquid scintillation counter (Beckman Instruments, Irvine, CA, USA). Interleukin-2 secretion by PBMC after stimulation Peripheral blood mononuclear cells obtained from study patients and normal individuals were treated with PHA or anti-CD3 antibody as described above and supernatants of these cultures were removed after 24, 48, and 72 h of stimulation. The presence of IL-2 in the supernatant was measured using IL-2 ELISA kits (Endogen, Woburn, MA, USA) in accordance with the manufacturer’s directions. Memory response Memory response was analyzed utilizing influenza peptides (CTELKLSDY for HLA-A1 and GILGFVFTL for HLAA2) for patients and control subjects who were HLA-A1 and/or HLA-A2 positive.12,13 The presence of memory was determined by the ability of PBMC to elaborate interferongamma upon presentation of the peptides by autologous antigen presenting cells in vitro. Therefore, peripheral blood mononuclear cells (1 × 105) were stimulated with


respective autologous or allogeneic HLA-A1 or -A2 matched antigen presenting cells (1 × 105) and pulsed with influenza peptides in microtiter plates for 24 h in tissue culture medium. Either monocyte lineage cells which had been derived from PBMC and cultured in granulocyte–macrophage colony-stimulating factor (GM-CSF, 1000 U/ml) as well as interleukin-4 (IL-4, 1000 U/ml) or allogeneic Epstein–Barr virus (EBV)-transformed B cells were used to present the influenza peptides. Antigen presenting cells were pulsed with 1 ␮g/ml of the appropriate peptide/million cells in PBS for 2 h and irradiated with 100 Gy from a CS137 source. Peripheral blood mononuclear cells obtained from patients and normal individuals were then stimulated with the antigen presenting cells in 96-well microculture plates (PBMC: antigen presenting cells = 10:1) in CM. After 24 h of culture, supernatants from all of the different wells were collected in duplicate and the presence of interferon-gamma was quantified using ELISA kits (Immunotech, Miami, FL, USA) in accordance with the manufacturer’s directions. Results Clinical outcome: safety, toxicity and efficacy The clinical characteristics of all 56 patients and their response status following autologous PBSCT are outlined in Table 1. Seventy-one percent of these 56 patients had received radiation therapy several months or years prior to PBSCT. Ten patients agreed to participate in the immunologic study. Their clinical features were similar to those of the entire cohort of patients. The median age of these 10 patients was 48 years and they had received a median of two and nine prior chemotherapy regimens and cycles, respectively. The median interval from diagnosis of breast cancer to PBSCT and diagnosis of metastases to PBSCT was 25 months and 9 months, respectively. A median of Table 1

Clinical characteristics

Number of patients Median age in years (range) Interval from diagnosis of breast cancer till stem cell infusion in months (range) Interval from diagnosis of metastatic disease till stem cell infusion in months (range) Median number of organ systems involved by breast cancer (range) Median number of chemotherapy regimens prior to myelo-ablative chemotherapy (range) Median number of chemotherapy cycles prior to myeloablative chemotherapy (range) Median duration of neutropenia (absolute neutrophil count ⬍1 × 109/l) following stem cell infusion in days (range) Disease status 2 months following stem cell infusion Complete response Partial response No response Progressive disease Actuarial survival at 1223 days (n = 56) Actuarial survival of complete responders at 1009 days (n = 14)

56 45 (25–61) 32 (2–175)

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

Actuarial survival at 1223 days is 13.7% (n = 56)

0

200

600

400

800

1000

1200

1400

Days post transplant Figure 1 Actuarial survival of entire cohort of patients with breast cancer following myeloablative chemotherapy and autologous stem cell infusion. The actuarial follow-up was 1223 days and the total number of study patients was 56.

two organ systems were involved with metastatic disease. Seven individuals had received radiation therapy prior to PBSCT. However, only two patients had been irradiated within 6 months of stem cell infusion. Therefore, it is unlikely that radiation therapy made any significant impact on the immunologic aspects of this study. The number of complete responders at 2 months following transplantation was 14 patients (25%). The overall response rate (complete and partial response) was 53.6%. The actuarial survival of the entire group of patients at 1223 days was 13.7% (Figure 1). However, the actuarial survival of the group of patients in complete remission following transplantation at 1009 days was 39.3% (Figure 2). Major toxicities of the myeloablative regimen included grade 4 granulocytopenia and thrombocytopenia, grade 3 alopecia, grade 1–3 diarrhea, grade 1–3 nausea/vomiting, and grade 3–4 stomatitis. Eighty-eight percent of patients developed a grade 1– 3 cutaneous auto-aggression syndrome which resembled cutaneous graft-versus-host disease. This erythematous, maculo-papular, desquamative eruption affected not only the palms and soles but also the intertriginous regions. It was likely not related to thiotepa toxicity because its median day of onset (day +11 following PBSCT ) occurred nearly 2 weeks after the completion of thiotepa therapy. Furthermore, the rash lasted for a median of 6 weeks and affected individuals required topical and systemic corticosteroids for relief of symptoms. Fifty patients (89.3%)

7 (2–153) 2 (1–6) 3 (1–4) 10 (3–22) 7 (3–11)

14 16 11 15

(25%) (28.6%) (19.6%) (26.8%) 13.7% 39.3%

Actuarial survival at 1009 days is 39.3% ( n = 14)

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0

200

400

600

800

1000

1200

Days post transplant Figure 2 Actuarial survival of patients achieving a complete remission following transplantation at a follow-up of 1009 days. The number of patients was 14.

839


840

Table 2

Opportunistic viral infections

Number of patients Number of patients with any viral infection Number of viral infections Number of patients with VZV infection Median interval from stem cell infusion till VZV infection in days (range) Number of patients with CMV infection/excretion Median interval from stem cell infusion till CMV infection/excretion in days (range) Number of patients with HSV infection Median interval from stem cell infusion till HSV infection in days (range) Number of patients with adenovirus infection Interval from stem cell infusion

24 29 13 73

56 (42.9%) (51.8%) (23.2%) (6–284)

7 (12.5%) 37 (18–81) 4a (7.1%) 32 (15–83) 1 (1.8%) 24

VZV = varicella zoster virus; CMV = cytomegalovirus; HSV = herpes simplex virus. a Four patients had five herpes simplex virus infections.

developed a fever greater than 38°C when neutropenic and bacteremia was documented in 19 patients (33.9%). There was one treatment-related death from veno-occlusive disease of the liver. There were 29 documented instances of viral infection or excretion in 24 individuals (Table 2). Cytomegalovirus disease in the form of gastritis was identified in one individual. Two other patients developed CMV viremia without disease. Additionally, CMV viruria alone was detected in four other patients. Twenty-four (83%) of these 29 viral infections occurred within 100 days of stem cell infusion. Immunological recovery Figure 3 demonstrates the absolute number of PBMC/ml of blood obtained from healthy individuals (n = 6), from cancer patients (n = 10) and from transplant recipients (n = 10) before and after PBSCT. Immediately after transplantation,

a burst of mononuclear cell growth was observed with some patients demonstrating a 10-fold increase in number as compared to control subjects. When these PBMC were characterized phenotypically (Table 3), most of the cells were positive for both CD1a and MO1 markers indicative of precursor cells. Although the absolute number of these CD1a/MO1+ cells declined over the next 2 months, their number still remained relatively elevated until approximately 60 days following transplantation. The number of T lymphocytes including CD3+, CD4+ and CD8+ cells was reduced, with CD4 demonstrating the most profound suppression lasting to day 60 after stem cell infusion. Transplanted patients also exhibited a significant decrease in the number of CD3-positive lymphocytes as well as of T cell receptor ␣, ␤- or WT31-positive cells. Furthermore, the number of cells expressing the Fc receptor (CD16) was significantly reduced when compared to controls (data not shown). Functional analysis of the recovering immune system Table 4 describes the functional properties of PBMC obtained from transplanted patients at different time intervals. The mononuclear cells did not proliferate in response to PHA or anti-CD3 antibody, even in the presence of exogenous IL-2. Only a marginal proliferative response was observed when exogenous IL-2 was added, indicating an inability of such cells to respond to IL-2 even after stimulation. This was in marked contrast to the responses observed in normal individuals and non-transplanted cancer patients, where a significant amount of IL-2 was detected in the samples (Table 5). This severe functional deficit of the immune system continued to be demonstrated at 1 year following PBSCT. Table 3

Phenotypic analysis of PBMC

Source of PBMC

Healthy donor ( n = 6) Melanoma

( n = 5)

Breast cancer ( n = 5) Day -10 Day +10 Day +15 Day +20

n = 10

Day +25 Day +30 Day +35 Day +40 Day +60 0

2

4

6

8

10

12

Number of cell × 106

Figure 3 Number of peripheral blood mononuclear cells per milliliter of blood. The numbers are a mean of values obtained from individuals studied in each category.

Normal volunteers (n = 6) Mean (range) Melanoma and breast cancer patients (n = 8) Mean (range) Transplanted patients (n = 10) Day −10 Mean (range) Day +10 Mean (range) Day +20 Mean (range) Day +30 Mean (range) Day +50 Mean (range) Day +60 Mean (range)

Phenotypes (%) CD1a

CD3

CD4

CD8

MO1

5 (0–10) 10

66 (55–72) 60

45 (40–65) 35

21 (20–30) 25

6 (0–1) 12

(5–16)

(52–70)

(20–45)

(20–35)

(6–20)

35 (28–40) 76 (62–100) 58 (46–72) 60 (48–76) 46 (40–56) 32 (25–40)

36 (30–42) 22 (16–28) 34 (28–48) 22 (15–30) 20 (12–32) 26 (20–40)

2 (0–4) 0 − 4 (0–8) 3 (0–6) 5 (0–8) 8 (0–16)

18 (12–22) 4 (0–10) 12 (4–20) 10 (8–18) 18 (16–26) 28 (20–40)

23 (18–29) 83 (70–100) 62 (50–78) 52 (55–90) 38 (25–40) 26 (20–40)

PBMC = peripheral blood mononuclear cells.


Table 4 Mean proliferation of peripheral blood mononuclear cells with different stimuli (fold increase over baseline control) Days after PBSCT

PHA

PHA + IL-2

Anti-CD3 antibody

−10 +10 +15 +20 +25 +30 +35 +40 +60 +1 year +2 years Non-transplanted cancer patients Normal volunteers

9.6 1.1 1.2 1.1 1.1 1.2 1.2 1.0 2.2 2.1 10.7 7.1 53.5

39.8 1.2 2.1 3.7 2.6 1.8 2.2 2.5 19.6 19.5 14.4 13.1 76.0

14.9 1.5 1.5 2.1 1.8 1.5 2.5 2.0 3.6 4.0 23.4 7.3 36.5

PBSCT = peripheral blood stem cell transplantation; PHA = phytohemagglutinin; IL-2 = interleukin-2.

Table 5 PBMC

Interleukin-2 secretion (pg/ml/2 × 105 cells) by stimulated

Source of PBMC

Day −10 Day +10 Day +20 Day +30 Day +40 Day +50 Day +365ⴱⴱ Day +730ⴱⴱⴱ Normal volunteers

Control

PHA

Anti-CD3 antibody

0 0 0 0 0 0 0 0 0

20* (0–46) 0 0 0 0 0 30 (0–68) 20 (0–45) 120 (100–200)

25 (0–55) 0 0 0 0 0 20 (0–40) 40 (0–40) 100 (80–150)

*These numbers represent the mean of all values obtained from 10 patients at each time point. Numbers in parenthesis denote range. On days 365 and 730 only four (**) and five (***) patients respectively were tested.

Memory response Memory response was evaluated in a selected number of cases based on HLA-A1 or HLA-A2 phenotype because influenza matrix peptides restricted to these two MHC class 1 molecules are known.12,13 A significant amount of interferon-gamma (100–200 pg/ml/2 × 105 cells) was detected in the culture supernatant when PBMC obtained from normal individuals were stimulated with the appropriate peptides. In contrast, PBMC obtained from transplanted patients failed to elaborate any interferon-gamma in the culture supernatants during the first 60 days following transplantation. Even at 2 years after transplantation, interferongamma production remained significantly impaired (Table 6). Discussion Numerous studies have outlined the sequence of immunologic reconstitution following allogeneic, syngeneic and autologous BMT.8–10 Both quantitative and qualitative

Table 6 Memory response (interferon-␥ production) by PBMC when stimulated by influenza peptide Source of PBMC Normal volunteers (n = 6) Cancer patients (5)b Day −10 (n = 6) Day +10 (n = 6) Day +20 (n = 4) Day +40 (n = 4) Day +60 (n = 5) Day +730 (n = 4)

Mean interferon-␥ levels pg/ml 160 30 0 0 0 0 0a 0a

a One patient in each group showed some memory response when presented with allogeneic EBV-transformed B cells. b Interferon-␥ concentration varied from 20 to 35 pg/mg.

deficits of T lymphocytes, B lymphocytes and, to a lesser extent, natural killer (NK) cells exist for prolonged periods.14 Fewer studies have been performed in the autologous PBSCT setting. The peripheral blood stem cell (PBSC) harvest product contains a normal number (compared to peripheral blood) of T lymphocyte subsets, B lymphocytes, NK cells, and CD25-expressing cells, as well as a normal CD4:CD8 ratio.10,15–17 Harvested T lymphocytes secrete IL2 when stimulated by PHA, and exhibit lymphokine-activated killer (LAK) activity when incubated with IL-2.15,18,19 Hematologic recovery is usually swift and complete following autologous PBSCT. Similarly, immunologic recovery is more rapid following autologous PBSCT compared to autologous and allogeneic BMT.20–23 Natural killer cells (NK cells) (CD3− and CD56+) are among the earliest cells to recover numerically following autologous PBSCT10 and exceed the normal value by 6 weeks before returning gradually to baseline.16 The number of NK cells is significantly greater following autologous PBSCT compared to autologous BMT for up to 5 months after transplantation.15,20,21 The absolute lymphocyte count recovers by 17 days following stem cell infusion22 and CD3+ cells normalize quantitatively by 2–4 weeks.16,20 CD8+ cells recover numerically by 3 weeks whereas CD4+ cells remain subnormal for 60 days or longer10,16,20,22 leading to a reversal of the CD4:CD8 ratio for 4 or more months.16,18,20,22 Cytotoxic T cells (CD8+, CD11b−) recover rapidly whereas suppressor T lymphocytes (CD8+, CD11b+) remain subnormal for a prolonged period.16 Similarly, helper-inducer cells (CD4+, CD45RO+) and memory T lymphocytes (CD4+, CD29+) recover rapidly while suppressor-inducer cells (CD4+, CD45RA+) remain subnormal for a prolonged period.16,20,21 Recovery of activated CD3+ cells, CD25+ cells and T cell receptor gamma/delta-bearing lymphocytes is also faster following autologous PBSCT compared to autologous BMT.16,21 Re-establishment of immunologic function is also prompt following autologous PBSCT. Response of T lymphocytes to stimulation by IL-2 and the ability of these cells to secrete IL-2 in response to stimulation by PHA recovers within 30 days after autologous PBSCT.15,18,21 Our findings are in marked contrast to the data reported in the literature. Although there was an initial burst of precursor cells, comprised mainly of CD1A+, MO1+, CD3− cells, the regeneration of T lymphocytes was markedly

841


842

delayed with fewer than 30% of circulating cells being CD3+ by day +60 and less than 10% of circulating cells being CD4+ at that time point. Even more dramatic was the virtual absence of T cell function with a severely impaired ability of T lymphocytes to proliferate or synthesize IL-2 in response to lectin or receptor-mediated stimulation lasting for over a year, and an undetectable memory response to influenza peptides for up to 2 years following autologous stem cell infusion. It is likely that the use of busulfan, a bi-functional alkylator that not only results in complete ablation of the bone marrow but also has a profound effect on all bone marrow-derived cells such as thymic dendritic cells,24,25 contributed significantly to the prolonged immunosuppression seen in our patients. One could speculate that complete ablation of the bone marrow requires a complete rebuilding of the immune system from the stem cell, not unlike the situation observed after allogeneic transplantation where the kinetics of immune recovery are similar to those seen in our patient population.10 Conversely, destruction of the dendritic cells may severely impair the thymic microenvironment and with it the thymic function as well, leading to a delay in immune reconstitution until the thymic architecture has been rebuilt.23,26 The number of opportunistic viral infections identified following autologous PBSCT with our myeloablative regimen was significant. The majority of these infections occurred within the first 4 months, the period of maximum immunocompromise following autologous PBSCT. Although there was no mortality from opportunistic viral infections in this series, there was considerable morbidity and inconvenience to the patient because of the need for therapy and, occasionally, hospitalization. Certainly, the incidence, if not the severity, of such infections after our myeloablative regimen resembled the scenario after allogeneic BMT. It is likely that our chemotherapy was not only myeloablative but also extremely immunoablative. This would suggest that in the autologous PBSCT setting, the type of chemotherapy regimen may be responsible for the subsequent degree and duration of immunosuppression.10 The high response rate achieved with our myeloablative regimen was gratifying. Unfortunately, few patients remained in remission for prolonged periods of time. Only the subset of patients in complete remission following transplantation were long-term survivors whereas none of the patients with residual disease after autologous PBSCT had prolonged survival. Although resistance of clones of malignant cells to chemotherapy is the most obvious explanation for such relapses and disease progression, it has been postulated that an inadequately functioning immune system may also contribute to early cases of relapse especially in patients who fail to achieve a complete remission following autologous PBSCT.10 In fact, many of the relapses seen in our patients occurred early after transplantation during the period of maximal immunosuppression. Adoptive immunotherapy appears to work most effectively when there is minimal residual disease,14 a status that is achieved in patients with breast cancer in clinical complete remission or near complete remission immediately after autologous PBSCT. However, our study indicates that the immune sys-

tem may be incapable of responding to immunologic manipulation in the immediate post-transplant period. We have demonstrated that a prolonged period of severe immunologic compromise follows autologous PBSCT, similar to autologous and allogeneic BMT, if a highly myeloablative conditioning chemotherapy regimen is utilized. The clinical effects of this marked immunosuppression include opportunistic viral infections and, possibly, early relapse of breast cancer. We believe that the intensity of myeloablative chemotherapy regimens, such as the one utilized in this study, has been maximized and no significant advances will be forthcoming in this direction. Considerable research is ongoing with immune modulators in the post-transplant setting.27 However, such an approach is unlikely to be successful because of the marked immunosuppressed state observed shortly after transplantation. If these patients could be reconstituted immunologically, immunotherapy would have more of a chance to be effective. One approach which is currently being studied at our center, is to infuse aliquots of pre-ablative PBMC (T cells, B cells and NK cells) at frequent and defined time intervals during the first 4 months post transplant as a method of adoptive transfer of mature immune cells. It is not yet clear whether such a strategy will help reconstitute these patients more fully and rapidly and decrease the incidence of opportunistic infections and disease relapse. If this method is able to produce rapid reconstitution of the immune system, these patients may become better candidates for immunotherapeutic approaches.

Acknowledgements This work was supported by a grant from the University of Connecticut Health Center Research Council (6-33071) and in part by National Institute of Health (CA 613980). We are indebted to Mrs Joyce Fritz for her expertise in the preparation of this manuscript.

References 1 Kessinger A, Armitage JO. The evolving role of autologous peripheral stem cell transplantation following high dose therapy for malignancies. Blood 1991; 77: 211–213. 2 Kessinger A, Bierman P, Vose J, Armitage J. High dose cyclophosphamide, carmustine and etoposide followed by autologous peripheral blood stem cell transplantation for patients with relapsed Hodgkin’s disease. Blood 1991; 77: 2322–2325. 3 Elias A, Ayash, L, Anderson K et al. Mobilization of peripheral blood progenitor cells by chemotherapy and granulocyte–macrophage colony-stimulating factor for hematologic support after high dose intensification for breast cancer. Blood 1992; 79: 3036–3044. 4 Teshima T, Harada M, Takamatsu Y et al. Cytotoxic drug and cytotoxic drug/G-CSF mobilization of peripheral blood stem cells and their use for autografting. Bone Marrow Transplant 1992; 10: 215–220. 5 Bezwoda WR, Seymour L, Dansey RD. High-dose chemotherapy with hematopoietic rescue as primary treatment for metastatic breast cancer. A randomized trial. J Clin Oncol 1995; 13: 2483–2489. 6 Antman KH, Rowling PA, Vaughan WP et al. High dose chemotherapy with autologous hematopoietic stem-cell sup-


7

8

9

10

11 12

13 14 15

16

17

port for breast cancer in North America. J Clin Oncol 1997; 15: 1870–1879. Laport GF, Grad G, Grinblatt DL, Williams SF. High-dose chemotherapy consolidation with autologous stem cell rescue in metastatic breast cancer: a 10 year experience. Bone Marrow Transplant 1998; 21: 127–132. Shiobara S, Harada M, Mori T et al. Difference in posttransplant recovery of immune reactivity between allogeneic and autologous bone marrow transplantation. Transplant Proc 1982; 14: 429–433. Witherspoon RP, Kopecky K, Storb R et al. Immunological recovery in 48 patients following syngeneic marrow transplantation for hematological malignancy. Transplantation 1982; 33: 143–149. Roberts MM, To LB, Gillis D et al. Immune reconstitution following peripheral blood stem cell transplantation, autologous bone marrow transplantation and allogeneic bone marrow transplantation. Bone Marrow Transplant 1993; 12: 469–475. Miller AB, Hoogstraten B, Staquet M, Winkler A. Reporting results of cancer treatment. Cancer 1981; 47: 207–214. Bednarek MA, Sauma SY, Gammon MC et al. The minimum peptide epitope from the influenza virus matrix protein: extra and intracellular loading of HLA-A2. J Immunol 1991; 147: 4047–4053. Sidney J, Gray HM, Kubo T, Sette A. Practical, biochemical and evolutionary implications of the discovery of HLA Class I super motifs. Immunol Today 1996; 17: 261–267. Bilgrami S, Silva M, Cardoso A et al. Immunotherapy with autologous bone marrow transplantation: rationale and results. Exp Hematol 1994; 22: 1039–1050. Neubauer MA, Benyunes M, Thomson JA et al. Lymphokineactivated killer (LAK) precursor cell activity is present in infused peripheral blood stem cells and in the blood after autologous peripheral blood stem cell transplantation. Bone Marrow Transplant 1994; 13: 311–316. Ashihara E, Shimazaki C, Yamagata N et al. Reconstitution of lymphocyte subsets after peripheral blood stem cell transplantation: two color flow cytometric analysis. Bone Marrow Transplant 1994; 13: 377–381. Ageitos AG, Ino K, Ozerol I et al. Restoration of T and NK

18

19

20 21

22

23

24 25

26

27

cell function in GM-CSF mobilized stem cell products from breast cancer patients by monocyte depletion. Bone Marrow Transplant 1997; 20: 117–123. Kiesel S, Pezzuto A, Korbling M et al. Autologous peripheral blood stem cell transplantation: analysis of autografted cells and lymphocyte recovery. Transplant Proc 11989; 21: 3084–3088. Verbik DJ, Jackson JD, Pirruccello SJ et al. Functional and phenotypic characterization of human peripheral blood stem cell harvests: a comparative analysis of cells from consecutive collections. Blood 1995; 85: 1964–1970. Rosillo MC, Orturio F, Moraleda JM et al. Immune recovery after autologous or rh-G-CSF primed PBSC transplantation. Eur J Haematol 1996; 56: 301–307. Talmadge JE, Reed E, Ino K et al. Rapid immunologic reconstitution following transplantation with mobilized peripheral blood stem cells as compared to bone marrow. Bone Marrow Transplant 1997; 19: 161–172. Henon PR, Liang H, Beck-Wirth G et al. Comparison of hematopoietic and immune recovery after autologous bone marrow or blood stem cell transplants. Bone Marrow Transplant 1992; 9: 285–291. Witherspoon RP. Immunological reconstruction of the allogeneic marrow, autologous marrow or autologous peripheral blood stem cell transplantation. In: Atkinson K (ed). Clinical Bone Transplantation. Cambridge University Press: Cambridge, 1994, pp 62–72. Dunn CDR. The chemical and biological properties of busulphan (‘myeleran’). Exp Hematol 1974; 2: 101–117. Tutschka PJ. Marrow ablation: the need for space, immunosuppression, and malignant cell eradication. In: Atkinson K (ed). Clinical Bone Marrow Transplantation. Cambridge University Press: Cambridge, 1994, pp 13–18. Glazier A, Tutschka PJ, Farmer E. Studies on the immunobiology of syngeneic and autologous graft versus host disease in cyclosporine-treated rats. Transplant Proc 1983; 15: (Suppl. 1): 3035–3041. Guillaume T, Rubinstein DB, Syman M. Immune reconstitution and immunotherapy after autologous hematopoietic stem cell transplantation. Blood 1998; 92: 1471–1490.

843