Biopsy-Verified Bronchiolitis Obliterans and Other Noninfectious Lung ...

3 downloads 0 Views 932KB Size Report
Lung biopsy is the gold standard for diagnosis. This study ... planted between 2000 and 2010 who underwent lung biopsy for suspected BO. Of those, 23 were ...
Biol Blood Marrow Transplant 21 (2015) 531e538

Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org

Biopsy-Verified Bronchiolitis Obliterans and Other Noninfectious Lung Pathologies after Allogeneic Hematopoietic Stem Cell Transplantation Hilde Hylland Uhlving 1, 2, 3, *, Claus B. Andersen 4, Ib Jarle Christensen 5, Magdalena Gormsen 6, Karen Damgaard Pedersen 6, Frederik Buchvald 1, 7, Carsten Heilmann 1, Kim Gjerum Nielsen 1, 7, Jann Mortensen 8, Claus Moser 9, Henrik Sengeløv 10, Klaus Gottlob Müller 1, 2 1

Department of Pediatrics and Adolescent Medicine, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Institute for Inflammation Research, The Finsen Center, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 3 Research Unit Women’s and Children’s Health, The Juliane Marie Center, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 4 Department of Pathology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 5 The Finsen Laboratory & Biotech Research and Innovation Center (BRIC), Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 6 Department of Radiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 7 Pediatric Pulmonary Service, Department of Pediatrics and Adolescent Medicine, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 8 Department of Clinical Physiology, Nuclear Medicine and PET, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 9 Department of Microbiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 10 Department of Hematology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 2

Article history: Received 25 August 2014 Accepted 1 December 2014 Key Words: Pulmonary pathology Pulmonary function tests Bronchiolitis obliterans Bronchiolitis obliterans syndrome Chronic graft-versus-host disease Hematopoietic stem cell transplantation

a b s t r a c t Bronchiolitis obliterans (BO) is a serious complication of allogeneic hematopoietic stem cell transplantation (HSCT). Lung biopsy is the gold standard for diagnosis. This study describes the course of BO and assesses the congruity between biopsy-verified BO and a modified version of the National Institutes of Health’s consensus criteria for BO syndrome (BOS) based exclusively on noninvasive measures. We included 44 patients transplanted between 2000 and 2010 who underwent lung biopsy for suspected BO. Of those, 23 were diagnosed with BO and 21 presented other noninfectious pulmonary pathologies, such as cryptogenic organizing pneumonia, diffuse alveolar damage, interstitial pneumonia, and nonspecific interstitial fibrosis. Compared with patients with other noninfectious pulmonary pathologies, BO patients had significantly lower values of forced expiratory volume in 1 second (FEV1), FEV1/forced vital capacity, and maximal mid-expiratory flow throughout follow-up, but there was no difference in the change in pulmonary function from the time of lung biopsy. The BO diagnosis was not associated with poorer overall survival. Fifty-two percent of patients with biopsy-verified BO and 24% of patients with other noninfectious pulmonary pathology fulfilled the BOS criteria. Pathological BO diagnosis was not superior to BOS criteria in predicting decrease in pulmonary function beyond the time of biopsy. A lung biopsy may provide a characterization of pathological patterns that can extend our knowledge on the pathophysiology of HSCT-related lung diseases. Ó 2015 American Society for Blood and Marrow Transplantation.

INTRODUCTION Bronchiolitis obliterans (BO) is a serious complication of allogeneic hematopoietic stem cell transplantation (HSCT), clinically defined by airflow obstruction and pathologically characterized by progressive fibrosis and obliteration of the bronchioles [1,2]. Lung biopsy is the gold standard for

Financial disclosure: See Acknowledgments on page 537. * Correspondence and reprint requests: Hilde Hylland Uhlving, Department of Pediatrics and Adolescent Medicine 4072, Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark. E-mail address: [email protected] (H.H. Uhlving).

http://dx.doi.org/10.1016/j.bbmt.2014.12.004 1083-8791/Ó 2015 American Society for Blood and Marrow Transplantation.

diagnosis but is invasive and not widely used in the daily clinical routine. Instead, multiple diagnostic criteria based on clinical disease manifestations have been applied, which may partly explain the variability in the reported incidence (1.2% to 14.6%) [3-14] and mortality (27% to 90%) [3-5,7,1013]. To improve the basis for comparative research without running the risks associated with lung biopsy, in 2005 the National Institutes of Health (NIH) published criteria for BO syndrome (BOS), a clinical equivalent of biopsy-verified BO based on pulmonary function (PF) tests and radiology [2]. According to these criteria, BOS is diagnosed in patients

532

H.H. Uhlving et al. / Biol Blood Marrow Transplant 21 (2015) 531e538

where infectious etiology has been excluded and who manifest a forced expiratory volume in 1 second (FEV1) < 75% of predicted normal value, FEV1/forced vital capacity (FVC) ratio < .7, residual volume > 120% of predicted, and high-resolution computed tomography (HRCT) with signs of air trapping, bronchiectasis, or peribronchial thickening [2]. However, in a retrospective validation study, only 18% of patients (4/22) with clinically identified or biopsy-proven BO fulfilled the criteria [15]. Modified criteria have therefore been proposed to improve applicability [15]. Although biopsy-verified BO is diagnostic for chronic graft-versus-host disease (cGVHD), fulfillment of the BOS criteria is not by itself diagnostic for cGVHD but requires at least 1 distinctive extrapulmonary manifestation for the diagnosis of cGVHD [16,17]. Still, biopsy-based studies are sparse, and our knowledge on the clinical course of BO and the level of congruity with BOS is limited. Likewise, it is not clarified if lung biopsy represents a more sensitive or specific prognostic tool than the BOS criteria. The aims of this study were (1) to describe the course of biopsy-verified BO, (2) to assess the congruity between biopsy-verified diagnosis and a modified version of the NIH BOS criteria, and (3) to compare the prognostic value of these diagnostic methods in the prediction of pulmonary outcome and survival. METHODS Patient Population Patients treated with allogeneic HSCT in Denmark between January 1, 2000 and December 31, 2010 who survived beyond 3 months from transplantation were included if a lung biopsy was performed during follow-up and if this biopsy showed signs of noninfectious pulmonary pathology. Infectious pathology was an exclusion criterion. The study was approved by the Danish National Board of Health (J.nr.3-3013-47/1) and the Danish Data Protection Agency (J.nr.2007-58-0015). Data Collection Clinical data were collected from the Danish reports to the Center for International Blood and Marrow Transplant Research and patient files, until September 2013. The diagnosis of cGVHD was verified by review of patient files to ensure agreement with NIH criteria [2]. Lung Biopsies Only patients having open (n ¼ 34) or transbronchial (n ¼ 10) biopsies were included in the study. All biopsies were reviewed by the same experienced pulmonary pathologist and were classified as having either BO according to NIH’s histopathological criteria [1] or other noninfectious pulmonary pathologies, such as cryptogenic organizing pneumonia (COP, previously known as bronchiolitis obliterans organizing pneumonia), diffuse alveolar damage, interstitial pneumonia (including lymphocytic interstitial pneumonia), and nonspecific interstitial fibrosis (peribronchial or alveolar septal). Microbiological agents in biopsies, bronchoalveolar lavages, sputum, and blood samples were evaluated by an experienced microbiologist and led to exclusion unless considered as contamination (ie, S. epidermidis) or in case of pathogens not associated with respiratory symptoms (ie, normal flora of the oropharynx). HRCT Chest HRCTs performed before diagnostic biopsies were described independently by 2 senior radiologists who were blinded to the pathological diagnosis. The final interpretation was made by consensus. Based on previous descriptions of HRCTs in HSCT recipients [18], the presence of the following radiological parameters was established: bronchiectasis, bronchial wall thickening, air trapping (or mosaic pattern if no expiratory scan was available), consolidation, ground glass opacities, and centrilobular opacities. PF Tests PF tests followed the recommendations from the European Respiratory Society and American Thoracic Society [19-21] using the ScreenPro and MasterLab (Jaeger, Hoechberg, Germany). Diffusion capacity of the lungs for carbon monoxide (DLCO) and total lung capacity were measured by the

single-breath helium dilution technique. DLCO was corrected for hemoglobin value. The results are presented as percent of predicted value, based on agematched reference materials for spirometry [22], DLCO [23,24], and static lung volumes [21,23]. For longitudinal analysis, PF tests were sorted into time intervals relative to the time of lung biopsy: within 3 months before lung biopsy, 3 to 9 months, 9 to 18 months, and last available measurement beyond 18 months from the lung biopsy. When more than 1 measurement was available per patient per time interval, the median value was chosen. Residual volume, which is needed for a full BOS classification according to the NIH criteria [2], was available in only a few patients. We therefore applied a previously published modified version of the NIH criteria to establish the BOS diagnosis: FEV1 < 75% of predicted normal and FEV1/FVC ratio < .7 and a decrease in FEV1 by 10% compared with pretransplant values [4].

Statistics Associations were tested using the Fisher exact test and Wilcoxon rank sum test. Longitudinal differences and changes in PF between BO and nonBO patients from time of biopsy were estimated in a linear mixed model with repeated measures. Non-normally distributed data were log transformed for analysis. To detect bias due to different biopsy techniques, results were validated in subset analyses that only included patients diagnosed with open lung biopsy. To compare the value of lung biopsy and BOS criteria in predicting longterm decline in PF, a decrease in FEV1 of more than 20% from (pre-HSCT) baseline to 18 months postbiopsy was defined as the outcome parameter. Based on the definition of BOS in lung transplant recipients [25,26], a 20% decrease in FEV1 from baseline was chosen as primary endpoint for a clinically significant decrease in PF. As a second endpoint, we compared the associations between BO and BOS diagnoses and death at 18 months postbiopsy. The 18-month threshold was based on the fact that this time point was reached by all participants at the time of this study. Associations were tested in logistic regression models. For survival analyses, the BO and non-BO patients in our cohort were compared with the background population of Danish HSCT recipients transplanted in Denmark between 2000 and 2010 who survived beyond 3 months from HSCT (n ¼ 698). Survival rates were calculated from time of HSCT and time of lung biopsy. Overall survival probabilities were plotted in Kaplan-Meier curves. Estimates were compared with Cox proportional hazard models. Differences in relapse rates and treatment-related mortality were analyzed using competing risks methodology. Statistical analyses were performed in SPSS Statistics (version 19, IBM, Armonk, NY) and SAS (version 9.2; SAS Institute, Cary NC). P < .05 were considered to be significant.

RESULTS Of the 742 patients who underwent HSCT in Denmark between 2000 and 2010 and survived beyond 3 months, 59 patients had an open or a transbronchial lung biopsy performed on suspicion of BO (Figure 1). Forty-four patients were diagnosed with noninfectious pulmonary pathology and were included in this study. Of these, 23 patients had BO and 21 patients presented noninfectious lung pathology other than BO, including COP (n ¼ 6), diffuse alveolar damage (n ¼ 3), unspecific interstitial fibrosis (n ¼ 11), and interstitial pneumonia (n ¼ 1). Characteristics of the study population are presented in Table 1. In 5 patients, BO was diagnosed in a second biopsy. In these patients, the first biopsy had shown bronchopneumonia (n ¼ 1), COP (n ¼ 1), diffuse alveolar damage (n ¼ 1), or unspecific inflammation (n ¼ 2). Ninety-one percent of BO cases (21/23) and 62% of nonBO cases (13/21) (P ¼ .031) were diagnosed by open lung biopsy. The remaining cases were diagnosed with transbronchial biopsies. The most common biopsy-related morbidity included pain and a transient increase in the requirement for O2 support. Minor pneumothorax was seen on x-ray after both open (8/34, 23.5%) and transbronchial biopsies (2/10, 20.0%). In 1 patient who received a transbronchial biopsy, pneumothorax required transthoracic drainage. In 3 patients who received open lung biopsies (3/ 34, 8.8%) the pneumothorax was combined with the development of subcutaneous emphysema. Infections related to

H.H. Uhlving et al. / Biol Blood Marrow Transplant 21 (2015) 531e538

533

Figure 1. Study population. TBB indicates transbronchial biopsy.

the incision site were seen in 6 patients (6/34, 17.6%), and 1 patient developed a minor bleeding (1/34, 2.9%) after open lung biopsy. Three patients died during the hospital stay at which the biopsy was performed. Causes of death were bacterial pneumonia, interstitial pneumonia, and BO, respectively.

HRCT and Clinical Characteristics at Time of Lung Biopsy The decision to perform a lung biopsy was based on clinical symptoms (cough, dyspnea, increased oxygen need), reduced PF, and/or pathological findings on HRCT (especially air trapping, bronchiectasis, and peribronchial thickening), combined with an individualized risk assessment of this invasive procedure. The decision was made by transplant physicians in collaboration with thoracic surgeons but without specific radiological or clinical cut-off-points to guide the decision. Cough and dyspnea were present in most patients, both BO and non-BO, before lung biopsy. A larger proportion of the non-BO patients were in need of O2 supplement (Table 2). PF tests performed a median of 7 days (range, 1 to 90) before the lung biopsy showed a significantly larger decrease in the FEV1/FVC ratio compared with pre-HSCT values in patients who were later diagnosed with BO compared with patients with other pulmonary pathologies (Table 3, Figure 2). HRCT was performed in 77.2% of patients (34/44), of whom 82.3% (28/34) presented findings that led to a suspicion of BO (air trapping/mosaic pattern, bronchiectasis, peribronchial thickening), whereas infectious pathology (consolidation, ground glass opacities) was suspected in 35.3% of patients (12/34). We found no significant differences in the presentation of air trapping, peribronchial thickening,

and bronchiectasis in HRCTs performed in patients with BO compared with non-BO pathologies (Table 2). The accordance between pathological BO diagnosis and the modified BOS criteria is described in Table 4. The sensitivity of the modified BOS criteria for biopsy-verified BO was 52.4% and the specificity 76.5%. The positive predictive value was .73 and the negative predictive value .57. When only patients diagnosed by open lung biopsy (as opposed to transbronchial biopsy) were included, the sensitivity and specificity were 57.9% and 63.6%, respectively. Treatment In most patients, prednisolone in combination with other immunosuppressive drugs was used as first-line therapy after biopsy-based diagnosis (Table 5). There was a tendency toward a higher intensity of treatment in the BO group, because a larger proportion of these patients received a combination of 3 or more immunosuppressive drugs once the diagnosis had been established (9/23 versus 1/21, P ¼ .048). Long-Term Lung Function Follow-up The last available PF tests were performed at a median of 47 months (range, 18 to 112) for BO patients and of 44 months (range, 18 to 122) for non-BO patients after the lung biopsy. From time of biopsy and throughout follow-up, BO patients exhibited lower values of FEV1, FEV1/FVC, maximal mid-expiratory flow (MMEF), FVC, and DLCO compared with non-BO patients. However, no significant differences between the 2 groups in the change in PF parameters during follow-up were found (Figure 2). The results were validated in subanalyses only including patients diagnosed by open lung biopsy (34/44). In this analysis we also found

534

H.H. Uhlving et al. / Biol Blood Marrow Transplant 21 (2015) 531e538

Table 1 Characteristics of the Study Population (N ¼ 44)

Median recipient age, yr (range) Diagnosis malignant AML ALL CML Lymphoma Other malignancies Nonmalignant hematological diseases Other diseases HLA-identical sibling donor Matched unrelated donor (10/10 or 9/10 match) Unrelated donor (>1 allele mismatch) Median donor age, yr (range) Sex mismatch Male to female Female to male CMV mismatch Positive donor to negative recipient Negative donor to positive recipient Stem cell source Bone marrow Peripheral blood Umbilical cord Median cell dose/kg (range) Chemotherapy BU TBI CY Etoposide FLU Conditioning regimen BUþCY TBIþCY TBIþetoposide TBIþFLU Other Acute GVHD occur Grade I Grade II Grade III Grade IV

Table 2 Clinical Characteristics at Time of Lung Biopsy for BO and Non-BO Patients

BO (n ¼ 23)

Non-BO (n ¼ 21)

40.7 (.6-61.8)

49.7 (.7-59.4)

19 4 4 3 3 5 3

(82.6) (17.4) (17.4) (13.0) (13.0) (21.7) (13.0)

18 3 6 2 d 7 2

(85.7) (14.3) (28.6) (9.5)

P Value of Difference .43 1.0

(33.3) (9.5)

1 (4.3) 11 (47.8)

1 (4.7) 10 (47.6)

7 (30.4)

9 (42.9)

5 (21.7)

2 (9.5)

38.3 (15.4-57.8) 37.3 (20.0-69.0)

.47

.80

4 (17.4) 6 (26.1)

2 (9.5) 12 (57.1)

.67 .065

3 (13.0)

5 (23.8)

.45

6 (26.1)

3 (14.3)

.46

8 (34.8) 15 (65.2) d 7.8 (2.4-27.8)

7 (33.3) 14 (66.7) d 5.9 (.3-12.7)

1.0 1.0 .11

9 14 13 3 7

(39.1) (60.9) (56.5) (13.0) (30.4)

9 4 3 7

(39.1) (17.4) (13.0) (30.4)

d 13 4 8 1 d

(56.5) (17.4) (34.8) (4.3)

7 13 13 3 5 7 5 3 5 1 9 3 5 1

(33.3) (61.9) (61.9) (14.3) (23.8) (33.3) (26.3) (14.3) (23.8) (4.8) (42.9) (14.3) (23.8) (4.8)

.76 .75 .77 1.0 .74 .76 .71 1.0 .74 d .55

d

AML indicates acute myeloid leukemia; ALL, acute lymphoblastic leukemia; CML, chronic myelogenous leukemia; CMV, cytomegalovirus; BU, busulfan; TBI, total body irradiation; CY, cyclophosphamide; FLU, fludarabine. Values are number of case with percents in parentheses, unless otherwise indicated.

significantly lower values of FEV1 (P ¼ .0020), FEV1/FVC (P < .0001), MMEF (P ¼ .0003), and DLCO (P ¼ .012) in the group of BO patients, but there was no difference between BO and non-BO patients in the change in PF during follow-up from time of biopsy. In a bivariate logistic regression model, we compared the association between a 20% decrease in FEV1 from pre-HSCT to 18 months postbiopsy and the diagnoses of biopsy-based BO and clinically defined BOS, respectively. The odds ratio (95% confidence interval) for the association with a 20% decrease in FEV1 was 3.1 (.6 to 15.5, P ¼ .16) for patients with biopsy-verified BO and 3.1 (.6 to 16.5, P ¼ .18) for patients with BOS. Neither BO nor BOS were significantly associated with death at 18 months postbiopsy (odds ratio, 1.6 [95% CI,

BO Months from HSCT to lung biopsy, median (range) Respiratory symptoms, n (%) Cough Dyspnea Fever Hypoxia Lung function values, mean (SD) FEV1 % predicted FVC % predicted FEV1/FVC % predicted MMEF % predicted TLC % predicted DLCO % predicted HRCT findings, n (%)* Air trapping or mosaic pattern Bronchiectasis Peribronchial thickening Consolidation Ground glass opacities Centrilobular opacities cGVHD in other organ system, n (%)y Skin Mouth Eyes Other Immunosuppressive therapy, n (%) Cyclosporine Tacrolimus Prednisolone Mycophenolate mofetil Rapamune

Non-BO

13.7 (3.3-44.6)

10.0 (2.8-83.4)

22 20 3 3

(95.7) (86.9) (13.0) (13.0)

20 21 5 10

46.7 57.4 82.4 33.7 70.8 52.3

(14.0) (14.6) (20.1) (25.9) (11.8) (14.1)

63.8 63.4 101.2 105.2 71.3 45.1

P Value of Difference .41

(95.2) (100.0) (23.8) (47.6)

1.0 .23 .43 .020

(28.8) (25.9) (18.0) (60.4) (19.5) (23.2)

.054 .61 .002 .001 .85 .16

11 (55.0)

11 (78.6)

17 (85.0) 17 (85.0)

12 (85.7) 12 (85.7)

13 13 14 13

(65.0) (65.0) (70.0) (56.5)

11 11 5 7

(78.6) (78.6) (35.7) (33.3)

3 13 7 1 19

(13.0) (56.5) (30.4) (4.3) (82.6)

2 6 4 0 13

(9.5) (28.6) (19.0)

11 1 12 10 3

(47.8) (4.3) (52.2) (43.5) (13.0)

5 3 6 4

(23.8) (14.3) (28.6) (19.0)

(61.9)

.28 1.0 1.0 .47 .47 .080 .14

.18

d

TLC indicates total lung capacity. * Percentage values calculated based on the number of patients with available scans. y Only changes classified as diagnostic or distinctive of cGVHD according to the NIH consensus criteria [2] are included.

.3 to 9.2], P ¼ .58), and odds ratio, 1.6 (95% CI, .2 to 10.1], P ¼ .64, respectively). Survival BO patients and non-BO patients had similar overall survival rates (Figure 3). Likewise, both groups had similar survival rates as the background population of HSCT recipients. There was no difference in treatment-related mortality: cumulated incidence at 3 years from HSCT was .13 (95% CI, .03 to .30) for BO patients, .095 (95% CI, .015 to .27) for non-BO patients, and .14 (95% CI, .13 to .17) for the background population (P ¼ .77). The cumulated incidence of relapse mortality for BO patients was .048 (95% CI, .003 to .202), for non-BO patients .26 (95% CI, .06 to .41), and for the background population .25 (95% CI, .22 to .29) (P ¼ .18). Survival analysis with the date of lung biopsy as the starting point did not show any difference in overall survival between BO and non-BO patients (Figure 3). DISCUSSION This is to date the largest study of BO in a cohort of HSCT patients defined clinically by pulmonary manifestations leading to lung biopsy. Compared with HSCT patients with

H.H. Uhlving et al. / Biol Blood Marrow Transplant 21 (2015) 531e538

535

Table 3 Decrease in PF from Pre-HSCT Baseline to Time of Biopsy in BO and Non-BO Patients BO

FEV1 % predicted FEV1/FVC % predicted MMEF % predicted FVC % predicted TLC % predicted DLCO % predicted

Non-BO

Mean Decrease (SD) from Baseline to Time of Biopsy

95% CI of Difference

Mean Decrease (SD) from Baseline to Time of Biopsy

95% CI of Difference

39.7 17.2 50.8 29.0 23.6 21.2

31.3-48.1 9.0-25.4 30.7-70.8 21.3-36.7 16.9-30.4 10.2-32.2

33.9 2.7 25.3 29.5 25.5 30.3

21.8-46.0 5.4-10.8 4.4-54.9 17.7-41.2 10.8-40.2 13.6-47.0

(18.5) (18.0) (34.7) (16.9) (12.7) (17.3)

(23.5) (15.7) (38.6) (22.8) (19.1) (28.9)

P Value of Difference Between BO and Non-BO Patients

.40 .013 .050 .94 .77 .35

Figure 2. Development of lung function in patients with biopsy-verified BO compared with patients with other noninfectious pulmonary pathologies (non-BO). *P value relates to the difference between BO and non-BO patients in PF parameters from time of lung biopsy and throughout follow-up. **P value relates to the difference between BO and non-BO patients in change in PF parameters from time of lung biopsy.

536

H.H. Uhlving et al. / Biol Blood Marrow Transplant 21 (2015) 531e538

Table 4 Fulfillment of Modified NIH Criteria for BOS at Time of Biopsy BO FEV1 < 75% of predicted value FEV1/FVC ratio < .7 >10% decrease in FEV1 from baseline Modified NIH criteria fulfilled

21 11 19 11

(100) (52.4) (90.5) (52.4)

Non-BO

P

13 4 14 4

.032 .10 .64 .10

(76.5) (23.5) (82.4) (23.5)

Values in parentheses are percentages calculated based on number of patients with available measurements.

other noninfectious pulmonary pathologies, patients with BO exhibited significantly reduced PF parameters from time of lung biopsy and throughout follow-up. The BO diagnosis was not associated with reduced survival. The NIH BOS criteria were originally established to provide a noninvasive tool for BO diagnosis. The sensitivity and specificity of the BOS criteria for pathological BO in our study were relatively low, because only half the patients with BO and 24% of the patients with other noninfectious pulmonary pathologies met a modified version of the BOS criteria. This corresponds to the report from Holbro et al. [27] where 7 of 14 patients with biopsy-verified BO and 3 of 10 with other pathology (lymphocytic bronchiolitis) fulfilled the BOS criteria. The HRCT patterns of air trapping, bronchiectasis, and peribronchial thickening, included in the original NIH BOS criteria, were present in up to 86% of the patients in our cohort but had low specificity for the biopsy-verified BO diagnosis. Accordingly, bronchiectasis and bronchial wall thickening have been reported in restrictive [28] and in obstructive [18,29] HSCT-related lung diseases but appear not to be specific to BO or BOS. Air trapping, which has been a reported almost universal finding in HSCT patients with airflow obstruction [18,29,30], may have been underestimated in our cohort, because only half the patients had available expiratory images. Lung biopsy has the potential advantage of discriminating between infectious and noninfectious pulmonary pathologies that may require different treatment strategies. However, in this cohort of patients with noninfectious pathologies, we did not find the pathological BO diagnosis to

Table 5 First-Line Therapy for Pulmonary Complications after Lung Biopsy Drugs Used

BO

Non-BO*

P

High-dose methylprednisolone Oral prednisolone Cyclosporine Mycophenolate mofetil Tacrolimus Rapamune Infliximab Rituximab Initial drug combinations Prednisolone alone Cyclosporine alone Mycophenolate mofetil alone PredþCy PredþMPM Predþother PredþCyþMPM PredþCyþother PredþMPMþother

8 19 13 12 2 3 1 1

(34.8) (82.6) (56.5) (52.2) (8.7) (13.0) (4.3) (4.3)

8 (38.1) 17 (81.0) 8 (38.1) 3 (14.3) 2 (9.5) 0 d d

1.0 1.0 .25 .011 1.0 .23 d d

2 2 1 4 2 3 5 2 2

(8.7) (8.7) (4.3) (17.4) (8.7) (13.0) (21.7) (8.7) (8.7)

7 d d 6 2 3 1 d d

.064 d d .48 1.0 1.0 .19 d d

(33.3)

(28.6) (9.5) (14.3) (4.8)

Pred indicates prednisolone; Cy, cyclosporine; MPM, mycophenolate mofetil. Values in parentheses are percentages of the total number of patients. * Two non-BO patients did not receive any immunosuppressive therapy after lung biopsy.

Figure 3. (A) Overall survival in time from HSCT in patients with biopsyverified BO compared with patients with other noninfectious pulmonary pathology (non-BO) and HSCT patients without lung biopsy from the same period. 1P value of difference between the 3 groups. (B) Overall survival in time from lung biopsy.

be a better predictor of long-term pulmonary impairment than the BOS criteria that were based on PF tests alone. The biopsy discriminated BO patients from non-BO patients, who exhibited significantly higher values of FEV1, FEV1/FVC, and MMEF from time of lung biopsy and during follow-up. However, the change in PF beyond the time of biopsy did not differ significantly between BO and non-BO patients. In contrast, Holbro et al. [27] found that although BO patients exhibited no significant increment in FEV1, patients with other noninfectious pulmonary pathology (lymphocytic bronchiolitis) exhibited a significant and sustained improvement from 6 months postbiopsy. Based on these retrospective studies, we cannot reach a final conclusion on the role of lung biopsy in the diagnosis of post-HSCT pulmonary complications, but our study underlines the value of spirometry as a valid predictive parameter of long-term pulmonary morbidity.

Survival Three previous studies have described the course of biopsy-verified BO in HSCT patients. In 2 minor studies of HSCTs conducted between 1976 and 1985, patients with BO had severely reduced pulmonary capacity, and 70%-100% of these patients died [31,32]. Holbro et al. [27] described HSCT patients transplanted between 1989 and 2010 and found that 50% (7/14) of the BO patients died. Overall survival was significantly reduced in patients with BO compared with patients with other pulmonary pathology (lymphocytic bronchiolitis) [27]. In our cohort, BO diagnosis was not associated with reduced survival compared with non-BO patients or with HSCT recipients not having a lung biopsy. Although the time of follow-up after lung biopsy was similar in our study and Holbro’s study [27], the criteria for performing a lung biopsy might differ between the centers. Although the number of cases in the current study is too small to draw definite conclusions, changes in transplant

H.H. Uhlving et al. / Biol Blood Marrow Transplant 21 (2015) 531e538

protocols and improved therapy over the past decades may explain the better survival of the BO patients observed in our cohort.

Pathogenesis The pathogenesis of BO is not fully understood, and it is conceivable that some of the other noninfectious pulmonary pathologies may represent early stages of BO. Based on pathological studies, Yousem [33] hypothesized that lymphocytic bronchiolitis, diffuse alveolar damage, COP, and BO represented different stages of GVHD-related lung disease. In support of this hypothesis, 4 of our patients with BO had previous biopsies showing diffuse alveolar damage, COP, and unspecific inflammation. Accordingly, several case studies of patients with autopsy-verified BO found COP, lymphocytic bronchiolitis, and interstitial pneumonia in lung biopsies performed when the patients were still alive [34-36]. In a study of serial lung biopsies in lung transplant recipients, COP and interstitial pneumonia in the early post-transplant period were associated with the later occurrence of BO [37]. Because of the pathological and clinical similarities [38], it is conceivable that similar relations exist between pathological patterns encountered in HSCT recipients as well. Whether it is possible to inhibit the pathological processes leading to BO by increased immunosuppressive therapy at an earlier stage is a topic for further research, where lung biopsies, when performed in a protocolled manner, may improve our insights into the pathophysiology of these complications. A major strength of this study is the availability of detailed clinical data in all patients. The patients were transplanted over a 10-year period during which the main elements of the protocol and supportive therapy were largely unchanged. Still, the study has weaknesses. The performance of lung biopsy may have been avoided in the most severely ill patients, resulting in bias in the study population. Because a higher proportion of BO compared with non-BO patients were diagnosed by open lung biopsy, which is assumed to have higher sensitivity than transbronchial biopsy [39], some cases of BO may have been misclassified. The validity of our results, however, was confirmed in subset analyses, where only patients with open lung biopsy were included. The HRCT scanning protocols and image quality have changed during the study period, which make systematic evaluation and comparison difficult. Our evaluation may not have had a sufficient level of accuracy to distinguish differences between BO and non-BO, but imaging per se may not be sensitive enough to detect these differences anyway. Prospective studies with uniform scanning protocols and validated scoring systems are warranted to evaluate the applicability of HRCT in BO and BOS diagnosis. Furthermore, new techniques to improve sensibility and the quantification of small airways disease on HRCT, such as spirometric volume control [40] and quantitative biomarkers such as parametric response mapping [41], might prove valuable in the diagnosis of BO. The BOS criteria applied in the current study were from a previously published, modified version of the NIH criteria [4] with the advantage only to rely on spirometry, which is available at most centers. However, several modifications of the NIH BOS criteria have been applied in different studies [4,5,7,8,15,17,18,42-47], which makes comparison difficult and indicates a need for a more simple and widely accepted classification.

537

In conclusion, patients with biopsy-verified BO had significantly lower values of PF than non-BO patients from time of biopsy and throughout follow-up, but the change in PF did not differ significantly between the groups. Although the BOS criteria had low sensitivity and specificity for pathological BO, the pathological BO diagnosis was not superior to the BOS criteria in the prediction of long-term decrease in PF. Accordingly, the BOS criteria, or the assessment of PF as such, may provide important noninvasive alternatives to lung biopsy for the prediction of long-term pulmonary morbidity. However, the BOS criteria do not allow distinction between infectious and noninfectious pathologies, which could influence the treatment strategy of individual patients, and do not provide information about the etiology of the pulmonary disease. Future, prospective studies should focus on strategies for early detection of noninfectious pulmonary complications to investigate whether the pathological process leading to BO can be reversed by increased immunosuppressive therapy at an earlier stage. ACKNOWLEDGMENTS Financial disclosure: This study was supported by grants from the Danish Children’s Cancer Association, the Gangsted Foundation, the Danish Lung Association, and the Capital Region of Denmark. Conflict of interest statement: There are no conflicts of interest to report. REFERENCES 1. Shulman HM, Kleiner D, Lee SJ, et al. Histopathologic diagnosis of chronic graft-versus-host disease: National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-versus-Host Disease. II. Pathology Working Group Report. Biol Blood Marrow Transplant. 2006;12:31-47. 2. Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-Versus-Host Disease. I. Diagnosis and Staging Working Group report. Biol Blood Marrow Transplant. 2005;11:945-956. 3. Forslow U, Mattsson J, Gustafsson T, Remberger M. Donor lymphocyte infusion may reduce the incidence of bronchiolitis obliterans after allogeneic stem cell transplantation. Biol Blood Marrow Transplant. 2011;17:1214-1221. 4. Au BK, Au MA, Chien JW. Bronchiolitis obliterans syndrome epidemiology after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2011;17:1072-1078. 5. Bergeron A, Godet C, Chevret S, et al. Bronchiolitis obliterans syndrome after allogeneic hematopoietic SCT: phenotypes and prognosis. Bone Marrow Transplant. 2013;48:819-824. 6. Duque-Afonso J, Ihorst G, Wasch R, et al. Identification of risk factors for bronchiolitis obliterans syndrome after reduced toxicity conditioning before hematopoietic cell transplantation. Bone Marrow Transplant. 2013;48:1098-1103. 7. Ditschkowski M, Elmaagacli AH, Koldehoff M, et al. Bronchiolitis obliterans after allogeneic hematopoietic SCT: further insight-new perspectives? Bone Marrow Transplant. 2013;48:1224-1229. 8. Gazourian L, Coronata AM, Rogers AJ, et al. Airway dilation in bronchiolitis obliterans after allogeneic hematopoietic stem cell transplantation. Respir Med. 2013;107:276-283. 9. Santo Tomas LH, Loberiza FR Jr, Klein JP, et al. Risk factors for bronchiolitis obliterans in allogeneic hematopoietic stem-cell transplantation for leukemia. Chest. 2005;128:153-161. 10. Nakaseko C, Ozawa S, Sakaida E, et al. Incidence, risk factors and outcomes of bronchiolitis obliterans after allogeneic stem cell transplantation. Int J Hematol. 2011;93:375-382. 11. Clark JG, Crawford SW, Madtes DK, Sullivan KM. Obstructive lung disease after allogeneic marrow transplantation. Clinical presentation and course. Ann Intern Med. 1989;111:368-376. 12. Dudek AZ, Mahaseth H, DeFor TE, Weisdorf DJ. Bronchiolitis obliterans in chronic graft-versus-host disease: analysis of risk factors and treatment outcomes. Biol Blood Marrow Transplant. 2003;9:657-666. 13. Yoshihara S, Tateishi U, Ando T, et al. Lower incidence of Bronchiolitis obliterans in allogeneic hematopoietic stem cell transplantation with reduced-intensity conditioning compared with myeloablative conditioning. Bone Marrow Transplant. 2005;35:1195-1200.

538

H.H. Uhlving et al. / Biol Blood Marrow Transplant 21 (2015) 531e538

14. Schultz KR, Green GJ, Wensley D, et al. Obstructive lung disease in children after allogeneic bone marrow transplantation. Blood. 1994;84: 3212-3220. 15. Williams KM, Chien JW, Gladwin MT, Pavletic SZ. Bronchiolitis obliterans after allogeneic hematopoietic stem cell transplantation. JAMA. 2009;302:306-314. 16. Uhlving HH, Buchvald F, Heilmann CJ, et al. Bronchiolitis obliterans after allo-SCT: clinical criteria and treatment options. Bone Marrow Transplant. 2012;47:1020-1029. 17. Hildebrandt GC, Fazekas T, Lawitschka A, et al. Diagnosis and treatment of pulmonary chronic GVHD: report from the consensus conference on clinical practice in chronic GVHD. Bone Marrow Transplant. 2011;46: 1283-1295. 18. Gunn ML, Godwin JD, Kanne JP, et al. High-resolution CT findings of bronchiolitis obliterans syndrome after hematopoietic stem cell transplantation. J Thorac Imaging. 2008;23:244-250. 19. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J. 2005;26:319-338. 20. MacIntyre N, Crapo RO, Viegi G, et al. Standardisation of the singlebreath determination of carbon monoxide uptake in the lung. Eur Respir J. 2005;26:720-735. 21. Wanger J, Clausen JL, Coates A, et al. Standardisation of the measurement of lung volumes. Eur Respir J. 2005;26:511-522. 22. Stanojevic S, Wade A, Stocks J, et al. Reference ranges for spirometry across all ages: a new approach. Am J Respir Crit Care Med. 2008;177: 253-260. 23. Koopman M, Zanen P, Kruitwagen CL, et al. Reference values for paediatric pulmonary function testing: the Utrecht dataset. Respir Med. 2011;105:15-23. 24. Cotes JE, Chinn DJ, Quanjer PH, et al. [Standardization of the measurement of transfer factor (diffusing capacity). Work Group on Standardization of Respiratory Function Tests. European Community for Coal and Steel. Official position of the European Respiratory Society]. Rev Mal Respir. 1994;11(Suppl. 3):41-52. 25. Estenne M, Maurer JR, Boehler A, et al. Bronchiolitis obliterans syndrome 2001: an update of the diagnostic criteria. J Heart Lung Transplant. 2002;21:297-310. 26. Verleden GM, Raghu G, Meyer KC, et al. A new classification system for chronic lung allograft dysfunction. J Heart Lung Transplant. 2014;33: 127-133. 27. Holbro A, Lehmann T, Girsberger S, et al. Lung histology predicts outcome of bronchiolitis obliterans syndrome after hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2013;19: 973-980. 28. Song I, Yi CA, Han J, et al. CT findings of late-onset noninfectious pulmonary complications in patients with pathologically proven graftversus-host disease after allogeneic stem cell transplant. AJR Am J Roentgenol. 2012;199:581-587. 29. Sargent MA, Cairns RA, Murdoch MJ, et al. Obstructive lung disease in children after allogeneic bone marrow transplantation: evaluation with high-resolution CT. AJR Am J Roentgenol. 1995;164:693-696. 30. Jung JI, Jung WS, Hahn ST, et al. Bronchiolitis obliterans after allogenic bone marrow transplantation: HRCT findings. Korean J Radiol. 2004;5: 107-113. 31. Chan CK, Hyland RH, Hutcheon MA, et al. Small-airways disease in recipients of allogeneic bone marrow transplants. An analysis of 11 cases and a review of the literature. Medicine (Baltimore). 1987;66:327-340.

32. Holland HK, Wingard JR, Beschorner WE, et al. Bronchiolitis obliterans in bone marrow transplantation and its relationship to chronic graft-vhost disease and low serum IgG. Blood. 1988;72:621-627. 33. Yousem SA. The histological spectrum of pulmonary graft-versus-host disease in bone marrow transplant recipients. Hum Pathol. 1995;26: 668-675. 34. Yokoi T, Hirabayashi N, Ito M, et al. Broncho-bronchiolitis obliterans as a complication of bone marrow transplantation: a clinicopathological study of eight autopsy cases. Nagoya BMT Group. Virchows Arch. 1997; 431:275-282. 35. Trisolini R, Bandini G, Stanzani M, et al. Morphologic changes leading to bronchiolitis obliterans in a patient with delayed non-infectious lung disease after allogeneic bone marrow transplantation. Bone Marrow Transplant. 2001;28:1167-1170. 36. Nakamoto-Matsubara R, Nishikii H, Yamada K, et al. Early pathologic findings of bronchiolitis obliterans after allogeneic hematopoietic stem cell transplantation: a proposal from a case. Case Rep Hematol. 2012; 2012:957612. 37. Burton CM, Iversen M, Carlsen J, Andersen CB. Interstitial inflammatory lesions of the pulmonary allograft: a retrospective analysis of 2697 transbronchial biopsies. Transplantation. 2008;86:811-819. 38. Philit F, Wiesendanger T, Archimbaud E, et al. Post-transplant obstructive lung disease (“bronchiolitis obliterans”): a clinical comparative study of bone marrow and lung transplant patients. Eur Respir J. 1995;8:551-558. 39. Chan A, Allen R. Bronchiolitis obliterans: an update. Curr Opin Pulm Med. 2004;10:133-141. 40. Kongstad T, Buchvald FF, Green K, et al. Improved air trapping evaluation in chest computed tomography in children with cystic fibrosis using real-time spirometric monitoring and biofeedback. J Cyst Fibros. 2013;12:559-566. 41. Galban CJ, Boes JL, Bule M, et al. Parametric response mapping as an indicator of bronchiolitis obliterans syndrome after hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2014;20: 1592-1598. 42. Kuzmina Z, Eder S, Bohm A, et al. Significantly worse survival of patients with NIH-defined chronic graft-versus-host disease and thrombocytopenia or progressive onset type: results of a prospective study. Leukemia. 2012;26:746-756. 43. Vogl UM, Nagayama K, Bojic M, et al. Lung transplantation for bronchiolitis obliterans after allogeneic hematopoietic stem cell transplantation: a single-center experience. Transplantation. 2013;95: 623-628. 44. Kuzmina Z, Krenn K, Petkov V, et al. CD19(þ)CD21(low) B cells and patients at risk for NIH-defined chronic graft-versus-host disease with bronchiolitis obliterans syndrome. Blood. 2013;121:1886-1895. 45. Nakasone H, Kanda J, Yano S, et al. A case-control study of bronchiolitis obliterans syndrome following allogeneic hematopoietic stem cell transplantation. Transplant Int. 2013;26:631-639. 46. Yanagawa N, Sakai F, Kamata N, et al. High-resolution computed tomography features of pulmonary chronic graft-versus-host disease following hematopoietic stem cell transplantation: histopathological correlation. Jpn J Radiol. 2011;29:116-128. 47. Norman BC, Jacobsohn DA, Williams KM, et al. Fluticasone, azithromycin and montelukast therapy in reducing corticosteroid exposure in bronchiolitis obliterans syndrome after allogeneic hematopoietic SCT: a case series of eight patients. Bone Marrow Transplant. 2011;46:1369-1373.