Prevalence,Risk Factors,and Outcomes of Bronchiolitis Obliterans After Allogeneic Hematopoietic Stem Cell Transplantation

Introduction

Allogeneic hematopoietic stem cell transplantation (HSCT) has been established as a therapeutic approach for malignant and nonmalignant hematological disorders. The use of reduced intensity conditioning regimens, advances in graft-versus-host disease (GVHD) management, and improved anti-infectious therapy/prophylaxis has lead to increasing survival rates, and therefore, late transplant-related complications have become more frequent. Pulmonary complications develop in 30%–60% of allogeneic HSCT recipients and are considered as a major cause of morbidity and mortality. Bronchiolitis obliterans (BO) is now the primary noninfectious pulmonary complication of allogeneic hematopoietic stem cell transplantation and lung transplantation. ¹ The NIH consensus criteria considered BO as the only pulmonary manifestation of chronic GVHD ² (cGVHD) and the reported prevalence of BO in allogeneic HSCT recipients ranges between 2% and 10%.^3,4 Despite the improvement in BO management, mortality remains high (between 13% and 23%) and most patients die for respiratory failure or subsequent infections. ⁵

Materials and Methods

In this study, we retrospectively analyzed the incidence, risk factors, and outcome of BO in a single-institution cohort of pediatric patients who underwent allogeneic HSCT. Between January 2000 and November 2012, 89 transplanted pediatric patients were retrospectively evaluated. The institutional review board (IRB) of our hospital approved the study. HSCT recipients were in mandatory continuous ambulatory aftercare at our center, during which GVHD characteristics were documented. The diagnosis of BO was made according to the NIH consensus criteria for cGVHD. ² BO was classified as mild (mild symptoms such as shortness of breath after climbing one flight of steps and/or forced expiratory volume in 1 s (FEV1) 60%–79%), moderate (moderate symptoms with shortness of breath after walking on flat ground and/or FEV1 40%–59%), and severe (severe symptoms with shortness of breath at rest, requiring O₂ and/or FEV1 < 40%). ² High-resolution computed tomography and exclusion of infectious etiological factors by bronchoalveolar lavages or induced sputum analyses were included in the diagnostic routine. Pulmonary function tests (PFTs) were regularly performed every 3–6 months after transplantation and additionally in symptomatic patients with nonresolving symptoms such as exertional dyspnea and dry cough.

Differences in frequencies of discrete variables were tested using 2-sided Fisher's exact test. Univariate and multivariate analysis was conducted on the following risk factors: age at HSCT, sex, high risk, type of donor and type of diagnosis, conditioning regimen, total body irradiation, busulfan-based regimen, acute and cGVHD, donor sources, ABO blood group incompatibility, Rh incompatibility, human leukocyte antigen (HLA), cytomegalovirus (CMV) reactivation. The median follow-up time was of 60 months (range 4–153). Log-rank test was used to analyze the overall survival of the BO-positive versus BO-negative patients. All P-values were 2 sided, and differences were considered statistically significant when a P-value was <0.05.

Results

Clinical characteristics, demographic features, and transplantation-related data of patients are summarized in Table 1. Seven patients met the diagnostic criteria for BO. The prevalence of BO was 7.8%. Four patients (58%) presented with moderate forms of BO, 1 patient (14%) was identified as having mild form, and 2 patients (28%) were categorized with severe BO. Pretransplant PFTs, all BO patients did not present any abnormality: mean FEV1 was 90% (±9.1), mean FEV1/forced vital capacity (FVC) ratio was 91% (±4.1), mean FEF 25%–75% value was 90% (±3.7), and mean residual volume (RV) value was 109% (±18.1). Pretransplant PFT results in patients with and without BO were not significantly different as shown in Table 2. The changes between the pretransplant PFT results and PFT results at the time of BO diagnosis are summarized in Table 3. PFT variables such as FEV1, FEV1/FVC ratio, and FEF 25%–75% decreased significantly (P < 0.001). RV, which is a marker of air trapping, increased significantly (P < 0.001). In particular at the time of BO diagnosis, the mean FEV1 value was 46% (±17.7), mean value FEF 25%–75% was 23% (±16.1), mean FEV1/FVC ratio was 66% (±16.3), and mean RV value was 180% (±25.6). Clinical risk factors associated with BO were the presence of chronic hepatic and/or cutaneous GVHD (P < 0.0001). All patients received treatment consisting of steroids (patients start treatment with prednisone usually at 1 mg/kg/day in a single oral dose. If a patient is responding or showing signs of a stable disease, a prednisone taper starts by reducing the dose by 25% per week over 4 weeks, to a target dosage of 1 mg/kg every other day), inhaled beta agonists, increase in the dosage of systemic immunosuppressive treatment, repeated courses of oral antibiotics [azithromycin (AZI)] and supportive anti-infectious therapy/prophylaxis. One patient had to be hospitalized in pediatric intensive care, where the patient had necessitated oxygen therapy, oxygen therapy with high flow, and noninvasive ventilation. Six patients showed improvement in BO and 1 patient (14%) died, after developing BO, of pulmonary insufficiency during hospitalization in pediatric intensive care. In particular, this patient presented a decreased pretransplant FEV1/FVC ratio value, associated to an FEV1 value <30% of that predicted at the time of diagnosis. The 5-year survival at the median follow-up period after diagnosis of BO (range 4–153 months) was not significantly different between the 2 groups (60%, SE 5.7% versus 80%, SE 17 = 0.38).

Discussion

In this cohort of pediatric patients who underwent allogeneic HSCT, we found a BO prevalence of 7.8%, which was consistent with previous studies.^3–5 cGVHD, both in hepatic and cutaneous form, was demonstrated in the present study to be frequently associated with BO as described in various studies on BO before. ⁵ The higher incidence of BO in children with hepatic and cutaneous GVHD could be potentially interpreted on the basis that cGVHD is an immunoregulatory disorder, which involves almost any organ of the body; therefore, BO could be described as the pulmonary manifestation of a systemic disease. ⁶ Blazar et al. explained the pathophysiology of cGVHD underlining the major role of 6 hallmarks distinctive of this syndrome. cGVHD appears as an autoimmune-like systemic syndrome frequently linked with fibroproliferative changes in almost any organ as oral and ocular mucosal surfaces the skin, lung, kidneys, liver and gut. In particular, the mechanisms unique for this condition are as follows: the damage to the thymus, associated with the conditioning regimen and with the occurrence of a GVHD earlier in the post-HSCT course, which determines a dysregulated negative selection of alloreactive CD4⁺ T cells; the deviation of the Th2 cytokine pattern causing a release of fibrogenic cytokines such as interleukin (IL)-2, IL-10, and transforming growth factor (TGF)-β; the tissue fibroblast proliferations resulting from the macrophage activation through the release of TGF-β and platelet-derived growth factor (PDGF); and the lower Treg levels and the dysregulation of B cells leading to emergence of autoreactive B cells and production of autoreactive antibodies. On the basis of these speculations, BO could be interpreted as pulmonary manifestation of a systemic alloreactive immune response. ⁶ Although intensified immunosuppression has been used to treat BO, this approach has had negligible impact on disease course and can substantially increase the risk of serious toxicity, especially the risk of serious and fatal opportunistic infections. While initial, intense corticosteroid therapy can significantly benefit hematopoietic stem cell transplantation recipients with acute onset of GVHD or restrictive, late-onset, noninfectious pulmonary complications (organizing pneumonia, late interstitial pneumonia), such therapy does not appear to provide sustained benefit for BO associated with cGVHD. ⁶ AZI, a neomacrolide antibiotic with both antibacterial and immunomodulatory properties, has been used to treat BO in hematopoietic stem cell transplantation recipients, and a growing body of clinical research shows that a substantial number of patients may respond to AZI or AZI-based combination therapy. The new perspectives of the pathogenetic mechanisms of BO could allow future therapeutic approaches such as tyrosine kinase inhibitors.^7,8 BO remains difficult to diagnose and once clinical symptoms develop, airway obstruction and small airway disease often are significant and irreversible. Trimonthly follow-up PFTs for a follow-up period of 12 months in patients who underwent allogeneic HSCT helped in the early diagnosis of BO. ⁹ PFTs presented an important role in the management of BO, not only in supporting an early diagnosis but also in suggesting the possible evolution of the condition and survival. The only patient who died in our group, which had necessitated respiratory instrumental treatment, presented specific PFT features, a decreased pretransplant FEV1/FVC ratio value, and an FEV1 value <30% of that predicted at the time of diagnosis. As reported in literature, significant risk factors associated with mortality were the decreased pretransplant FEV1/FVC and an FEV1 value <30% of that predicted at the time of diagnosis, ¹⁰ which could be considered as “warning signs” for possible early treatment/hospitalization in the pediatric intensive care. Early diagnosis and intensive treatment may improve the prognosis of patients with BO resulting in a global overall survival not statistically different from the other transplanted patients.