Abstract
The non-infectious pulmonary complications idiopathic pneumonia syndrome (IPS) and bronchiolitis obliterans syndrome (BOS) are important causes of morbidity and mortality after allogeneic hematopoietic cell transplantation (allo-HCT). To identify risk factors for these complications, we retrospectively analyzed baseline characteristics and longitudinal data from 633 pediatric and young adult allo-HCT recipients. Risk factors for IPS included non-malignant diagnosis (HR 2.97; 95% CI 1.27–6.97; P = 0.01) and adenovirus reactivation (HR 2.75; 95% CI 1.24–6.14; P = 0.01). Conditioning with busulfan-cyclophosphamide ±melphalan (BuCy ±Mel) associated with increased IPS risk compared to busulfan-fludarabine ±clofarabine (HR 5.15; 95% CI 2.36–11.24; P < 0.001) and non-myeloablative regimens (HR 9.78; 95% CI 2.48–38.61; P = 0.001). BuCy ±Mel conditioning also related to increased BOS risk relative to total body irradiation (TBI)(HR 5.11; 95%CI 1.65–15.83; P = 0.005) and non-myeloablative regimens (HR 19.68; 95% CI 2.53–155.76; P = 0.005). Moreover, elevated endothelial activation and stress index (EASIX), white blood cell, and lymphocyte counts before day 100 post-transplant correlated with IPS. For BOS, high EASIX and increased activated or effector memory CD4 + T-cells were additional significant correlates. In conclusion, IPS and BOS were associated with both distinct and overlapping risk factors in this study. These findings may inform future strategies to identify high-risk patients and develop targeted preventive interventions.
Subject terms: Risk factors, Translational research
Highlights
The clinical entities IPS and BOS are associated with both distinct and overlapping risk factors
Being alert for risk factors of IPS and BOS might result in earlier diagnosis and treatment, decreasing the risk of irreversible lung damage
Introduction
Non-infectious pulmonary complications after allogeneic hematopoietic cell transplantation (allo-HCT) are important drivers of transplantation-related morbidity and mortality. Idiopathic pneumonia syndrome (IPS), an early onset pulmonary complication, and bronchiolitis obliterans syndrome (BOS), a late onset pulmonary complication, are both commonly defined as alloimmune lung syndromes (allo-LS) with an incidence in pediatric patients of 2–16% [1–7] and 2–11% [3, 5, 6, 8–10], respectively. Allo-LS is thought to be mediated and implicated by alloreactive donor T cells against lung tissue due to endothelial cell activation injury [1, 5, 11], leading to widespread alveolar injury in the case of IPS and a strongly affected small airway epithelium in the case of BOS [12]. Allo-LS is fatal in 50–80% of the cases [3–5, 7, 10] and timely management or prevention of these clinically significant complications is therefore crucial to improve allo-HCT outcome.
Current treatment options for allo-LS include systemic corticosteroids and supportive care, including mechanical ventilation and/or supplemental oxygen [1, 13]. Yanik et al showed that TNF-alpha inhibition therapy with etanercept is beneficial in IPS [1]. Additional treatment of BOS is often a choice of 2-3 immunosuppressive agents such as calcineurin inhibitors (cyclosporine, tacrolimus), prednisone or ruxolitinib, and a combination of fluticasone inhalation therapy, azithromycine and montelukast (FAM). Antifibrotic agents such as tyrosine kinase inhibitors are also being used more recently [13, 14]. Nevertheless, both IPS and BOS remain difficult to treat and have a high mortality rate [3–5, 7, 10]. More insights in risk factors for allo-LS might point towards new prophylactic or early intervention measures.
Most studies on allo-LS to date are limited by small sample sizes, are performed in adults, and include both autologous and allogeneic HCT. As a result, the understanding of etiology, pathology and treatment options for allo-LS in children remains limited. Studies in children show that microbiome attenuation and enrichment of genes associated with NK/T cell activation or respiratory virus (RV) infection prior to and early after allo-HCT are associated with the development of (fatal) allo-LS [5, 6, 15]. In addition, damage to the respiratory endothelium and epithelium is believed to propagate inflammation in lung tissue, increasing the risk of allo-LS development [5, 15]. In preclinical mouse models, defibrotide was able to reduce the incidence of IPS associated with a reduction of endothelial activation [16]. Luft et al have identified the endothelial activation and stress index (EASIX) biomarker panel to be a good surrogate for endothelial injury. This score has been shown to predict mortality after acute graft-versus-host disease (aGvHD) and after allo-HCT [17–19] but has not yet been investigated as a biomarker for allo-LS. Moreover, the occurrence of acute and chronic GvHD are correlated with allo-LS [3, 4, 7, 10], although it has been suggested that the immunosuppressive therapies used for aGvHD might protect against the development of allo-LS [5, 6]. Studies that discriminated between BOS or IPS after allo-HCT in children are based on small numbers of patients [1, 7, 10, 20–22] and studies in larger cohorts are needed to identify potential biomarkers for the development of both BOS and IPS.
Here, we aim to identify risk factors for the development of either BOS or IPS in pediatric and young adults following their first allo-HCT. We conduct a retrospective analysis to investigate the association of different patient characteristics, such as diagnosis and transplant-preparative regimen, and longitudinal markers, including EASIX and immune cell subsets, with BOS and IPS occurrence.
Patients and methods
Patients and procedures
All patients receiving their first allo-HCT in the Wilhelmina Children’s Hospital or the Princess Máxima Center for Pediatric Oncology, the Netherlands, between 2004 and 2024, were included. Patients were enrolled and data were prospectively collected after written informed consent in accordance with the Helsinki Declaration. The use of clinical and laboratory data was approved by the local ethical committee (biobank protocol numbers 05/143 and 11/063-k).
Patients were treated according to (inter)national protocols as previously described [23–26]. After 2019, patients with ALL received total body irradiation (TBI)-based conditioning regimens due to the results of the FORUM study [27]. Viral reactivations and viral respiratory disease were determined and treated as previously described [24, 28]. Additional information is given in the supplemental methods.
Diagnosis and treatment of allo-LS
IPS was diagnosed according to the American Thoracic Society and defined as evidence of widespread lung injury by clinical symptoms and radiological abnormalities in the absence of a direct infectious cause, cardiac dysfunction, fluid overload and renal failure [29]. BOS was diagnosed according to the National Institutes of Health Consensus Criteria on Chronic GvHD disease 2014 and characterized by forced expiratory volume in 1 second (FEV1)/Forced Vital Capacity <0.7, FEV1 < 75% and/or evidence of air trapping on pulmonary function tests (PFT) or high-resolution computed tomography (HRCT, in the absence of a direct infectious cause [30]. Information on treatment protocols is given in the supplemental methods.
Immune kinetics
Absolute numbers and percentages of neutrophils, monocytes, basophils, eosinophils, and lymphocytes were measured at least 3 times per week during the first month, at least every other week up to 12 weeks, and monthly thereafter up to 6 months, followed by every 3 months until 1 year, and twice in the second year after allo-HCT. After a lymphocyte count of 400 cells/µL was reached, additional measurements were performed for CD19 + B-cells, CD16 + CD56+ natural killer (NK)-cells, CD3 + T-cells, CD4 + T-cells, CD8 + T-cells, and CD69+ activated-, CD45RO+ effector memory (EM)-, CD45RA+ naïve T-cell subsets. In addition, IgA, IgG and IgM were measured with Siemens Atellica Technology according to the manufacturer’s instructions.
EASIX scores
EASIX scores were calculated as follows: lactate dehydrogenase (LDH; U/L)*Creatinine (mg/dL)/thrombocytes (109/L) [19], which were routinely measured in clinical care according to local protocols.
Statistical analysis
Associations between patient characteristics and allo-LS were analyzed with Fine and Gray competing risk models, accounting for graft failure and death. Covariables included age, year of allo-HCT, sex, diagnosis, serotherapy, prior grade II–IV aGvHD, cell source, donor type, HLA match, conditioning, ATG prophylaxis (with/without dose adjustment or model-based dosing), viral reactivation ( > 1000 copies/mL of CMV, EBV, AdV, HHV-6) before allo-LS, and respiratory virus (RV) infection before or early after allo-HCT [5, 6]. Variables with P < 0.1 entered multivariable models; forward selection identified independent predictors. Optimal cut-offs for longitudinal parameters were defined by maximally selected rank statistics and tested in Fine and Gray models. Longitudinal markers were assigned to specific days ( ± 7; ±14 for day 150). Analyses were performed using R4.03 with packages ggplot2, survival, survminer, cmprsk, riskRegression, dplyr, forestplot, prodlim, adjustedCurves, pammtools, and lme4.
Results
Patients and outcome
A total of 633 children and young adults receiving an allo-HCT for malignant (56%) and non-malignant (44%) disease were included. The median age was 8.5 (range 0.1 – 22.7) years at time of allo-HCT. Most donors were unrelated (77%). Grafts were derived from BM (48%), CB (49%), and PB (3%). The median follow-up time was 3.27 (range 0.01–20.1) years. Two hundred sixty nine out of 464 patients treated with ATG received dose adjusted ATG or MBD [26]. A total of 78 patients (12%) developed allo-LS, of which 34 were diagnosed with IPS (44%) and 44 with BOS (56%). The median time to IPS and BOS was 62 (interquartile range [IQR] 29–86) and 113 (IQR 90 – 242) days, respectively. RV screening prior to and early after allo-HCT was performed for 389 and 186 patients, respectively. Patient characteristics stratified by allo-LS diagnosis are described in Table 1. The 5-year overall survival (OS) was 72% and the 5-year event-free survival (EFS) was 63% for patients not diagnosed with allo-LS (Fig. S1). For IPS and BOS patients, the 5-year OS were 52% and 66% and the 5-year EFS were 52% and 61%, respectively (Fig. S1).
Table 1.
Patient Characteristics.
| No allo-LS (n = 555) | Allo-LS (n = 78) | ||
|---|---|---|---|
| IPS (n = 34) | BOS (n = 44) | ||
| Female, n (%) | 233 (42%) | 16 (47%) | 19 (43%) |
| Age at allo-HCT, years; median (range) | 8.7 (0.1–22.7) | 4.65 (0.2–18.2) | 9.6 (0.2–19.0) |
| Allo-HCT year, (%) | |||
| 2004–2008 | 86 (15%) | 18 (53%) | 10 (22%) |
| 2009–2013 | 114 (21%) | 8 (23%) | 8 (18%) |
| 2014–2018 | 171 (31%) | 2 (6%) | 13 (30%) |
| 2019–2024 | 184 (33%) | 6 (18%) | 13 (30%) |
| Diagnosis, n (%) | |||
| Malignancy | 315 (57%) | 11 (32%) | 27 (61%) |
| Non-malignant | 240 (43%) | 23 (68%) | 17 (39%) |
| Bone marrow failure | 111 (20%) | 3 (9%) | 3 (7%) |
| Metabolic inborn error | 64 (12%) | 16 (47%) | 7 (16%) |
| Primary immune deficiency | 61 (11%) | 3 (9%) | 7 (16%) |
| Autoimmune disease | 4 (0%) | 1 (3%) | 0 (0%) |
| Cell source, n (%) | |||
| CB | 270 (49%) | 21 (62%) | 21 (48%) |
| BM | 269 (48%) | 12 (35%) | 21 (48%) |
| PB | 16 (3%) | 1 (3%) | 2 (4%) |
| Conditioning regimen, n (%) | |||
| TBI-based | 111 (20%) | 4 (12%) | 4 (9%) |
| BuCy +/- Mel | 56 (10%) | 16 (47%) | 11 (25%) |
| BuFlu +/- Clo | 281 (51%) | 11 (32%) | 28 (64%) |
| Non-myeloablative | 107 (19%) | 3 (9%) | 1 (2%) |
| Serotherapy, n (%) | 417 (75%) | 28 (82%) | 29 (66%) |
| ATG, n (%) | |||
| No ATG | 161 (29%) | 8 (24%) | 15 (34%) |
| ATG with DA or MBD | 245 (44%) | 5 (15%) | 19 (43%) |
| ATG without DA or MBD | 149 (27%) | 21 (62%) | 10 (23%) |
| HLA match, n (%) | |||
| Full | 335 (60%) | 14 (41%) | 25 (57%) |
| Partial | 220 (40%) | 20 (59%) | 19 (43%) |
| 1 HLA mismatch | 164 (30%) | 15 (44%) | 11 (25%) |
| 2 HLA mismatches | 53 (10%) | 5 (15%) | 6 (14%) |
| >2 HLA mismatches | 3 (0%) | 0 (0%) | 2 (4%) |
| Donor type, n (%) | |||
| Related | 126 (23%) | 5 (15%) | 13 (30%) |
| Unrelated | 429 (77%) | 29 (85%) | 31 (70%) |
| aGVHD prior to allo-LS, n (%) | |||
| Grade II-IV | 132 (24%) | 8 (24%) | 8 (18%) |
| Viral reactivation prior to allo-LS, n (%) | |||
| CMV | 71 (13%) | 4 (12%) | 7 (16%) |
| EBV | 23 (4%) | 4 (12%) | 4 (9%) |
| AdV | 45 (8%) | 10 (30%) | 4 (9%) |
| HHV-6 | 96 (17%) | 10 (30%) | 8 (18%) |
| RV infection prior to allo-LS, n (%) | |||
| Prior to allo-HCT (n = 389) | 160 (41%) | 19 (70%) | 16 (62%) |
| Early after allo-HCT (n = 186) | 84 (57%) | 14 (61%) | 10 (67%) |
Allo-LS allo-immune lung syndromes; BO: bronchiolitis obliterans; IPS: idiopathic pneumonia syndrome; allo-HCT allogeneic hematopoietic
cell transplantation; CB cord blood; BM: bone marrow; PB peripheral blood; TBI total body irradiation; BuCy +/- Mel busulfan-cyclophosphamide +/- melphalan; BuFlu +/- Clo busulfan-fludarabine +/- Clofarabine; aGvHD acute graft-versus-host disease; DA dose adjustment; MDB model-based dosing; HLA human leukocyte antigens; CMV cytomegalovirus; EBV Epstein-Barr virus; AdV adenovirus; HHV-6 human herpesvirus 6; RV respiratory virus. RV data prior to and early after allo-HCT was available for 389 and 186 patients respectively.
Patient baseline characteristics and allo-LS development
To study potential risk factors for BOS and IPS, univariable Fine and Gray competing risk models were computed and showed that age, sex, serotherapy, aGvHD grade II-IV, cell source, donor type, and reactivation of CMV or HHV-6 were not risk factors for BOS or IPS (Fig. S2). Year of allo-HCT, non-malignant diagnosis, AdV reactivation, ATG without dose adjustment or MDB, RV infection prior to allo-HCT, and partial matched HLA were associated with higher risk of IPS but not of BOS in univariable analysis (Fig. S2A, B). Conditioning with BuCy +/- Mel was correlated with higher risk of both IPS and BOS, when compared with TBI and non-myeloablative regimens. BuCy +/- Mel was also correlated with higher risk of IPS, but not of BOS, when compared with BuFlu +/- Clo conditioning in univariable analysis (Fig. S2A, B).
Multivariable analysis identified non-malignant diagnosis (HR 2.97; 95% CI 1.27 – 6.97; P = 0.01; Fig. 1a) and AdV reactivation (HR 2.75; 95% CI 1.24 – 6.14; P = 0.01; Fig. 1b) as risk factors for IPS. In addition, conditioning with BuCy +/- Mel was associated with higher risk of IPS compared with BuFlu +/- Clo (HR 5.15; 95% CI 2.36 – 11.24; P < 0.001; Fig. 1c) or with non-myeloablative regimens (HR 9.78; 95% CI 2.48 – 38.61; P = 0.001; Fig. 1c) in multivariable analysis. Conditioning with BuCy +/- Mel was also associated with a higher risk of BOS, when compared with TBI (HR 5.11; 95% CI 1.65–15.83; Fig. 1c) or with non-myeloablative regimens (HR 19.68; 95% CI 2.53–155; P = 0.005; Fig. 1c).
Fig. 1. Non-malignant diagnosis and adenovirus reactivation are risk factors for IPS, while type of conditioning is a risk factor for both BOS and IPS.
Confounder-adjusted cumulative incidence plots of IPS (left) and BOS (right) from the day of allo-HCT stratified by diagnosis type (a), adenovirus reactivation (b), and conditioning regimen (c). Groups were compared using multivariable Fine and Gray Competing Risk models. P values < 0.05 were considered statistically significant.
Association of longitudinal biomarkers with allo-LS development
EASIX scores, immune cell data, and immunoglobulin levels were available for 539 patients, of whom 43 developed BOS and 31 developed IPS (Supplementary Table 1). We evaluated the effect of these parameters as continuous data over time during the first 100 days after allo-HCT on the risk of developing IPS and BOS. From the patients without allo-LS, 30 and 64 patients were, based on the IQR, considered still at risk of developing IPS and BOS, respectively, and were excluded from further analysis. We assessed EASIX (Fig. S3), immune cell reconstitution (Fig. S4) and immunoglobulin levels (Fig. S5) over time within the first 100 days after allo-HCT for patients without allo-LS versus patients that developed IPS or BOS. These longitudinal markers were analyzed separately from patient characteristics because of missing data. We applied maximally selected rank statistics to determine the cut-off points associating with IPS and BOS development, which we then tested in univariable and multivariable Fine and Gray Competing Risk models. Time points considered were day 30, day 50 and day 70 after allo-HCT, with additional analysis on day 100 and day 150 for BOS.
An EASIX score >12 on day 50 and day 70 after allo-HCT was associated with IPS in univariable analysis (Fig. 2a). Absolute numbers of lymphocytes >2619/µL on day 50 and >2323/µL on day 70, as well as white blood cell counts >10761/µL on day 50 and >12377/µL on day 70 were associated with IPS in univariable analysis (Figs. 2b, c, respectively). Multivariable analysis and the association of CD3 + T-cells, CD8 + T-cells and CD8 + EM T-cells could not be assessed due to insufficient data [31].
Fig. 2. High EASIX, and high lymphocyte and white blood cell counts on day 50 and day 70 are associated with IPS.
Cumulative Incidence plots of IPS on day 50 (left) and day 70 (right) stratified by EASIX (a), lymphocyte count (b), and white blood cell count (c). Groups were compared using univariable Fine and Gray Competing Risk models. P values < 0.05 were considered statistically significant.
An EASIX score >3.57 on day 100 (Fig. 3a) and absolute numbers of CD4 + T-cells >220/µL on day 30 (Fig. 3b) and >56/µL on day 70 (Fig. 3c) were associated with BOS in multivariable analysis with conditioning regimen as covariable. CD4 + T-cell counts >343/µL on day 100 were also associated with BOS, however, multivariable analysis could not be performed due to limited number of events and is therefore univariably computed and visualized (Fig. 3d) [31]. Moreover, CD4 + EM T-cell numbers >166/µL on day 30 (Fig. 3e) and activated CD4 + T-cells >14/µL on day 70 (Fig. 3f) were identified to correlate with BOS in multivariable analysis. On day 100, activated CD4 + T-cells >45/µL were also associated with BOS; again, multivariable analysis could not be performed for day 100 (Fig. 3G).
Fig. 3. High EASIX on day 100 and high CD4 + T cell counts on day 30, 70 and 100 are associated with BO.
Confounder-adjusted cumulative incidence plots (a–c, e, f) and cumulative incidence plots (d, g) of BOS stratified by EASIX on day 100 (a), CD4 + T cells on day 30 (b), day 70 (c) and day 100 (d), CD4 + EM T cells on day 30 (e), and activated CD4 + T cells on day 70 (f) and day 100 (g). Groups were compared with either univariable (d, g) or multivariable (a-c, e, f) Fine and Gray Competing Risk models. P values < 0.05 were considered statistically significant.
Discussion
To our knowledge, this is the first study that identified risk factors for BOS and IPS separately in a large cohort of pediatric and young adult patients receiving their first allo-HCT. With the limitations of a retrospective cohort study considered, our data show that non-malignant disease, AdV reactivation, and BuCy +/- Mel conditioning were risk factors for IPS. BuCy +/- Mel conditioning was also a risk factor for BOS. Interestingly, high EASIX scores before day 100 were associated with both IPS and BOS. Overall high absolute numbers of lymphocytes and white blood cells before day 100 were associated with IPS, while specifically high early activated and EM CD4 + T-cell numbers were associated with BOS development. The identification of these risk factors allows timely awareness for these severe complications after allo-HCT, which might be crucial for successful stabilization of the disease before irreversible lung damage can occur.
For both IPS and BOS, BuCy +/- Mel conditioning and high EASIX scores before day 100 were identified as risk factors. The higher toxicity profile of BuCy +/- Mel versus BuFlu was already shown in patients receiving an allo-HCT for malignant and non-malignant disease in our center; the replacement of Cy to Flu in the Bu-based regimen in 2009 was shown to result in less toxicity and equal efficacy, including a significantly lower risk of IPS [25]. Both Bu and Cy are metabolized through conjugation with glutathione (GSH) catalyzed by glutathione S-transferase and a combination of these drugs results in GSH depletion [32, 33], thereby increasing the risk of toxicity [25, 34–36]. The correlation between a higher early EASIX score after allo-HCT with both BOS and IPS supports the hypothesis that endothelial activation and injury is involved in the pathophysiology, as previously described for IPS [11, 12]. EASIX scores can be calculated with a simple formula and might therefore be a useful biomarker for allo-LS. It should be noted, however, that EASIX is not specific to allo-LS since a relation with other endothelial dysfunction syndromes has been found as well [17, 18].
For IPS, we also identified non-malignant disease and AdV reactivation as risk factors, which was not observed in previous studies [5, 7, 20]. This is likely explained by limited number of events and the stratification by various diagnosis types in the earlier studies. Nevertheless, having an inborn error of metabolism as indication for transplantation was previously found to be a borderline risk factor [6]. The number of patients transplanted for metabolic inborn errors was also enriched in the IPS group in our cohort (Table 1), indicating that IPS risk may be higher in this specific diagnostic group. It remains to be studied whether patients with non-malignant disease are at higher risk of IPS due to distinct disease characteristics, lung damage due to the disease, differences in treatment, or other factors that contribute to immune reactions against the lung after allo-HCT. AdV reactivation might result in endothelial dysfunction, possibly triggering immune responses against the lung at the time of immune recovery. We also identified high early lymphocyte and white blood cell counts to be associated with IPS, supporting the hypothesis that immune cells play a role in IPS [1, 5, 11].
For BOS, we identified higher early CD4 + T-cell numbers as an additional risk factor. It should be noted that potential selection bias from the evaluations of the longitudinal biomarkers cannot be excluded in this retrospective cohort analysis. Nevertheless, it would be of interest to know if these CD4 + T-cells are targeting lung tissue already at this timepoint, or if these are increased from an immune activation event such as a viral infection. There is, to the best of our knowledge, only one publication on immune cells in relation to BOS in the pediatric population [21]. However, this small cohort study only compared mean neutrophil values without taking time-to-event data and covariables into account. It is important that future studies include time-to-event data to adequately study or confirm (known) risk factors. Furthermore, the stem cell sources in our cohort were CB and BM, while many other transplant centers use PB and perform GvHD prophylaxis with ptCy, which could especially influence the immune recovery biomarkers identified in this study. Therefore, it is important to evaluate if the identified risk factors also apply to IPS and BOS risk in other pediatric HCT cohorts.
Age, sex, serotherapy, aGvHD grade II-IV, cell source, donor type, and reactivation of CMV or HHV-6 were found not to be risk factors for BOS or IPS, which is in line with previous reports [3, 5–7, 20, 36]. While a correlation between aGvHD and IPS [4, 7] or BOS [10] incidence has been described, we and others did not find aGvHD to be a risk factor [3, 6]. Moreover, RV infection prior to and early after allo-HCT has previously been shown to correlate with allo-LS in our cohort [5, 6], whereas we did not observe an effect in the current study. This is likely explained by the implementation of prophylactic measures for RV infection in our cohort since 2014, which has been shown to reduce TRM [6, 37]. In our cohort, the association with EASIX suggests that endothelial stress contributed to the pulmonary phenotype. Therefore, overlap with subclinical or evolving transplant-associated thrombotic microangiopathy (TMA) cannot be fully excluded. Nonetheless, the clinical presentation and timing of the cases met established criteria for IPS and BOS rather than primary VOD or TMA. In addition, pre-HCT pulmonary function tests were not available in a standardized fashion and could therefore not be incorporated into the risk-factor analysis for IPS and BOS. Therefore, future prospective studies should integrate baseline lung function and lung-compartment biomarkers with longitudinal markers such as EASIX and immune reconstitution to better define how pre-existing lung injury modifies the risk and trajectory of allo-LS.
In conclusion, the clinical entities IPS and BOS were associated with both distinct and overlapping risk factors in this retrospective cohort study in children and young adults receiving their first allo-HCT. The identification of these risk factors for BOS and IPS allows further investigation of potential preventative measures to limit these morbidities, after confirmation in independent allo-HCT cohorts. Changing the BuCy +/- Mel regimen to the BuFlu regimen, postponing allo-HCT and/or tapering immune suppression more slowly for patients with RV infection [6], and optimization of immunosuppressant therapy exposure to prevent adenovirus reactivation after allo-HCT are indicated to be important strategies to limit allo-LS risk. Being alert for these risk factors might result in an earlier diagnosis and treatment, decreasing morbidity and mortality.
Supplementary information
Author contributions
LD was involved in the study design, performed statistical analyses and wrote the manuscript; CAL, SN, CCHdK, CLS, ABV, and ADRH were involved in the study design; ABV, DB, and CAL provided clinical data and reviewed the manuscript; WKL assisted with creating the clinical database and reviewed the manuscript; SH provided routine laboratory data and reviewed the manuscript; CCHdK, CLS, MB, JJB, ADRH, and SN reviewed the manuscript.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Linde Dekker, Birgitta A. Versluys, Caroline A. Lindemans, Stefan Nierkens.
Supplementary information
The online version contains supplementary material available at 10.1038/s41409-026-02829-w.
References
- 1.Yanik GA, Grupp SA, Pulsipher MA, et al. TNF receptor inhibitor therapy for the treatment of children with idiopathic pneumonia syndrome (IPS): A joint Pediatric Blood and Marrow Transplant Consortium (PBMTC) and Children’s Oncology Group (COG) study (ASCT0521). Biol Blood Marrow Transpl. 2015;21:67–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wenger DS, Triplette M, Crothers K, et al. Incidence, Risk Factors, and Outcomes of Idiopathic Pneumonia Syndrome after Allogeneic Hematopoietic Cell Transplantation. Biol Blood Marrow Transpl. 2020;26:413–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Park M, Koh KN, Kim BE, Im HJ, Seo JJ. Clinical features of late onset non-infectious pulmonary complications following pediatric allogeneic hematopoietic stem cell transplantation. Clin Transpl. 2011;25:168–76. [DOI] [PubMed] [Google Scholar]
- 4.Keates-Baleeiro J, Moore P, Koyama T, Manes B, Calder C, Frangoul H. Incidence and outcome of idiopathic pneumonia syndrome in pediatric stem cell transplant recipients. Bone Marrow Transplant. 2006;38:285–9. [DOI] [PubMed] [Google Scholar]
- 5.Versluys AB, Rossen JWA, van Ewijk B, Schuurman R, Bierings MB, Boelens JJ. Strong Association between Respiratory Viral Infection Early after Hematopoietic Stem Cell Transplantation and the Development of Life-Threatening Acute and Chronic Alloimmune Lung Syndromes. Biol Blood Marrow Transplant. 2010;16:782–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Versluys B, Bierings M, Murk JL, et al. Infection with a respiratory virus before hematopoietic cell transplantation is associated with alloimmune-mediated lung syndromes. J Allergy Clin Immunol. 2018;141:697–703.e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sano H, Kobayashi R, Iguchi A, et al. Risk factor analysis of idiopathic pneumonia syndrome after allogeneic hematopoietic SCT in children. Bone Marrow Transplant 2014 49:1. 2013;49:38–41. [DOI] [PubMed] [Google Scholar]
- 8.Griese M, Rampf U, Hofmann D, Führer M, Reinhardt D, Bender-Götze C. Pulmonary complications after bone marrow transplantation in children: Twenty-four years of experience in a single pediatric center. Pediatr Pulmonol. 2000;30:393–401. [DOI] [PubMed] [Google Scholar]
- 9.Wieringa J, Van Kralingen KW, Sont JK, Bresters D. Pulmonary function impairment in children following hematopoietic stem cell transplantation. Pediatr Blood Cancer. 2005;45:318–23. [DOI] [PubMed] [Google Scholar]
- 10.Duncan CN, Buonanno MR, Barry EV, Myers K, Peritz D, Lehmann L. Bronchiolitis obliterans following pediatric allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant 2008 41:11. 2008;41:971–5. [DOI] [PubMed] [Google Scholar]
- 11.Altmann T, Slack J, Slatter MA, et al. Endothelial cell damage in idiopathic pneumonia syndrome. Bone Marrow Transplant Nat Publ Group. 2018;53:515–8. [DOI] [PubMed] [Google Scholar]
- 12.Fraebel J, Engelhardt BG, Kon Kim T. Noninfectious Pulmonary Complications after Hematopoietic Stem Cell Transplantation. Transpl Cell Ther. 2023;29:82–93. [DOI] [PubMed] [Google Scholar]
- 13.Ilan U, Brivio E, Algeri M, et al. The Development of New Agents for Post-Hematopoietic Stem Cell Transplantation Non-Infectious Complications in Children. J Clin Med. 2023;12:1–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Williams KM, Cheng GS, Pusic I, et al. Fluticasone, Azithromycin, and Montelukast Treatment for New-Onset Bronchiolitis Obliterans Syndrome after Hematopoietic Cell Transplantation. Biol Blood Marrow Transplant. 2016;22:710–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zinter MS, Lindemans CA, Versluys BA, et al. The pulmonary metatranscriptome prior to pediatric HCT identifies post-HCT lung injury. Blood. 2021;137:1679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Klein OR, Ktena YP, Pierce E, et al. Defibrotide modulates pulmonary endothelial cell activation and protects against lung inflammation in pre-clinical models of LPS-induced lung injury and idiopathic pneumonia syndrome. Front Immunol. 2023;14. 10.3389/fimmu.2023.1186422 [DOI] [PMC free article] [PubMed]
- 17.Kulkarni UP, Pai AA, Kavitha ML, et al. Endothelial Activation and Stress Index-Measured Pretransplantation Predicts Transplantation-Related Mortality in Patients with Thalassemia Major Undergoing Transplantation with Thiotepa, Treosulfan, and Fludarabine Conditioning. Transpl Cell Ther. 2022;28:356.e1–356.e6. [DOI] [PubMed] [Google Scholar]
- 18.Luft T, Benner A, Jodele S, et al. EASIX in patients with acute graft-versus-host disease: a retrospective cohort analysis. Lancet Haematol. 2017;4:e414–e423. [DOI] [PubMed] [Google Scholar]
- 19.Luft T, Benner A, Terzer T, et al. EASIX and mortality after allogeneic stem cell transplantation. Bone Marrow Transpl. 2020;55:553–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Abugideiri M, Nanda RH, Butker C, et al. Factors Influencing Pulmonary Toxicity in Children Undergoing Allogeneic Hematopoietic Stem Cell Transplantation in the Setting of Total Body Irradiation-Based Myeloablative Conditioning. Int J Radiat Oncol Biol Phys. 2016;94:349–59. [DOI] [PubMed] [Google Scholar]
- 21.Gassas A, Krueger J, Zaidman I, et al. Infections and neutrophils in the pathogenesis of bronchiolitis obliterans syndrome in children after allogeneic stem cell transplantation. Pediatr Transpl. 2016;20:303–6. [DOI] [PubMed] [Google Scholar]
- 22.Gassas A, Schechter T, Krueger J, et al. Serum Krebs Von Den Lungen-6 as a Biomarker for Early Detection of Bronchiolitis Obliterans Syndrome in Children Undergoing Allogeneic Stem Cell Transplantation. Biol Blood Marrow Transplant. 2015;21:1524–8. [DOI] [PubMed] [Google Scholar]
- 23.Versluijs AB, de Koning CCH, Lankester AC, et al. Clofarabine-fludarabine-busulfan in HCT for pediatric leukemia: an effective, low toxicity, TBI-free conditioning regimen. Blood Adv. 2022;6:1719–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bartelink IH, Belitser SV, Knibbe CAJ, et al. Immune Reconstitution Kinetics as an Early Predictor for Mortality using Various Hematopoietic Stem Cell Sources in Children. Biol Blood Marrow Transplant. 2013;19:305–13. [DOI] [PubMed] [Google Scholar]
- 25.Bartelink IH, Van Reij EML, Gerhardt CE, et al. Fludarabine and Exposure-Targeted Busulfan Compares Favorably with Busulfan/Cyclophosphamide-Based Regimens in Pediatric Hematopoietic Cell Transplantation: Maintaining Efficacy with Less Toxicity. Biol Blood Marrow Transpl. 2014;20:345–53. [DOI] [PubMed] [Google Scholar]
- 26.Admiraal R, Nierkens S, Bierings MB, et al. Individualised dosing of anti-thymocyte globulin in paediatric unrelated allogeneic haematopoietic stem-cell transplantation (PARACHUTE): a single-arm, phase 2 clinical trial. Lancet Haematol. 2022;9:111-31. Accessed February 22, 2022. www.thelancet.com/haematology [DOI] [PubMed]
- 27.Peters C, Dalle JH, Locatelli F, et al. Total Body Irradiation or Chemotherapy Conditioning in Childhood ALL: A Multinational, Randomized, Noninferiority Phase III Study. J Clin Oncol. 2021;39:295–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.De Koning C, Admiraal R, Nierkens S, Boelens JJ. Human herpesvirus 6 viremia affects T-cell reconstitution after allogeneic hematopoietic stem cell transplantation. Blood Adv. 2018;2:428–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Panoskaltsis-Mortari A, Griese M, Madtes DK, et al. An official American Thoracic Society Research Statement: Noninfectious lung injury after hematopoietic stem cell transplantation: Idiopathic pneumonia syndrome. Am J Respir Crit Care Med. 2011;183:1262–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Paczesny S, Hakim FT, Pidala J, et al. National institutes of health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: III: The 2014 biomarker working group report. Biol Blood Marrow Transplant. 2015;21:780–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein2~3~4 AR. A Simulation Study of the Number of Events per Variable in Logistic Regression Analysis. Vol 49.; 1996. [DOI] [PubMed]
- 32.Deleve LD, Wang X Role of Oxidative Stress and Glutathione in Busulfan Toxicity in Cultured Murine Hepatocytes. Pharmacology. 2000;60:143-54. www.karger.comwww.karger.com/journals/pha [DOI] [PubMed]
- 33.Hassan Z, Hellström-Lindberg E, Alsadi S, Edgren M, Hägglund H, Hassan M. The effect of modulation of glutathione cellular content on busulphan-induced cytotoxicity on hematopoietic cells in vitro and in vivo. Bone Marrow Transpl. 2002;30:141–7. [DOI] [PubMed] [Google Scholar]
- 34.Morris PE, Bernard GR. Significance of Glutathione in Lung Disease and Implications for Therapy. Am J Med Sci. 1994;307:119–27. [DOI] [PubMed] [Google Scholar]
- 35.Langenhorst JB, Boss J, van Kesteren C, et al. A semi-mechanistic model based on glutathione depletion to describe intra-individual reduction in busulfan clearance. Br J Clin Pharm. 2020;86:1499–509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Liu KX, Poux N, Shin KY, et al. Comparison of Pulmonary Toxicity after Total Body Irradiation- and Busulfan-Based Myeloablative Conditioning for Allogeneic Hematopoietic Stem Cell Transplantation in Pediatric Patients. Transpl Cell Ther. 2022;28:502.e1–502.e12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Ottaviano G, Lucchini G, Breuer J, et al. Delaying haematopoietic stem cell transplantation in children with viral respiratory infections reduces transplant-related mortality. Br J Haematol. 2020;188:560–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.