The limited impact of preexisting rheumatologic diseases on outcomes after allogeneic hematopoietic stem cell transplantation: A nationwide Propensity–Matched analysis

Abstract

Abstract Background Rheumatologic diseases (RDs), including rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and other connective tissue disorders, have historically been considered high-risk comorbidities in the setting of allogeneic hematopoietic stem cell transplantation (allo-HSCT). The Hematopoietic Cell Transplantation Comorbidity Index (HCT-CI) assigns a score of 2 to RD, reflecting its presumed association with increased non-relapse mortality (NRM). This assumption is based on data from earlier transplant eras, when disease control and immunosuppressive management were inadequate. Given recent advances in transplant-related supportive care and RD treatment, reassessment of the prognostic impact of preexisting RD in contemporary risk models is warranted. Methods We performed a retrospective cohort study using the Transplant Registry Unified Management Program (TRUMP) database, a comprehensive national registry that captures data from over 200 transplantation centers in Japan. Eligible patients were aged 16 years or older, and underwent their first allo-HSCT between 2006 and 2018 for hematologic malignancies. Patients with a prior diagnosis of RD at the time of transplantation were compared with those without RD. To minimize confounding, we performed 1:4 (RD:control) propensity score matching (PSM) based on the following variables; previous history of RD, patient age, patient sex, patient-donor sex combination, disease type, disease stage at allo-HSCT, TBI, ATG/ALG, type of GHVD prophylaxis, HCT-CI score calculated excluding history of RD, donor type, cell source, conditioning regimen, time from diagnosis to allo-HSCT, performance status (PS) at allo-HSCT, and year at allo-HSCT. The primary endpoints were overall survival (OS) and NRM; secondary outcomes included cumulative incidence of relapse (CIR), graft-versus-host disease (GVHD), and neutrophil engraftment. Results Among 25,606 patients, 224 (0.9%) had a documented history of RD. After PSM, 216 RD patients and 864 matched controls were analyzed. data on the specific types of RD and their corresponding treatments were largely unavailable. Median age was 58 years, and the median follow-up among survivors was 41.1 months. The standardized mean difference for all variables except HSCT year was below the threshold of 0.1; there were no statistically significant differences in any baseline characteristics. The 3-year OS was 41.1% in the RD group versus 44.7% in the control group (P = 0.36). NRM was 31.8% versus 26.0% (P = 0.21), and CIR was 31.8% versus 34.3% (P = 0.43), showing no statistically significant differences. Rates of grade II–IV acute GVHD were comparable between the RD (33.2%) and control (32.7%) groups, as were chronic GVHD rates (22.9% vs. 24.3%). Day-30 neutrophil engraftment occurred in 89.8% of RD patients and 87.0% of controls (P = 0.16). In univariate Cox regression, RD was not a significant predictor for OS, NRM, or relapse. Subgroup analyses stratified by donor type and conditioning regimen showed consistent findings, with no excess mortality or GVHD attributable to RD history. Furthermore, we evaluated a modified Hematopoietic Cell Transplantation Comorbidity Index (mHCT-CI), in which RD was excluded, to determine whether the prognostic utility of the HCT-CI was preserved without including rheumatologic disease in the scoring system. The mHCT-CI retained strong predictive performance: patients with an mHCT-CI ≥3 had significantly worse 3-year OS (37.1%) and higher NRM (31.1%) compared to those with an mHCT-CI <3 (OS 51.7%, NRM 21.6%; both P < 0.01). These findings reaffirm the predictive validity of the HCT-CI even after excluding RD and suggest that RD may no longer be a necessary component of the index in contemporary transplant practice. Conclusions The modified Hematopoietic Cell Transplantation Comorbidity Index (mHCT-CI), in which a history of rheumatologic disease (RD) is excluded, demonstrated preserved prognostic accuracy for transplant outcomes. These findings suggest that a history of RD should not necessarily preclude transplantation and that existing prognostic models such as the HCT-CI may need to be revised to reflect current transplant practices. This study did not obtain information on the use of biologic agents or molecular inhibitors commonly prescribed for rheumatologic diseases. To better define the prognostic relevance of RD in modern transplant settings, future studies should collect and analyze treatment-specific data.