Article · Transplantation and Cellular Therapy · January 31, 2019

A Novel Method to Evaluate Outcomes of Chronic Graft Vs Host Disease by Using National Population-Based Real-World Data

Jonas Mattsson, Thomas Webb, Nishan Sengupta, Nurgul Batyrbekova, Mats Remberger, Frida Schain

Abstract

Introduction Chronic graft vs host disease (cGVHD) affects 30-40% of patients (pts) post HSCT (Socie G, et al. Blood . 2014;124:374). Corticosteroids are associated with significant toxicity, and severe cGVHD is associated with low quality of life, despite advances in overall survival (OS). Swedish national registries allow population-based longitudinal studies of pts with cGVHD. Objectives To identify real-world cGVHD pts and investigate the medical resource utilization (MRU) and burden of disease in Sweden. Methods Nationwide population-based high-quality registries were linked and used to identify cGVHD pts (Fig 1) who underwent HSCT from 2006 to 2015 using a novel method. To correct for clinical underdiagnosis, we defined pts with cGVHD as those who, ≥ 6 months (mos) post HSCT, received systemic steroid treatment for ≥ 3 mos alone (mild) or with extracorporeal phosphoresis and/or immunosuppressive treatment (moderate/severe [mod/sev]). Kaplan-Meier analysis of OS was conducted for pts surviving ≥ 100 days post HSCT. We used relative survival framework to compare estimated excess mortality rate ratios (EMRRs). MRU rates were summarized. Results We included 1439 pts who received HSCT from 2006 to 2015; 37.3% were classified as non-cGVHD, 24.0% as mild cGVHD, and 38.6% as mod/sev cGVHD. At a median follow-up of 3.8 years, OS in non-cGVHD pts was inferior to pts with mild or mod/sev cGVHD (Fig 2A); for pts surviving > 100 days post HSCT, 1-year OS was 62.5%, 87.3%, and 94.8%, and 5-year OS was 54.0%, 63.1%, and 66.9%, respectively. Multivariate analysis adjusted for baseline characteristics such as sex, age, calendar period, source, and donor revealed EMRRs of 9.33 (95% CI, 6.12-14.21) for non-cGVHD and 2.55 (95% CI, 1.56-4.2) for mild vs mod/sev cGVHD pts; the difference was greatest during the first year (≥ 6 mos post HSCT; Fig 2B). Mod/sev cGVHD pts had more outpatient hospital visits than non- or mild cGVHD pts (Fig 3), and experienced significantly more health problems, including but not limited to musculoskeletal/connective tissue, eye/adjacent organ, cutaneous/subcutaneous, and digestive diseases, as well as certain infectious/parasitic diseases (incidence rate ratio 0.89 [95% CI, 0.7-1.11] and 0.79 [95% CI, 0.64-0.98], respectively, vs mod/sev cGVHD). Conclusions Analysis of Swedish registry data using a novel method suggests that cGVHD is underdiagnosed. Pts with non- and mild cGVHD had greater risk of death than pts with mod/sev cGHVD, mainly during the first year, potentially due to infections (possibly related to overimmunosuppression). Pts with mod/sev cGVHD lived longer but with more health problems and greater MRU. These data highlight the importance of evaluating immune function and developing new, efficient cGVHD therapies.

Hematopoietic Stem Cell TransplantationImmunodeficiency and Autoimmune DisordersTransplantation: Methods and Outcomeshematopoietic stem cell transplantationreal-world datamedical resource utilization

A Novel Method to Evaluate Outcomes of Chronic Graft Vs Host Disease by Using National Population-Based Real-World Data

Abstract

More by Nurgul Batyrbekova