Development and Preliminary Testing Of The Post-Transplant Multimorbidity Index (PTMI)

Background Comorbid health conditions, both those present before transplant and those acquired as late effects of allogeneic hematopoietic stem cell transplantation (alloHSCT) and the treatment of chronic graft-versus-host disease (cGVHD), represent an important covariate. While the Hematopoietic Cell Transplantation-specific Comorbidity Index (HCT-CI) (Sorror et al. Cancer, 2008; 112(9):1992-2001) has been developed to predict non-relapse mortality and overall survival in allo-HSCT recipients, many of the chronic comorbid conditions that develop in the allogeneic post-transplant setting are not represented in the HCT-CI, and its predictive validity has had limited study in long-term transplant survivors. The objectives of this study were to: (i) establish a new measure of multimorbidity for the post-transplant setting (Post-Transplant Multimorbidity Index [PTMI]), and (ii) explore its content validity. Methods Based on a review of the literature and the most commonly used comorbidity measures, the PTMI was developed to reflect a broad and inclusive list of comorbid conditions that may arise in the post-transplant setting. Preliminary definitions to establish the presence of each of these conditions were developed. To enhance longitudinal comparisons, the conditions and definitions from the HCT-CI were nested within the PTMI. The conditions and definitions were iteratively refined (SM and DW) through application in a cohort of 30 post-transplant patients. Subsequently, the PTMI, HCT-CI, Charlson Comorbidity Scale (CCS), and the Functional Comorbidity Index (FCI) were comparatively evaluated in a cohort of 50 alloHSCT survivors referred for comprehensive cGVHD consultation. The evaluation cohort was a mean age of 42 (range 23-68) years and a median of 33.5 (range 13-208) months post-transplant. All but two (late acute GVHD n=1; no cGVHD n=1) had active cGVHD that had been present for a median of 27 (range 3-197) months. A majority had severe cGVHD (NIH global severity score 1 n=4; 2 n=8; 3 n=36) and 90% were currently receiving systemic immunosuppression. Results Applying the PTMI, HCT-CI, CCS and FCI to our evaluation cohort yielded a mean of 5 (SD±2.5), 1.5 (SD±1.23), 1.39 (SD±0.78), and 2.18 (SD±1.16) co-occurring conditions, respectively. On average the HCT-CI, CCS, and FCI missed the identification of one or more comorbidities 71%, 75%, and 43% of the time. Conditions that were prevalent in the cohort but missed by the other comorbidity measures included osteoporosis, avascular necrosis, hypertriglyceridemia, hypothyroidism, BMI<22, and secondary solid malignancy (excluding nonmelanoma skin cancer) after transplant. Using the PTMI, the two most prevalent comorbid conditions were osteoporosis (64%) and underweight/sarcopenia (BMI<22) (46%), whereas the FCI identified osteoporosis and depression as most prevalent, and the HCT-CI psychiatric disturbance and peptic ulcer, and the CCS history of malignancy and peptic ulcer disease, as the two most prevalent comorbidities, respectively. Conclusions Our results offer preliminary evidence that the PTMI improves the identification of chronic comorbid conditions in long-term transplant survivors with cGVHD, and demonstrate that the choice of a measure is an important methodologic issue in the design of epidemiologic studies of comorbidity in post-transplant survivors with cGVHD. Prospective studies evaluating the measurement properties of the PTMI are ongoing in post-transplant survivors with and without active chronic GVHD. With continued testing and refinement, we anticipate that this new measure will have utility for risk-adjustment and stratification in both observational studies and clinical trials in the post-transplant setting. Disclosures: No relevant conflicts of interest to declare.