Background The hematopoietic transplant comorbidity index (HTC-CI) has been developed to determine treatment-related morbidity following allogeneic hematopoietic stem cell transplantation (alloHSCT) and includes obesity and diabetes as risk factors. On the other hand, chronic low-grade inflammation which is regularly associated with obesity and represents a mechanism of insulin resistance might mediate beneficial immune effects as demonstrated in the context of immune-check point inhibition cancer treatment. Literature on the role of a high body mass index (BMI) prior to alloHSCT remains controversial likely due to the complexity of the involved mechanisms that also comprise of increased catabolic rates and the immunonutritional status. In this study, we evaluated clinical outcomes in a large cohort of consecutive patients who underwent alloHSCT. Specifically, we analyzed pre-transplant BMI and immunonutritional scores as well as their dynamic changes in the early post-transplant phase with regard to survival and toxicities.

Materials and Methods Clinical records of 664 consecutive patients undergoing alloHSCT between 2012 and 2017 at the Department of Medicine I, University Hospital of Cologne, Germany, were retrospectively analyzed. Patients were categorized into four BMI classes and three immunometabolic risk groups according to the modified Glasgow Prognostic Score (mGPS) measured pre-transplant and on day 30 post-transplant. Overall survival (OS), non-relapse mortality (NRM) and the development of a clinically relevant acute graft-versus-host disease (GvHD) ≥2 grade were compared using Kaplan-Meier survival analysis. Additional analyses stratified for sex and focused on a disease and transplant setting homogenized cohort.

Results Median BMI of the cohort was 24.6 (15.1-50.4) kg/m². OS and NRM differed significantly between BMI classes (OS p = 0.02; NRM p = 0.05), with a significant survival benefit of overweight (median OS: 21 and 22 months in normal weight and obese, > 50% alive after 60 months in overweight). In contrast to the male cohort, in females also obesity had a favourable impact (p = 0.50; median OS: 16 months in normal weight; 35 months in overweight and >50% alive after 60 months in obese). mGPS classes, both determined pre-transplant and on day 30, experienced significantly different OS and NRM (OS p < 0.001; NRM p < 0.002), in which hypoalbuminemia combined with elevated C-reactive protein (mGPS 2) correlated with worst OS, NRM and a tendency of higher GvHD incidence. The extend of mGPS increase from day 0 to 30 impacted all outcomes significantly (OS p = 0.02; NRM p = 0.05; GvHD p = 0.01). Especially an increase to an mGPS of 2 was associated with significantly worse OS, NRM and higher GvHD incidence (median OS: 13 and 7 months in mGPS 0->2 and 1->2, respectively; 49 months in mGPS 0->1).

Conclusions Our data suggest a more complex role of metabolic pathologies as currently reflected by obesity and diabetes categories within the HTC-CI. Therefore, future prospective studies that include body composition as well as sensitive measures of disturbed glucose tolerance and metabolic rates are warranted to determine immunometabolic risk factors for alloHSCT outcomes.

H. Thurisch: None. K. Althoff: None. S. Leitzke: None. U. Holtick: None. C. Scheid: None. S. Oganesian: None. M. Funk: None. D. Cordas dos Santos: None. M. von Bergwelt-Baildon: None. S. Theurich: None.