Conference Abstracts - 2026 National Summit on Hematologic Cancers
Vol. 6, Issue Supplement 2, 2026 · S1-2
Pre-Transplant Growth Differentiation Factor-15 Independently Predicts Survival Following Allogeneic Hematopoietic Cell Transplantation
Rahul Thakur, MD,Nisha Nair, MPH,Chelsea Peterson, DO,Alicia Lieberman, MD,Hillary Jacobson, DPT,Renee McKenzie, DPT,Andrew Ray, PhD,Alex Niu, MD,Grant Schofield, MD,Ehsan Malek, MD,Maureen Ross, MD, PhD,Philip McCarthy, MD,Marco Davila, MD, PhD,Brian Betts, MD,Shernan Holtan, MD,Megan Herr, PhD
Submission received: 2026-08-16 / Accepted: 2026-08-25 / Published: 2026-09-25
Abstract
Background
Outcomes following allogeneic hematopoietic cell transplantation (allo-HCT) have improved substantially, yet identifying physiologic vulnerability before transplantation remains challenging. Growth differentiation factor-15 (GDF15) is a stress-responsive cytokine induced by mitochondrial dysfunction, tissue injury, chemotherapy, and inflammation. Through GFRAL-RET signaling, GDF15 regulates appetite and energy homeostasis and contributes to cancer-associated cachexia. Recent clinical studies demonstrating activity of GDF15 inhibition suggest this pathway is therapeutically actionable. We evaluated whether pre-transplant GDF15 predicts outcomes following allo-HCT.
Methods
Adults undergoing allo-HCT between 2018–2025 with available pre-conditioning serum were included. GDF15 was quantified by ELISA and categorized into tertiles. Outcomes included 1-year overall survival (OS), progression-free survival (PFS), non-relapse mortality (NRM), relapse, and acute/chronic GVHD. Survival was analyzed by Kaplan-Meier/log-rank methods and competing-risk outcomes by cumulative incidence and Gray's test. Multivariable regression evaluated associations with OS and NRM.
Results
Among 152 patients (median age 61 years; 58% ≥60), 64.5% had AML/MDS, 53.3% were transplanted in complete remission, and 86.8% received reduced-intensity/non-myeloablative conditioning. Patients were divided into GDF15 tertiles (<548, 548–1183, and >1184 pg/mL). Higher pre-transplant GDF15 was associated with significantly inferior 1-year OS (≥1184: 51%, 548-1183: 68%, <548: 88%; log-rank P=0.0001) and PFS (≥1184: 47%, 548-1183: 63%, <548: 82%; log-rank P=0.0004) in univariable analyses. On multivariable Cox regression, pre-transplant GDF15 remained independently associated with inferior 1-year OS (overall P=0.02). Compared with GDF15 <548, patients with GDF15 between 548–1183 and >1184 pg/mL had an increased risk of death (HR=2.6, 95% CI, 0.98–6.68; HR=3.7, 95% CI, 1.5–9.2, respectively), after adjusting for donor type and coronary artery disease. Pre-transplant GDF15 was also independently associated with 1-year NRM (overall P=0.04). Compared with the lowest tertile, the highest two tertiles had a higher 1-year NRM (HR=4.2, 95%CI, 1.4-13.2; P=0.01), after adjustment for donor type and age. Elevated GDF15 was not associated with chronic GVHD, but it was associated with a higher cumulative incidence of grade II–IV acute GVHD (Gray's test P=0.02). Prespecified age-stratified analyses demonstrated a consistent prognostic association across age groups without evidence of meaningful effect modification.
Conclusion
Elevated pre-transplant GDF15 independently predicts inferior survival and NRM following allo-HCT and identifies patients with increased biologic vulnerability beyond conventional clinical risk factors. As a stress-responsive cytokine with an emerging therapeutically targetable pathway, GDF15 represents a promising biomarker to refine pre-transplant risk assessment and provides a rationale for prospective studies evaluating biomarker-guided optimization and GDF15-directed therapeutic strategies before allo-HCT.