Phase II Study of Longitudinal Geriatric Assessment with Risk-Adapted Interventions to Reduce Non-Relapse Mortality in Allogeneic Hematopoietic Cell Transplantation for Older Patients with Advanced Myeloid Malignancies

Introduction:

Older patients are increasingly considered for higher-intensity, potentially curative therapy such as allogeneic hematopoietic cell transplantation (alloHCT). Despite the increased use of geriatric assessment (GA) to help identify deficits associated with transplant outcomes, it remains unknown whether prospective, GA-driven interventions could improve survival. In this study, we evaluate the hypothesis that a GA-driven, risk-adapted strategy including selection of conditioning intensity and longitudinal supportive care management reduces one-year non-relapsed mortality (NRM) among older patients with myeloid malignancies undergoing alloHCT and examine its impact on their functional decline and quality of life post-transplant. We present here the primary analysis of the study.

Methods:

We conducted an investigator-initiated, single center, phase II study of using GA to guide the selection of conditioning intensity and supportive care interventions for older patients (>60 years) undergoing first alloHCT (NCT04761770). Based on pre-transplant GA, patients with functional impairment and/or high comorbidity burden (HCT-CI/age >4) were categorized as vulnerable and received reduced intensity conditioning. Patients without functional impairment or high comorbidity burden were categorized as robust and received myeloablative conditioning. We allowed all types of donors and conditioning regimens. The geriatrics team saw patients at least weekly while admitted for transplant, and every 3 months thereafter as outpatient until 1-year post-transplant. The primary endpoint was one-year NRM of 10% comparing to the historical control of 25%, with an estimated study size of 32 patients. Important secondary endpoints included overall survival (OS); progression free survival (PFS); grade 3-5 toxicities; and changes in GA and quality of life. Standard statistical analysis was performed.

Results:

All patients remaining in remission completed at least one year of protocol mandated follow-up post-transplant. The median age of the cohort was 69.3 years (range 61.8 – 79.5), with 38% female and 16% non-white race. Diseases included acute myeloid leukemia (56%), myelodysplastic syndrome (28%), and myeloproliferative disorder (16%). Donors included matched related and unrelated (72%); mismatched unrelated (22%); and haploidentical donors (6%). According to the GA fitness criteria above, 8 patients (25%) received myeloablative conditioning and 24 patients (75%) received reduced intensity conditioning. Graft-versus-host-disease (GVHD) prophylaxis included post-transplant cyclophosphamide (41%), calcineurin inhibitors (41%), and CD34+ selection (19%).

The study met the primary endpoint with one-year NRM of 6.3% (2 events). One patient died of sepsis during transplant admission and the other died of immune-mediated cytopenia and hemorrhage at 9-month post-transplant. Both patients had received reduced intensity conditioning. Eleven patients had relapsed, including 5 out of 6 patients whose disease harbored TP53 mutation. With a median follow-up of 19.9 months for survivors, the one-year OS and PFS was 78% (95% CI 65 – 94) and 63% (95% CI 48 – 82), respectively. Half of the patients developed acute GVHD, of which only two were grade 3 (skin). Four patients (12.5%) developed moderate to severe chronic GVHD, and all had received calcineurin inhibitor based GVHD prophylaxis. CTCAE grade 3 and 4 non-hematologic toxicities were typical of early post-transplant care with the majority being infections, organ toxicities, and metabolic/electrolyte abnormalities. Finally, patients who remained in remission at one-year post-transplant maintained their physical independence and cognitive functioning based on serial GA.

Conclusions:

We showed here for the first time that a GA-driven, risk-adapted transplant approach that incorporates longitudinal GA-directed management reduces NRM and prevents functional decline for older patients undergoing alloHCT. While requiring a randomized, multicenter validation study, our results support the inclusion of GA-driven approaches into standard of care transplant planning to select and optimize older patients prior to and during transplant, and to enhance their recovery following transplant. Post-transplant relapse in this higher-risk group of patients remains an area of unmet need.