Patient Selection and Reduced-Intensity Conditioning for Allogeneic Transplant

Rawan Faramand, MD, describes how she selects a transplant approach in clinic. She starts with the disease and its risk of relapse and asks whether newer therapies can deepen remission before transplant. Practicing in Florida with an aging population, she assesses physiologic fitness rather than chronologic age alone, and she notes that safer platforms are allowing transplant to be offered to more patients, including older ones. Timing also matters: patients with high-risk disease may not be able to wait for a matched unrelated donor from the registry, and a haploidentical transplant is sometimes the most appropriate choice. For patients with a matched related or unrelated donor who are candidates for myeloablative conditioning, Dr Faramand considers Orca-T a very good option given its nonrelapse mortality and its reductions in acute and chronic graft-vs-host disease (GVHD), although she would like to see longer follow-up. Caspian Oliai, MD, MS, then addresses reduced-intensity conditioning, which is how many older patients and patients with comorbidities become eligible for a potentially curative transplant. Its downside is less disease control. He explains that the conditioning regimen suppresses low-level residual disease until the graft-vs-leukemia effect matures, which may take 3 to 6 months, and that beyond a certain disease burden the graft-vs-leukemia effect cannot catch up. With reduced-intensity conditioning, preserving that effect is critical, so prolonged high-level immunosuppression may raise relapse risk. Maintenance strategies with targeted inhibitors of FLT3, IDH, and now KMT2A help, but he identifies TP53-mutated disease as the largest remaining unmet need, where hypomethylating agents are commonly used without strong supporting evidence. Dr Oliai sees potential for Orca-T in this setting because single-agent GVHD prophylaxis and earlier immune reconstitution may strengthen the graft-vs-leukemia effect, although longer-term data will be needed.