Sitagliptin lowered acute GVHD after alternative donor transplant in phase III testing
Alternative donor transplantation remains one of the most difficult settings in hematologic oncology because the immune reconstitution needed to cure disease can also trigger donor T-cell attack on host tissues. In this randomized phase III trial, sitagliptin added to standard antithymocyte globulin-based prophylaxis reduced the risk of grade II-IV acute graft-versus-host disease by day +100, offering a clinically meaningful signal in a population at high baseline risk for inflammatory complications.
The biologic logic is straightforward. Dipeptidyl peptidase-4 inhibition has been explored as a way to modulate immune activation, and in the transplant setting that approach may blunt the cascade that allows donor lymphocytes to injure skin, liver, and gut. The signal in this study was strongest in the endpoints that matter most at the bedside: moderate-to-severe acute GVHD, severe acute GVHD, and intestinal GVHD.
Trial design and patient population
The study enrolled adults 18 to 60 years old with hematologic malignancies who underwent myeloablative haploidentical or unrelated donor hematopoietic stem-cell transplantation. Investigators randomized patients in a 1:1 ratio to a standard ATG-based prophylaxis regimen alone or the same regimen plus sitagliptin at 600 mg orally every 12 hours from day -1 through day +14. Standard prophylaxis included antithymocyte globulin, a calcineurin inhibitor, methotrexate, and mycophenolate mofetil.
The trial was prospective, multicenter, open-label, and randomized. From December 22, 2021, to June 16, 2023, 190 of 251 screened patients were enrolled. Median follow-up reached 29.8 months, with a range extending from 0.7 to 41.3 months. That follow-up is long enough to capture not only early acute GVHD, but also chronic GVHD, relapse, nonrelapse mortality, and survival outcomes that often determine whether a preventive strategy truly changes transplant practice.
Acute GVHD outcomes favored the sitagliptin arm
The prespecified primary end point diverged clearly between the two study groups. By day +100, the cumulative incidence of grade II-IV aGVHD was 14.7% with sitagliptin and 31.6% without it, with a subdistribution hazard ratio of 0.41 and a P value of .005.
That reduction is not a minor statistical fluctuation. It reflects a substantial drop in the inflammatory syndrome most likely to force high-dose steroid exposure, prolong hospitalization, and worsen early transplant morbidity. The benefit was also visible in more severe disease: grade III-IV aGVHD occurred in 6.3% of patients receiving sitagliptin versus 18.9% in the control group, corresponding to an sHR of 0.30 and a P value of .010.
Intestinal injury showed a similar pattern. Intestinal aGVHD developed in 12.6% of patients in the sitagliptin group compared with 29.5% in the control group, with a P value of .004. Gut involvement is especially consequential in this setting because diarrhea, malabsorption, mucosal injury, and infection risk often cascade together, making bowel-predominant GVHD one of the most clinically disruptive transplant toxicities.
Early GVHD-free, relapse-free survival improved, but longer-term endpoints did not separate
A more composite transplant outcome also moved in the favorable direction. Day +180 GVHD-free, relapse-free survival was 86.3% with sitagliptin and 72.6% without it, with an hazard ratio of 0.44 and a P value of .016. For transplant physicians, that endpoint matters because it captures the balance between disease control and treatment toxicity rather than focusing on a single organ complication.
Even so, the broader long-term picture did not change in a statistically detectable way. Two-year chronic GVHD, nonrelapse mortality, relapse-free survival, overall survival, and relapse outcomes did not differ between groups. Cytomegalovirus and Epstein-Barr virus reactivation rates were also similar, and the adverse event profile remained comparable across arms.
The lack of separation in survival endpoints should not be read as a failure of the intervention. Acute GVHD prevention is only one part of the transplant equation, and survival after allogeneic transplantation is influenced by relapse biology, conditioning intensity, infection, organ toxicity, and the downstream consequences of immunosuppression. A short prophylactic course that lowers early GVHD may improve morbidity without necessarily shifting 2-year survival in a study of this size.
Safety, engraftment, and transplant feasibility remained intact
No meaningful differences emerged in engraftment, relapse risk, or infection risk, which is an important reassurance for a strategy aimed at immune modulation rather than global suppression. In transplant care, any add-on agent must clear a high bar: it has to suppress harmful alloreactivity without delaying marrow recovery or opening the door to opportunistic viral reactivation. The similar safety profile suggests that sitagliptin can be layered onto ATG-based prophylaxis without an obvious toxicity penalty in this trial population.
For clinicians, the most relevant practical point is that the regimen did not appear to trade one problem for another. The lower acute GVHD rates were not accompanied by a detectable rise in relapse, nonrelapse mortality, cytomegalovirus or Epstein-Barr virus reactivation, or adverse events overall. In a setting where even small immune perturbations can create major consequences, that balance deserves attention.
How this study may influence transplant prophylaxis
This trial positions sitagliptin as a potential adjunct for alternative donor transplantation, particularly where the aim is to reduce early inflammatory injury without materially increasing infectious or relapse-related risk. The strongest signal was in acute disease, especially intestinal and severe manifestations, which fits the concept that early donor-host immune activation is the most modifiable phase of GVHD biology.
The study also leaves open several practical questions. The effect was demonstrated with a short peri-transplant course, and the benefit was clearest in early endpoints rather than durable long-term transplant outcomes. Open-label design can influence supportive care patterns, even in a randomized trial, and the population was restricted to adults receiving myeloablative haploidentical or unrelated donor transplants. Those features define the boundaries of applicability.
For now, the data support a narrow but meaningful conclusion: adding sitagliptin to ATG-containing prophylaxis reduced day +100 grade II-IV acute GVHD and improved early GVHD-free, relapse-free survival, without a detected penalty in relapse, infection, engraftment, or overall survival. The trial does not establish a universal new standard, and it does not answer whether the same benefit would appear in reduced-intensity conditioning, other donor settings, or different prophylaxis backbones.