Abstract
Ongoing acute graft versus host disease (aGVHD) after allogeneic stem cell transplantation (alloSCT) causes significant morbidity and mortality with no standard therapy except corticosteroids, which are broadly immunosuppressive and toxic. Cyclophosphamide (Cy) given day 3 and day 4 is effective for prophylaxis of aGVHD but has proven ineffective when given later as in ongoing aGVHD in both murine and human studies. We previously demonstrated that the MEK inhibitor trametinib (Tram) is effective for aGVHD prophylaxis in murine models of aGVHD through the novel mechanism of inhibition of naïve T cells (CD44−CD62L+). Our hypothesis is that ongoing aGVHD could be better inhibited by sequential Cy and post-Cy Tram inhibition (to prevent naïve T cells from re-initiating alloreactivity) without compromising graft-versus-malignancy (GVM) responses.
Mice were administered sequential Cy and Tram post SCT starting day 12. Efficacy of treatment was assessed in a murine model of MHC-mismatched alloSCT. Overall survival was assessed and clinical GVHD was scored three times weekly. Peripheral blood, spleen, and lymph node cells phenotypes were determined by flow cytometry. GVM was also assessed with A20 luciferase/YFP (A20luc/YFP) lymphoma cells imaged with in vivo imaging system (IVIS).
Overall survival following sequential Cy and Tram was identical to that seen with T cell depleted bone marrow alone and significantly superior to treatment using either single-agent Cy or Tram (p<0.05). Clinical GVHD scores were improved by sequential Cy and Tram compared to treatment with either single agent Cy (p=0.002) or single agent Tram (day 38 p=0.03). Timing of initiation of sequential Cy and Tram starting day 12 was significantly better than in the GVHD control group (p=0.002) and also significantly better than after later initiation at day 19 (p=0.005). Initiation of treatment on day 6 resulted in 100% mortality by day 15. Higher doses of Trametinib at 0.3mg/kg were significantly better than the GVHD control (p=0.006). Analysis of peripheral CD4+ and CD8+ T cells demonstrated a more physiologic CD4+CD8+ ratio and a more normal distribution of naïve and effector T cells in the sequential Cy and Tram within CD4+ and CD8+ T cell subsets. Sequential Cy and Tram did not adversely affect peripheral CD4+FOXP3+ regulatory T cells. Using A20 luciferase/YFP (A20luc/YFP) lymphoma cells in a transplant model demonstrated that GVM was still effective with sequential Cy and Tram.
Combination cyclophosphamide and trametinib demonstrates superior overall survival and limits GVHD severity without affecting regulatory T cells or GVM in murine allogeneic stem cell models. Utilizing Cy and Tram are readily translatable and support the notion that clinical approaches can be developed in the future.