Long Term Follow-up of Patients Treated with mTOR Inhibitors for Severe Chronic Gvhd Following Allogeneic Stem Cell Transplantation.

Abstract 2254

Poster Board II-231

Chronic graft-versus-host disease (cGVHD) remains a major cause of morbidity and mortality in hematopoietic transplant recipients. The mammalian target of rapamycin inhibitors (mTOR-I) sirolimus and everolimus are immunosuppressants that have activity in the prevention and treatment of acute GVHD. Sirolimus in combination with calcineurininhibitors (CNI) has also been shown to have activity in chronic GVHD, however at the cost of considerable toxicity. Since mTOR-I exert their action by blocking interleukin 2 receptor signaling and thus arresting the cells at G1 phase they are promising candidates for tolerance induction. In addition, they also have antiangiogenic and antiproliferative activity. We hypothesized that treatment of cGVHD with mTOR-I and without CNI would be more effective in inducing tolerance than the combination of both and would have a lower toxicity profile. In this retrospective analysis we report 31 consecutive patients (pts) with severe cGVHD according to the NIH consensus, in whom treatment with mTOR-I (everolimus n=20; sirolimus n=11) was initiated between 2004 and 2006, with a median follow-up after initiation of mTOR-I therapy for surviving pts of 46.5 months. Three pts terminated treatment prematurely (within 3 months after starting therapy) because of adverse events (1xsevere cough; 1xhypertriglyceridemia; 1xmouth ulcers) and they were excluded from further analysis. Of the remaining, 12 pts had de novo or quiescent cGVHD, 5 progressive cGVHD developing from acute GVHD and 11 had cGVHD following donor lymphocyte transfusions. Organ involvement included skin (scleroderma) in 21, lungs in 12, mucous membranes in 9, liver in 6, gut in 2 and eyes in 4 pts. Eighteen pts received mTOR-I in combination with steroids, three received a monotherapy and 7 pts received additional immunomodulatory drugs, no CNI however. To further reduce the risk of adverse events drug trough levels were monitored and the dosing was adjusted to low therapeutical levels (3-8 ng/ml). Eleven pts each had a complete and partial response, respectively, with an overall response rate of 78%. At the time of analysis 6 of the 15 surviving pts were complete off immunosuppression and steroids could be tapered and stopped in 5 additional pts. Best responses were seen in pts with de novo or quiescent cGvHD. Thirteen pts have died. Causes of death were: progressive cGvHD (n=5) relapse of the underlying disease (n=4), infections (n=2), secondary malignancy (n=1), unknown (n=1). No difference was observed in everolimus- and sirolimus-treated pts. With a median treatment time of 24 months (range 3 – 54 mo) the major adverse events possibly related to mTOR-I were hyperlipidaemia and impaired wound healing, especially in pts with ulcerative lesions of skin or mucous membranes. Only one of the pts developed thrombotic microangiopathy (TMA) and infectious complications were rare. In pts treated in combination with steroids frequently a shortened prothrombin time has been observed, suggesting an increased risk for thomboembolic events. Controlled trials comparing this approach with alternative strategies are warranted.