Graft-versus-host disease prophylaxis with tacrolimus and mycophenolate mofetil in HLA-matched nonmyeloablative transplant recipients is associated with very low incidence of GVHD and nonrelapse mortality.

Incidence of grade II-IV acute graft-versus-host disease (aGVHD) in nonmyeloablative (NMA) transplant recipients remains high. To date, the ideal prophylaxis regimen, which minimizes aGVHD and chronic GVHD (cGVHD), but does not abrogate graft-versus-tumor (GVT) response, has not been described. Because tacrolimus is more potent than cyclosporine (CSA), and because mycophenolate mofetil (MMF) is an effective immunosuppressant that does not lead to mucositis, we hypothesized that a combination of these 2 oral agents may be an effective GVHD prophylactic strategy. We, therefore, designed an outpatient prospective cohort study with a conditioning regimen consisting of fludarabine (Flu) 30 mg/m2 daily and cyclophosphamide (Cy) 300 mg/m2 daily for 5 days followed by infusion of blood stem cells. Tacrolimus 3mg twice a day was started on day (D) -8, adjusted to achieve levels 10-15 nmol/L, continued until D +50 and then tapered by D +100 or +180 according to estimated risk of relapse. MMF 1000 mg twice a day was started on D +1 and discontinued on D +50. To date, 131 patients (males/females: 75/56) with a median age of 54 years have received a 6/6 matched sibling transplant using this protocol. Indication for NMA transplant included age >55 years (24%), expected increased risk of toxicity (28%), or participation in a multiple myeloma (MM) sequential protocol (48%). Most common diagnoses included MM (N = 62), non-Hodgkin lymphoma (NHL, N = 46), and acute leukemia (N = 10). Following infusion of 6.8 x 10(6) CD34+ cells/kg (range: 0.30-22.3), neutrophil and lymphocyte engraftment occurred in 95% of patients by D +180. The estimated cumulative incidence of classical grade I-IV aGVHD by D +120 was 11.6% (95% confidence interval [CI]: 7.1-18.5). No grade IV aGVHD was observed. In addition, 15 patients (12%: CI: 7.4-19.2; median D +140) developed an overlap syndrome consisting of clinical and histologic features of both aGVHD and cGVHD simultaneously. The estimated cumulative incidence of extensive cGVHD was 76.1% (95% CI: 67.4-83.9) at 2 years, with clinical features at presentation similar to other reported series. In patients developing extensive cGVHD, the probability of remaining on immunosuppression at 5 years was 34.8% (95% CI: 16.4-57.3). With a median follow-up of 982 days, the estimated probabilities of nonrelapse mortality (NRM) and overall survival (OS) were 15.5% (95% CI: 9.0-26.1) and 62.7% (95% CI: 51.4-72.1). The cumulative incidence of relapse was 30% at 7 years. Following NMA transplant, disease-free survival (DFS) was highest in recipients with follicular NHL (79.8%: 95% CI: 57.6-91.2) and lowest in large cell NHLs (34.3%: 95% CI: 1.6-75.9). From this large group of patients treated with a uniform conditioning and GVHD prophylaxis regimen, we conclude that aGVHD prophylaxis with early use of tacrolimus and MMF is safe, effective, and associated with low NRM. Future strategies will need to focus on decreasing the incidence of extensive cGVHD without increasing the risk of relapse.