Initial studies of FK506 combined with methotrexate (MTX) in patients receiving unrelated donor BMT have demonstrated a possible-decrease in the incidence of severe GVHD but high rates of severe stomatitis and nephrotoxicity. With this background, we undertook a pilot study evaluating FK506 in combination with a lower than usual dose of MTX in an attempt to improve the tolerability of this immunoprophylaxis regimen. Between July 1993 and October 1994, 26 consecutive adults receiving unrelated donor BMT at Emory University Hospital were enrolled on this study. All patients received FK506 intravenously at an initial dose of 0.03 mg/kg/day beginning day -1 and continuing until oral FK506 was tolerated. Patients also received MTX intravenously at 5 mg/m2 on days 1, 3, 6, and 11. The preparative regimen administered to all but one patient included cyclophosphamide at 200 mg/kg over 4 days followed by total body irradiation (TBI) at 1400 cGy in twice daily fractions over 4 days. The median age of patients was 31 years (range: 19 to 52). Sixteen donor/recipient pairs were matched for HLA-A, -B, and -DR by serology and molecular typing. Ten paris were minor mismatches at either class I or class II. Twenty-two of 26 patients (85%) completed four doses of MTX on schedule. Nephrotoxicity was the most common adverse event associated with the administration of FK506: 88% of patients experienced a doubling of their serum creatinine. One patient died of central nervous system hemorrhage prior to engraftment. Twenty-four of the remaining 25 patients (96%) engrafted. Fourteen of 24 patients (50%) evaluable developed grades 2-4 acute GVHD. The rate of severe (grades 3-4) acute GVHD was 25%. Chronic GVHD developed in 11 of 20 (55%) evaluable patients. At a median follow-up of 461 days, 14 patients (54%) are alive. All are relapse-free with a median Karnofsky performance status of 90% (range: 70-100%). The cumulative probability of 2-year disease-free survival is 50% (95% confidence interval [CI]: 0.33 to 0.77); for low risk patients 67% (95% CI: 0.47 to 0.95) and for high risk patients 23% (95% CI: 0.049 to 1.00). These preliminary data indicate that FK506-based immunosuppression following unrelated donor BMT is effective in preventing severe acute GVHD and warrants comparison to CSA-based regimens.