Patients with very high risk hematologic malignancies who cannot find an HLA-matched related or unrelated donor can benefit from haplo-mismatched transplantation. The latter is, however, complicated by frequent and severe infectious complications and disease relapse due to delayed immune reconstitution. We have previously reported that photodynamic therapy (PDT) could selectively deplete donor alloreactive populations while preserving lymphocytes for immune responses. Indeed, the dibromorhodamine derivative TH9402 (Celmed BioSciences) has a propensity to accumulate in activated but not resting T cells. We present results of an ongoing Phase I clinical trial of haplo-mismatched allogeneic stem cell transplant (SCT) supplemented with donor lymphocyte infusions (DLIs) PDT depleted of host-reactive T cells. Nine high-risk patients with hematologic malignancies (5 AML relapsed or refractory, 2 MDS, 1 NHL relapsing after autologous SCT, 1 refractory CLL) entered the trial, 7 are evaluable for acute GVHD and reconstitution. Patients (4 M, 3 F) underwent transplantation with donor cells mismatched at 3 HLA Ags: 2 patients; 2Ags: 4 pts, and DR only: 1 pt). Donor mononuclear cells (MNCs) were incubated with recipient MNCs for 4 days, exposed to TH9402 PDT, stored frozen, and administered on day 30±3 after transplant at 3 graded DLI dose levels: 1×104 (1pt), 5×104 (3pts), and 1.3 x105 (3pts) CD3+ cells/kg. Anti-host cytotoxic T lymphocyte precursors (CTLp) were depleted from DLIs by approximately 1.5 logs, and flow cytometry showed greater than 90% elimination of activated T cells (CD4+CD25+ and CD8+CD25+) by TH9402 PDT. All stem cell grafts underwent in vitro immunomagnetic T cell depletion using CD34+ positive cell selection (Miltenyi). Median age at SCT was 57 years (range: 40–58). Five patients were in partial remission or had progressive disease, and 2 patients were in complete remission at the time of SCT. The myeloablative regimen consisted of TBI (1200 cGy), thiotepa (5 mg/kg) and fludarabine (40 mg/m2/day for 5 days) followed by infusion of CD3 depleted HSC grafts. A median of 9.2×106 CD34+ cells/kg were infused on day 0. No GVHD prophylaxis was administered. Evaluable patients showed durable hematologic engraftment: median time to >0.5×109 granulocytes/L was 10.5 days (8–20), and to >20×109 platelets/L without transfusion, 12 days (9–137) and achieved complete donor chimerism. No patient developed acute GVHD (grade II–IV), while 3 patients developed signs of chronic GVHD. Two patients died: one (cohort 1) of a post-transplant lymphoproliferative disease, and one (cohort 2), of relapsed AML. No other patient relapsed. Two pts (1 in cohort 2 and 1 in cohort 3) recovered greater than 0.3×109 CD3+ and CD4+ cells/L at 2 and 5 months post-DLI, and 4 pts had >0.2×109 CD3+ cells/L at 6 mo post-DLI. Although 4 patients developed infectious complications (HSV, CMV, Nocardia, Aspergillus), all resolved rapidly with appropriate therapy. The overall disease-free-survival and survival are 57% at 1 year (median follow-up: 9.4 mo). Our results indicate that the post-transplant infusion of a PDT treated DLI is feasible, does not induce acute GVHD, and may accelerate T cell reconstitution. This PDT strategy could represent an appealing alternative for patients in the higher age range who are at high risk for GVHD.