Favorable Impact of T Cell Depleted PBSCT on Transplant Outcomes Despite Delayed Donor T Cell Engraftment.

Abstract

Although T cell depleted stem cell transplantation can reduce risks from GVHD, it is associated with incomplete engraftment and higher relapse rates. To mitigate these negative effects we evaluated a T cell depleted PBSCT protocol followed by delayed add-back of donor T cells. Between 2001–2004, 56 patients with hematological malignancies received a preparative regimen of total body irradiation 12Gy, fludarabine 125mg/m2, and cyclophosphamide 120mg/kg followed by a T cell depleted PBSCT from an HLA identical sibling. The T cell dose was adjusted to 2 x 104/kg in all patients. Patients not developing grade <II GVHD received 107CD3 cells/kg between days 45–100. Lineage-specific T cell and myeloid chimerism was monitored day 14, 30, 45, 60 and 90 post-transplant. Patients were aged between 11–56 (median 30y). Twenty six with CML in CP, acute leukemia in first remission, or MDS RA were designated as standard risk (SR) for transplant outcome. Thirty with more advanced or refractory disease were designated high risk (HR). Overall survival was 65±7% at a median of 1 year. Disease risk was the major factor affecting outcome with lower relapse and transplant-related mortality (TRM) and higher disease-free survival (DFS) in SR versus HR patients (3.5±3.5 vs. 42±11%, 7.5±4 vs. 17.5±8% and 85±7 vs 36±9% respectively, p< 0.0001). One patient rejected the transplant and died following two further transplant attempts. The remaining patients achieved rapid 100% donor myeloid engraftment, sustained unless relapse occurred. Donor T cell engraftment was incomplete. 0nly 12.5% of patients had full donor T cell chimerism (>95% donor T cells) on day 14, rising to 50% of patients by day 90, with persistent mixed chimerism in some patients beyond 6 months. Predictably, acute GVHD grade I–IV by d60 was significantly correlated with full T cell chimerism (p< 0.0001). There was no significant difference in day 30 lymphocyte count, CMV reactivation or transplant CD34 dose between fully or partially T cell chimeric recipients at any time point. Full T cell chimerism was significantly more frequent in HR vs SR patients (on day 90: 82 vs 24%, p= 0.002). Kaplan-Meier outcome estimates for mixed vs full T cell chimerism were therefore analyzed separately for SR and HR patients. All outcomes (relapse, TRM and DFS) were more favorable (but not significantly so) for patients who had mixed chimerism at any time point between d14–90: of 13 evaluable HR patients fully chimeric at d30 there was a 74% probability of relapse vs a 58% relapse for15 mixed chimeric patients (p=NS). No patient with SR disease relapsed. There was a non-significant DFS advantage for d30 mixed vs full chimeric patients: 95% vs 85%, and 40% vs 20% for SR and HR patients respectively. These results indicate 1) full T cell chimerism is more frequent in HR patients (who have received large cumulative doses of prior chemotherapy) 2) T cell depleted PBSCT have a high frequency of prolonged mixed T cell chimerism. This does not predispose to late graft rejection, is favorable for reducing GVHD and does not negatively affect CMV reactivation, relapse, TRM and DFS.