Outcomes and Impact of Donor Lymphocyte Infusion after Reduced Intensity Conditioned-Alemtuzumab Allogeneic Hematopoietic Stem Cell Transplantation for Patients with Mature Lymphoid Malignancies

Allogeneic hematopoetic stem cell transplantation (allo-HSCT) offers long-term remission for patients with lymphoid malignancies. In particular, reduced intensity conditioned (RIC) protocols with Alemtuzumab T-cell depletion lead to durable engraftment and reduced graft versus host disease (GvHD). However, such regimens may be associated with reduced graft-versus-lymphoma (GvL) effect and higher incidence of disease progression, leading to increased use of donor lymphocyte infusion (DLI). This study examined transplant outcomes and use of DLI in a large UK RIC-Alemtuzumab conditioned HSCT cohort.

288 consecutive adult patients from three UK transplant centres (Birmingham, Glasgow, Oxford) undergoing first Alemtuzumab-based RIC HSCT for Hodgkin's lymphoma (57%), Non-Hodgkin's lymphoma (24%) and chronic lymphocytic leukemia (19%) between 2000 and 2012 were included. Overall survival (OS), current progression-free survival (CPFS), relapse incidence (RI), transplant-related mortality (TRM), incidence of GvHD and effects of DLI were analyzed.

Median age was 49y (range 17-68) and 60% were male. Patients were heavily pre-treated with 58% having received ≥3 prior lines of treatment and 49% having failed a previous autograft. 55% received stem cells from an HLA-matched unrelated donor and 94% received a peripheral blood stem cell harvest. The most frequent conditioning regimen was Fludarabine, Melphalan and Alemtuzumab (FMC) (70%) and the median total Alemtuzumab dose was 50mg (range 30-100). Median duration of follow-up for survivors was 64 months.

3y OS and CPFS were 53% and 50%, respectively. Factors negatively associated with OS in univariate (UV) analysis were previous autograft (p=0.01), no requirement for DLI (p=0.05), disease status not CR (p=0.05), previous treatment number ≥3 (p=0.007) and failure to engraft plts (p=0.001). On multivariate (MV) analysis, only previous autograft (HR 1.58 p=0.01) and failure to engraft plts (HR 2.3 p=0.0001) retained significance. Factors negatively associated with CPFS on UV analysis were no requirement for DLI (p=0.006), failure to engraft plts (p=0.0003) and previous treatment lines ≥3 (p=0.04). On MV analysis, no DLI remained a significant predictor of adverse CPFS (HR 1.8 p=0.01), as did failure to engraft plts (HR 1.77 p=0.005) and ≥3 treatment lines (HR 1.46 p=0.04).

At 3y, the cumulative incidence of relapse and TRM was both 27%. The incidence of grade 2-4 acute GvHD (aGvHD) and extensive chronic GvHD (cGvHD) was 22% and 18%, respectively. UV predictors of RI were unrelated donor (p=0.0009), HLA mismatch (p=0.04), age > median (p=0.04) and lack of aGvHD (p=0.04). The presence of cGvHD may have been protective against relapse (p=0.08). No variables retained significance on MV modelling. UV predictors of TRM were failure to engraft plts (p=0.001), HLA mismatch (p=0.005), age > median (p=0.004), previous autograft (p=0.02), FMC conditioning (p=0.03), ≥3 previous lines of treatment (p=0.001) and development of aGvHD (p=0.04). On MV modelling, failure to engraft plt (p=0.001), HLA mismatch (p=0.03), age > median (p=0.003) and treatment number ≥3 (p=0.04) retained significance.

62/288 patients received DLI; 37 receiving prophylactic DLI (pDLI) for mixed chimerism and 25 receiving therapeutic DLI (tDLI) for disease relapse. The median dose of DLI was 1x 10^6 CD3+ cells/kg (range 0.1 - 10 x 10^6). Median time from HSCT to first DLI was 10.5 months (range 3-30), to 2nd DLI 15 months (6-51) and to 3rd DLI 19 months (9-30). At 3y, OS after pDLI was 75% and after tDLI was 29%. 29/37 (78%) of pDLI and 14/25 (56%) of tDLI recipients achieved full or stable mixed chimerism. 3y RI after pDLI was 29% and after tDLI was 59% (p=0.02). Overall, the incidence of GvHD post-DLI was 32% (48% tDLI, 22% pDLI, p=ns). Median time to GvHD after last DLI was 46 days (range 17-249). There was no significant impact of post-DLI GvHD on RI (RI at 1y post DLI 18% in GvHD cohort vs 21% with no GvHD, p=ns), but GvHD increased TRM at 1y post DLI (17% in the GvHD cohort vs 3% with no GvHD, p=0.05).

This is the largest series to date with long-term follow-up examining outcomes after Alemtuzumab-based RIC HSCT for mature lymphoid malignancy. Our data indicate that allo-HSCT is effective and is associated with good long term outcomes. Use of pDLI for was associated with excellent outcomes and supports use of DLI in patients developing mixed chimerism following allo-HSCT for lymphoid malignancies.