Haploidentical Transplant Using Selective Ex-Vivo Tcrαβ Depleting and Memory T Cell Add-Back Results in More Favorable Gvhd-Free /Relapse Free Survival (GRFS) As Compared to Post-Transplant Cyclophosphamide for Adults with Hematological Malignancies - a Comparative Analysis of 170 Patients in a Multicenter Study in Singapore

Background: Outcome after haploidentical (Haplo) hematopoietic cell transplantation (HCT) is encouraging and has become alternative option to treat patients with high-risk hematological malignancies without human leukocyte antigen (HLA) matched donor. Currently, ex-vivo T-cell depletion (TCD) and unmanipulated graft with post-transplant cyclophosphamide (PTCy) remain the most commonly employed graft-versus-host disease (GVHD) prophylaxis strategies in patients receiving Haplo-HCT. The aim of this study is to compare the clinical outcome of adult patients with hematological malignancies receiving Haplo-HCT using these two different GVHD prophylaxis strategies.

Methods: We analysed 170 adults’ patients receiving haploidentical peripheral blood stem cell transplant for various hematological malignancies between Jan 2011 and July 2022, following myeloablative (MAC, n=21) or reduced intensity conditioning (RIC, N=149) regimen. 100 patients received Haplo-HCT with selective ex-vivo T cell (TCRαβ and CD45RA+) depleted grafts for GVHD prophylaxis (Haplo-TCD) (Koh LP et al. ASH 2018), whereas 70 patients received unmanipulated, T-cell-replete graft followed by PTCy (Haplo-PTCy), in combination with tacrolimus and mycophenolate mofetil for GVHD prophylaxis.

Results: Median time to neutrophil recovery was 12 days (range, 8-25) and 18 days (range 13-29) for patients receiving Haplo-TCD and Haplo-PTCY, respectively (p<0.001). Primary graft failure was seen in 1 patient in each of the 2 groups. Cumulative incidence (C.I.) of non-relapse mortality (NRM) (22% vs 25%; p=0.508), relapse (22% vs 26%; p=0.561), day 180 grade 2-4 acute GVHD (aGVHD) (35% vs 33%; p=0.539), day 180 grade 3-4 aGVHD (11% vs 3%; p=0.508) were not significantly different between the Haplo-TCD and Haplo-PTCy cohorts. However, C.I. of chronic GVHD (cGVHD) at 2 years was significantly lower in Haplo-TCD as compared with Haplo-PTCy (4% vs 26%, P<0.001).

At a median follow up of 19.7 months (range 0.3- 67.1) in surviving patients, the 2-year overall (OS), event-free (EFS), and GVHD-free/relapse-free (GRFS) survival were 62% vs 49% (p=0.083), 56% vs 42% (p=0.309) and 54% vs 31% (p=0.006) for Haplo-TCD and Haplo-PTCy, respectively.

Multivariable analysis showed that disease risk index (DRI), HCT-Comorbidity Index (HCT-CI) and GVHD prophylaxis strategies were independent risk factors for OS, EFS and GRFS. In contrast, NRM was only influenced by HCT-CI (p=0.007), whereas relapse was only influenced by DRI (p=0.0001). Notably, Haplo-TCD was associated with significantly lower risk of cGVHD in multivariable analysis (HR 0.14; 95% CI 0.04-0.15; p=0.0011), and this translates into significantly more favorable GRFS as compared to Haplo-PTCy (Figure 1).

We performed propensity scoring for matching and identified 54 patients in each cohort. With propensity score matching, the use haplo-TCD retained its significant prognostic impact with lower risk of cGVHD (HR 0.08; 95% CI 1-62; p=0.016), and resulting in more favorable 5 years GRFS (50% vs 30%; p=0.03) as compared to Haplo-PTCy recipients (Figure 2). There was no significant difference in OS, EFS, CI of relapse, NRM, acute GVHD 2-4 and 3-4 between the 2 groups.

Conclusions: As compared to unmanipulated graft and PTCy, Haplo-HCT using ex-vivo TCRαβ and CD45RA+ depleted graft resulted in comparable outcome, with the additional advantage of earlier neutrophil engraftment, lower risk of cGVHD and more favorable GRFS. The difference in the post-transplant outcome with these two GVHD prophylaxis strategies warrants further investigation and confirmation in prospective controlled trials.

No relevant conflicts of interest to declare.