S1625 REDUCTION OF GRAFT‐VERSUS‐HOST DISEASE USING DUAL LYMPHOCYTE SUPPRESSION WITH ANTI‐THYMOCYTE GLOBULIN AND LOW‐DOSE POST‐TRANSPLANT CYCLOPHOSPHAMIDE IN HAPLOIDENTICAL PATIENTS AT HIGH GVHD RISK

Background: In the setting of haplo HCT, further reducing GVHD merits examination, especially after HCT from maternal donor or collateral relatives with high risk of GVHD occurrence. With the aim to reduce the risk of GVHD without increasing the incidence of relapse, the combined use of antithymocyteglobulin (ATG) and post‐transplant cyclophosphamide (PTCy) has become feasible. However, the combination regimen has not been validated. Aims: There have been only two studies on the combination of ATG and high‐dose PTCy for hematological malignancy without control cohort or with short follow‐up. Our previous single‐center preliminary report on 40 patients seems to support the feasibility of the combined use of ATG/granulocyte colony‐stimulating factor (G‐CSF) and low‐dose PTCy (14.5 mg/kg on days 3 and 4 after HCT). Thus, these results need to be confirmed on a larger number of patients with hematological malignancies. Methods: In an attempt to assess the efficacy of this strategy in Haplo, we conducted a multicenter analysis in patients transplanted from maternal donor or collateral relatives and treated with ATG‐PTCy (low‐dose) regimen, and compared the results to contemporary cohort of patients with same characteristics who received ATG without low‐dose PTCy (ATG cohort). We identified 162 consecutive patients (ATG‐PTCy cohort = 73; ATG cohort = 89) between 2016 and July 2018. Results: All patients but one in ATG cohort achieved myeloid engraftment by day 30 post‐HCT. We found that both the cumulative incidence of 100‐d severe acute GVHD and 2‐y total chronic GVHD in ATG‐PTCy cohort was significantly lower than that in ATG cohort (5% vs 20%; P = 0.007; and 26% vs 49%; P = 0.001). And severe GVHD was the most common cause of death (33%) in the ATG cohort as compared to none in the ATG‐PTCy cohort. Furthermore, TRM was significantly lower in ATG‐PTCy cohort (4% vs 14%; P = 0.039). In contrast, myeloid recovery was significantly slower (median, 15d vs 12d; p < 0.0001) and 100‐d platelet recovery was significantly lower in the ATG‐PTCy group (87% vs 98%; P = 0.0002). The 100d cumulative incidence of CMV reactivation in ATG‐PTCy cohort was significantly higher than that in ATG cohort (73% vs 33%; P < 0.001). The 100‐d cumulative incidence of EBV reactivation and PTLD was comparable (25% vs 17%; P = 0.21; and 4% vs 2%; P = 0.49). In multivariate analysis, the combined treatment resulted in lower severe acute GVHD (Hazard Ratio 0.18; P = 0.004), chronic GVHD (HR 0.39; P = 0.010), and TRM (HR 0.24; P = 0.023) but slower myeloid and platelet recovery (HR 0.27 and 0.30; both P < 0.001). No significant differences were found in relapse rate and disease‐free survival according to treatment group. A propensity score technique was used to confirm results of main analysis. Summary/Conclusion: Unmanipulated haploidentical transplantation with ATG/G‐CSF and low‐dose PTCy as a GVHD prevention strategy results in lower rates of both acute and chronic GVHD without influencing the GVL effect for patients transplanted from maternal donor or collateral relatives with high risk of GVHD occurrence.