Achievement of High Concentration of Cyclosporine-a Is Associated with a Low Incidence of Acute Graft-Versus-Host Disease after Haploidentical Hematopoietic Cell Transplantation Using Post-Transplant Cyclophosphamide and Peripheral Blood Stem Cell Graft

Background: Allogeneic hematopoietic stem cell transplantation using a related haploidentical donor (Haplo-HCT) using post-transplant high-dose cyclophosphamide (PTCy) is increasingly used for patients lacking a matched related or unrelated donor. However, use ofperipheral blood stem cell graft (PBSC) is associated with an increased risk of acute GvHD (aGvHD) compared to bone-marrow graft. Therefore, while cyclosporine A (CsA) and mycophenolate mofetil (MMF) are traditionally initiated after completion of PT-Cy at day +5, we decided to initiate them at day -3 before transplant and to add a low dose of ATG to reinforce GvHD prophylaxis in those patients. With this background, we analyze retrospectively the impact of early initiation of CsA and of CsA concentration on patients' outcome in all patients who underwent Haplo-HCT with PBSC grafts and PTCy.

Patients and Methods: Sixty-one consecutive patients who underwent Haplo-HCT for hematological malignancies between October 2013 and August 2017 were included in this retrospective single-center study. All patients received G-CSF mobilized PBSC as grafts and post-transplantation immunosuppression with CsA and MMF. CsA was administered at a dose of 3mg/kg by continuous intravenous infusion starting from D -3 and changed to twice daily oral dosing as soon as tolerated.MMF was administered at a fixed oral dose of 2g per day starting from day 6 without adjustment. In the absence of GvHD, MMF and CsA were tapered over 4 weeks starting from day 30 and day 60, respectively. CsA blood trough concentrations were monitored 3 times per week during the intravenous treatment and at least once per week after switch to oral dosing. CsA doses were adjusted to achieve blood levels between 200 and 300 ng/mL and to prevent renal dysfunction. The primary endpoint was to determine the impact of the CsA concentration on the risk of grade II-IV and III-IV aGvHD.

Results: Median age was 53 (range, 15-72) years, with 16 male patients (26%) receiving a graft from a female donor. Diagnoses were myeloid (64%) or lymphoid malignancies (36%). According to the Disease Risk Index, patients were considered as low-risk, intermediate-risk, high-risk or very-high-risk (respectively 8%, 56%, 31% and 5%). Twenty-five patients (41%) with refractory disease received a sequential conditioning regimen while the remaining (n=36, 59%) received a RIC/RTC regimen based on fludarabine, busulfan and thiotepa. 51 patients (83%) received ATG (2.5-5 mg/Kg total dose) as part of the conditioning regimen. All patients received standard PTCy, nine at D+3 (15%) and 52 at D+3 and D+5 (85%). All patients engrafted at a median of 18 (range: 13-35) days after Haplo-HCT and the median follow-up among surviving patients was 21 (range: 13-53) months. The median concentrations of CsA at 1, 2, 3, and 4 weeks after Haplo-HCT were 272 (range: 114-911), 296 (range: 132-516), 251 (range: 111-485), and 246 (range: 36-375) ng/mL, respectively. At d180, the cumulative incidences (CIs) of grade II-IV and grade III-IV aGvHD were 39% and 18%, respectively. The CIs of chronic GvHD (cGvHD), extensive cGvHD and relapse were 41%, 19% and 35% at 18 months after Haplo-HCT, respectively. At 18 months after the transplant, the OS, PFS and GPFS rates were respectively 60%, 55% and 48%. In univariate analysis, patients having the lowest CsA concentration in the first week after Haplo-HCT had a significantly higher risk of grade II-IV aGvHD (49% vs 18%; P= .02), severe grade III-IV aGvHD (26% vs 0%; P = .03), cGvHD (P= .02) and extensive cGvHD (P= .04). We do not find statistically significant correlation between CsA concentration and relapse incidence, NRM, PFS, GPFS or OS. In multivariate logistic regression analysis, higher CsA concentration (> 301 ng/ mL; the cut-off value defined by ROC analysis) during the first week following Haplo-HCT was the only independent parameter significantly associated with a reduced risk of grade II-IV and grade III-IV aGvHD (respectively P = .04; RR .11; 95% CI, 0.05-0.94; and P < .0001; RR < .001; 95% CI, 0.000007-0.00006). There was no association with extensive cGvHD (P = .14; RR .11; 95% CI, 0.06-1.48).

Conclusion: We conclude that achievement of high concentration of CsA early after Haplo-HCT using PBSC graft is associated with a low incidence of aGvHD and that CsA should be initiated at time of transplant with adequate monitoring during the engraftment period to reduce the risk of grade II-IV aGvHD.