Evaluation of Immunosuppression Levels and Risk of Graft-Versus-Host Disease in Allogeneic Blood or Marrow Transplantation with Post-Transplant Cyclophosphamide.

Abstract

Post-transplantation cyclophosphamide (PTCy) is standard graft-versus-host disease (GVHD) prophylaxis for allogeneic blood or marrow transplantation (alloBMT), although optimal therapeutic levels of immunosuppression (IS) therapy combined with PTCy remain contested. Previously, with tacrolimus and methotrexate GVHD prophylaxis, week 1 tacrolimus levels >12 ng/mL were associated with a decreased incidence of grade 2-4 acute GVHD (aGVHD). We evaluated if associations between aGVHD and early IS levels were observed amongst patients receiving PTCy. This retrospective single-center study consisted of 349 patients who received PTCy and mycophenolate mofetil, with either tacrolimus (n=185) or sirolimus (n=164) from September 1, 2017, to September 30, 2019. The median age of patients receiving tacrolimus and sirolimus were 58 and 54 years, respectively. The primary diagnosed diseases for both cohorts were acute myeloid leukemia, acute lymphoblastic leukemia, myelodysplastic syndrome, and lymphoma. While most patients receiving tacrolimus were bone marrow graft sourced (78.4%), the majority of patients receiving sirolimus were peripheral blood sourced (80.5%). All patients were transplanted with FluCyTBI as the conditioning regimen. The primary outcome was grade 2-4 aGVHD incidence at 150 days post alloBMT between weekly IS levels <10 ng/mL versus ≥10 ng/mL throughout the four weeks post-alloBMT. Secondary endpoints included moderate to severe chronic GVHD (cGVHD) incidence, median overall survival (OS), relapse-free survival (RFS), and GVHD-free relapse-free survival (GRFS) at 2 years and the correlation between weekly IS levels <10 ng/mL versus ≥10 ng/mL throughout four weeks post-alloBMT. Patients receiving tacrolimus were compared to others in the tacrolimus cohort, and similarly for sirolimus. No correlation was found between IS levels at any individual week and cumulative aGVHD incidence for either tacrolimus or sirolimus. In the sirolimus cohort, no correlation for moderate to severe cGVHD was observed. However, at week 4, patients in the tacrolimus cohort with levels ≥10 ng/mL experienced significantly higher incidence of moderate to severe chronic GVHD than patients with weekly levels <10 ng/mL (20% vs 8%, p<0.001). When evaluating survival outcomes, post-alloBMT week 1 tacrolimus levels ≥10 ng/mL were associated with decreased OS (HR 3.84, 95% CI [1.16-12.67]; p=0.027), but no correlation was seen in RFS (HR 1.62, 95% CI [0.56-4.72]; p=0.377), or GRFS (HR 1.56, 95% CI [0.89-2.74]; p=0.124). Post-alloBMT week 1 sirolimus ≥10 ng/mL levels were associated with decreased OS (HR 2.74, 95% CI [1.37-5.48]; p=0.004) and GRFS (HR 1.93, 95% CI [1.19-3.12]; p=0.007), but not RFS (HR 1.60, 95% CI [0.78-3.30]; p=0.202). Overall, early IS levels in patients receiving PTCy-based GVHD prophylaxis did not correlate with aGVHD incidence, although IS levels ≥10 ng/mL were associated with compromised outcomes. Targeting IS levels <10 ng/mL may be optimal when using PTCy-based GVHD prophylaxis.