Association of Pharmacokinetic Biomarkers with Early Immune Recovery Following HLA-Haploidentical Hematopoietic Cell Transplantation

Guofang Shen^1*Timothy W. Synold¹Shanzay Khan²David A. Horne¹Christopher G. Kanakry²Ryotaro Nakamura³

• ¹Beckman Research Institute, City of Hope, Pasadena, United States
• ²National Institutes of Health National Cancer Institute, Bethesda, United States
• ³City of Hope Duarte Cancer Center, Duarte, United States

Background: Prophylactic immunosuppressants for graft-versus-host disease (GVHD) in allogeneic hematopoietic cell transplantation (alloHCT), including post-transplant cyclophosphamide (PTCy) and mycophenolate mofetil (MMF), exhibit complex pharmacokinetic profiles. Interindividual variations in pharmacokinetic exposure to these immunosuppressants or their metabolites may interfere with treatment outcomes. Method: We conducted a feasibility study (n=11) to investigate the pharmacokinetic/ pharmacodynamic relationship in patients undergoing HLA-haploidentical alloHCT with standard high-dose PTCy (50mg/kg/day on days +3/+4) combined with MMF and tacrolimus or sirolimus. Blood samples were collected to assess the variability of pharmacokinetic biomarkers including exposures (AUCs) of Cy, Carboxycyclophosphamide (cepm), N-Dechloroethyl Cyclophosphamide (dccy), 4-ketocyclophosphamide (ketocy), mycophenolic acid (MPA), and mycophenolic acid glucuronide (MPAG). Serial dynamic changes in immune cell populations, including regulatory T cells (Tregs), over the first 3 post-transplant weeks were monitored. Results: A transient reduction in the proliferation (Ki-67⁺) of activated (HLA-DR⁺) T cells coincided with Cy treatment. The ratio of Tregs to the CD4⁺ T-cell population increased in a time-dependent manner within the first 21 days post-transplant. Moderate interindividual variability was observed across all pharmacokinetic biomarkers. Serum creatinine and blood urea nitrogen levels positively correlated with exposure to Cy and MMF metabolites including Cepm, MPA, and MPAG. Using correlation analysis, we further confirmed the negative association between PK biomarkers and lymphocyte count, but not Treg percentage, suggesting that careful optimization of Cy and MMF dosing may have the potential to support immune recovery, although this requires further validation in larger studies.Using machine learning approach, we identified an association between exposure to several pharmacokinetic biomarkers and delayed immune cell recovery, underscoring the potential to improve immune recovery by optimizing Cy and MMF dosing. Conclusion: The relationship of pharmacokinetic biomarkers to immune and clinical outcomes warrants further investigation in larger studies, but holds promise for personalizing dosing of GVHD prophylaxis to improve patient outcomes after alloHCT.