Pharmacogenetic Predictors of Toxicity in Dose-Intensive Therapy for Pediatric Advanced-Stage Hodgkin’s Lymphoma.

Outcomes for patients with advanced-stage Hodgkin’s lymphoma (HL) remain suboptimal, as dose-intensified regimens produce relatively high cure rates at the cost of increased short- and long-term complications. Prospectively identifying the subset of patients who are at risk of therapy-related complications would allow refinement of treatment regimens. We hypothesized that functional polymorphisms in chemotherapy drug metabolizing enzymes will predict the risk of toxicity for pediatric patients with advanced-stage HL. Glutathione-S-transferase T1 (GSTT1) and glutathione-S-transferase M1 (GSTM1) participate in protection against oxidative damage. Previously, we demonstrated that genetic deletion of these enzymes, a common population polymorphism, increase the risk of regimen-related mortality in pediatric patients with acute myeloid leukemia. The human DNA repair gene XRCC1 is involved in the repair of DNA single-strand breaks, generated in response to alkylating agents and ionizing radiation, and a frequent single nucleotide polymorphism with reduced function (codon 399 arginine to glutamine substitution) has been associated with an increased risk of secondary malignancy. We examined the effects of polymorphic alleles of GSTT1, GSTM1 and XRCC1 on the risk of toxicity in 72 pediatric patients receiving chemotherapy for advanced stage (clinical stage IV, stages IIB/IIIB with bulk disease) Hodgkin’s lymphoma. Therapy consisted of intensified BEACOPP induction (4 cycles of bleomycin, etoposide, adriamycin, cyclophosphamide, vincristine, procarbazine, prednisone) followed by response-adapted continuation therapy. This regimen (Children’s Cancer Group Study 59704) is associated with a 4-year event-free survival rate of 95%. We found that polymorphic low/loss of function alleles were associated with an increased frequency of hematologic toxicities. We observed an increased frequency of therapy-associated thrombocytopenia in patients with absent GSTM1 (79% pts with grade 3/4 thrombocytopenia when GSTM1 absent vs 55% when present, p=0.0491). There was also an increased frequency of infection with neutropenia in patients with absent GSTT1 (57% when GSTT1 absent vs 23% when present, p=0.0522). The XRCC1 variant allele was associated with increased frequency of infection with neutropenia in a dose-dependent fashion (glutamine/glutamine: 67% with toxicity, arginine/glutamine: 27%, arginine/arginine: 13%; p= 0.0308). Similarly XRCC1 genotype was associated with increased frequency of thrombocytopenia (glutamine/glutamine: 67%, glutamine/arginine: 35%, arginine/arginine: 17%; p= 0.0498). Conclusion: These results, representing the only reported data in HL, suggest that deletion of GSTT1 and GSTM1, and variant alleles of XRCC1 may influence risk for toxicity with intensive BEACOPP chemotherapy. Prospective studies in future cohorts are needed to confirm these observations.