Peripheral Neuropathy in Children with High Risk Hodgkin Lymphoma (HL): The Role of Protocol-Stipulated Dose Modification in the Children's Oncology Group (COG) AHOD1331 Study

Abstract

Background Peripheral neuropathy (PN) is commonly associated with several classes of chemotherapy, including vinca alkaloids and the novel anti-CD30-directed antibody-drug conjugate, Brentuximab vedotin (Bv). The drug moiety in Bv, monomethyl auristatin E, a potent microtubule tubulin inhibitor, is known to cause PN. In the 2018 Echelon-1 study, rates of PN among adults with advanced stage HL were 67% in the Bv arm as compared to 43% in the control arm (Connors, 2018) with dose modifications at the discretion of the treating physician of 66% for Bv and 57% for vinblastine (VBL). We examined rates of PN and use of protocol-stipulated dose modification in the COG AHOD1331, a Phase 3 randomized comparison of standard multi-agent chemotherapy (ABVE-PC) vs. Bv-AVE-PC among children and adolescents with high risk HL (Castellino, 2022). The combination of vincristine (VCR) with Bv warrants the formal study of PN in this trial. Methods Eligible patients (pts) ages 2-21 years with high risk HL were enrolled from 2015-2019 (NCT02166463). After randomization, pts received 5 cycles of chemotherapy. Pts in the standard (std) arm received VCR on Day 1 and 8 of each 21-day treatment cycle; pts in the experimental (exp) arm received Bv (1.8 mg/kg) on Day 1 and VCR on Day 8. Clinician grading of PN was required at each cycle, using the Balis PN Scale, with protocol-stipulated dose modifications of VCR preceding modification of Bv in cases of PN > grade 2. Rates of PN were calculated by treatment cycle by study arm. Dose modifications were summarized for each drug by treatment cycle. Median time to first dose modification was estimated by treatment arm. A Cox model estimated hazard ratios (HR) and 95% confidence intervals (CI) for dose modifications by treatment arm and pt and clinical characteristics. Progression-free survival (PFS) was compared with Log-rank test by study arms and within arm by presence or absence of PN. Results 587 pts were enrolled on the study. While the proportion of pts overall developing PN did not differ by study arm (19%, p=0.86), dose modification of Bv, VCR, or both occurred more frequently on the exp arm (50/298, 16.8%) than on the std arm (12/289, 4.2%). Among pts on the exp arm, Bv was dose modified in 8% of pts; VCR was dose modified in 13.4% of pts. The median time to first dose modification was 40.8 days (31.5-73.5) for the exp arm vs. 75.5 days (59.0-94.8) for the std arm (p=.0043). During cycle 5, the rate of any PN was 3.1% in the exp arm vs. 7.8% in the std arm. Compared to the std arm, the HR for dose modification was 5.4 in the exp arm (95% CI, 2.3-12.6, p<0.0001); none of the other pt or disease characteristics were significantly associated with dose modification of the tubulin inhibitors. PFS was significantly better in the exp arm and did not differ by PN status (Figure). Conclusions The rate of dose modifications for PN was consistent with protocol-specified dose modification among children and adolescents treated with known tubulin inhibitors as part of a multi-drug regimen for high risk HL. The strategy to preserve Bv dose in the experimental arm was successful. Overall rates of PN were substantially lower than previously reported in adults treated with Bv. While there were no significant differences in the rate of PN between study arms, we found shorter median time to dose modification in the exp arm, corresponding to the second treatment cycle, as compared to the 3rd cycle for the std arm. Despite dose modifications of vincristine and Bv as management strategies for PN in the experimental arm, patients treated with Bv-AVEPC had superior disease control in children with high-risk HL. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal