Outcome and toxicity of ifosfamide, carboplatin, and etoposide versus gemcitabine and vinorelbine regimen for pediatric patients with relapsed or refractory Hodgkin's lymphoma.

Background: Pediatric classical Hodgkin lymphoma (CHL) is a curable disease; however, the optimal salvage regimen is unclear for relapsed/refractory (R/R) disease. This study aimed to compare response rates, toxicity, event-free survival (EFS), and overall survival (OS) of ifosfamide, carboplatin, and etoposide (ICE) with gemcitabine and vinorelbine (GV) regimen after first-line doxorubicin, bleomycin, vinblastine, dacarbazine (ABVD) in pediatric patients with R/R CHL. Methods: This is a retrospective cohort study of 132 pediatric patients with R/R CHL treated from July 2012 to December 2020 with ICE (n = 82) or GV (n = 50). Results: The median age at relapse was 13.9 years, and 68.2% were men. Rates of complete response, partial response, and progressive disease before consolidation were 50.6%, 3.7%, and 45.7% for ICE and 28.5%, 0%, and 71.5% for GV (P = 0.011). By multivariate analysis, regimen (P = 0.002), time to relapse (P = 0.0001), and B-symptoms (P = 0.002) were independent factors to lower response rates. Hematological toxicity, electrolyte disturbance, hemorrhagic cystitis, infectious complications, and hospital admission for fever neutropenia were statistically significant higher for the ICE regimen. Treatment-related mortalities were 2.4% for ICE and 2% for GV (P = 0.86). The 3-year EFS was 39.3% ± 11.4% for ICE and 24.9% ± 12.5% for GV (P = 0.0001), while 3-year OS was 69.3% ± 10.6% and 74% ± 12.9% (P = 0.3), respectively. By multivariate analysis, regimen (P = 0.0001), time to relapse (P = 0.011), B-symptoms (P = 0.001), and leukocytosis (P = 0.007) were significant for EFS, while anemia (P = 0.008), and progressive disease on early response evaluation (P = 0.022) were significant for OS. Conclusions: The ICE regimen had a better overall response rate and EFS, but higher toxicity, than GV; however, OS and mortality were similar.