Management and Outcomes for Patients With Hodgkin Lymphoma With Partial Metabolic Response After First-Line Systemic Therapy

PURPOSE: To evaluate real-world management strategies and outcomes for patients with Hodgkin lymphoma (HL) with a partial metabolic response (PMR) on an end-of-chemotherapy (EOC) fluorodeoxyglucose-positron emission tomography (PET) scan and assess whether dynamic changes in maximum standardized uptake value (SUVmax) could refine risk stratification. METHODS AND MATERIALS: Multicenter, retrospective cohort study of patients with HL treated from January 1, 2009, to September 30, 2021. PMR was defined as a Deauville score of 4 on the EOC-PET scan, with an SUVmax value lower than the staging PET scan. First-line chemotherapy was predominantly adriamycin, bleomycin, vinblastine, and dacarbazine (62.7%), or adriamycin, bleomycin, vinblastine, and dacarbazine-escalated bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisolone (28.4%). The primary endpoint was progression-free survival (PFS). Cox proportional hazards models adjusted for stage and EOC-avid sites quantified outcomes; SUVmax value kinetics (interim-to-EOC) were explored within the radiation therapy (RT) cohort. RESULTS: Among 836 patients with EOC-PET scans, 67 met PMR criteria; median follow-up was 3.2 years (range, 0.1-12.6). Post-EOC-PET scan management included involved-site RT (n = 38), salvage chemotherapy ± autologous stem-cell transplantation (n = 14), or observation with serial PET scans (n = 12). RT recipients had more early-stage disease (55.3% vs 21.4%), fewer EOC-avid sites (median, 1 vs 2), and lower EOC SUVmax values (median, 4.7 vs 10.2) than those receiving systemic therapy. Two-year PFS was 84.2% after RT, 52.7% after salvage chemotherapy, and 74.1% with observation (log-rank P = .057). On multivariable analysis, salvage chemotherapy (hazard ratio [HR], 5.82; 95% CI, 1.14-29.81; P = .03) and observation (HR, 5.74; 95% CI, 1.12-29.39; P = .03) were associated with higher progression risk versus RT. Within the RT cohort, rising SUVmax values between interim and EOC-PET scans (HR, 7.21; 95% CI, 1.17-44.35; P = .033) and higher absolute EOC SUVmax values (HR per unit, 1.35; 95% CI, 1.02-1.79; P = .036) predicted inferior PFS. CONCLUSIONS: Most patients with HL with PMR achieve durable remission with consolidative RT alone, avoiding salvage chemotherapy and transplantation. Dynamic changes in SUVmax values-especially a rising SUVmax value between interim and EOC-PET scans-identify a high-risk subset potentially warranting treatment intensification. Prospective studies integrating novel agents and PET scan metrics are needed to personalize therapy for this population.