ISHL10 Abstract T003

Long-term follow-up of SWOG S0816: response-adapted therapy of advanced stage Hodgkin Lymphoma using early interim FDG-PET imaging

S0816 was the first US intergroup clinical trial using early interim FDG–PET imaging to determine the utility of response-adapted therapy for adults with stage III–IV classical Hodgkin Lymphoma. After 2 initial cycles of ABVD, PET2-negative patients (Deauville 1-3) received an additional 4 cycles of ABVD, while PET2+ patients (Deauville score 4-5) switched to eBEACOPP for 6 cycles. Among 336 eligible and evaluable patients, the median age was 32 (18-60), with 52% stage III, 48% stage IV, and 49% IPS 0-2, 51% IPS 3-7. Central review of the interim PET2 scan was performed in 331 evaluable patients, with 271 (82%) PET2-negative and 60 (18%) PET2+. Of 60 eligible PET2+ patients, 49 switched to eBEACOPP as planned and 11 declined. The median follow-up of patients last known alive now exceeds 45 months (range 2.1-76 months). Eighty-two patients have either progressed or died, for an estimated 2-year progression-free survival of 79% (95%CI: 73.8%-82.7%). There were 61 failures in the PET2-negative group (2-year PFS 82%) and 20 failures in the PET2+ group (2-year PFS 64%), see figure. In the entire trial, only 19 patients died, for estimated 2-year overall survival of 98%. Long-term or late (post therapy) grade 3 adverse events on ABVD arm were uncommon including 1 case each of neuropathy, heart failure, DVT, diarrhea and prolonged neutropenia. There was one case of late grade 3 osteonecrosis on the eBEACOPP arm. Six patients (1.4%) have developed secondary malignancies, including 3 cases (1%) on ABVD (non-Hodgkin lymphoma, renal cell carcinoma, and melanoma) and 3 cases (6%) on eBEACOPP (non-Hodgkin lymphoma, renal cell carcinoma and skin cancer). We conclude that response-adapted therapy based on interim PET imaging after two cycles of ABVD appears promising and durable with a high 2-year PFS for PET2+ patients (64%), and outstanding overall survival. However, surprisingly, with prolonged follow-up, the majority of failures have occurred in PET2-negative patients, indicating the limitations to a PET-adapted approach. To further improve ABVD using precision medicine, we need to further refine predictive markers of failure at the time of Hodgkin lymphoma diagnosis.

Support: National Cancer Institute (NCI), National Clinical Trials Network (NCTN) grants CA180888, CA180819, CA180821, CA180820, CA121947