Abstract P007

Six Cycles of Pembrolizumab (PEM) Induction Followed by Abbreviated AVD: A Phase II Trial of Front-Line PET-adapted Therapy in Early Unfavorable and Advanced-Stage Classic Hodgkin Lymphoma (cHL)

Background: In NU16H08 (NCT03226249), 3 cycles of pembrolizumab (PEM) resulted in complete (37%) or near complete (25%) metabolic responses (CMR) in two thirds of untreated patients (pts) w/classic Hodgkin lymphoma (cHL) and 100% PFS/OS at 5yrs when followed by 4-6 cycles of AVD (ASH 2024, abst# 1664). Based on these findings, we hypothesized that PEM x 6 would yield a higher CMR rate than PEM x 3 and allow for a larger subset of pts to receive an abbreviated course of AVD while preserving efficacy.

Methods: Adult pts > 18yrs w/untreated stage III/IV cHL or stage I/II cHL w/at least 1 NCCN unfavorable risk factor received PEM X 6 followed by 2-6 AVD cycles (+/- PEM consolidation) based on response, stage, and bulk (see SCHEMA). The primary objective was CMR rate after PEM induction by PET scan, w/secondary objectives including efficacy of PEM followed by abbreviated AVD, safety, and PFS/OS. Exploratory objectives included clearance of circulating tumor DNA (ctDNA) as a predictor of response and association w/PET results.

Results: Thirty pts were enrolled: median age of 29 (25–33), 63% female, 50% stage III/IV, 50% w/early-stage unfavorable cHL (9 bulky, 5 B-symptoms, 7 ESR>50). 26 pts completed all PEM x 6; 2 pts went off study after PEM x 1 (1w/grade 3 encephalitis, 1 w/grade 3 headache). Two pts transitioned to AVD after PEM x 3 (1 w/grade 3 pancreatitis, 1 w/progression of disease). PET scans following PEM x 6 (PET #2) were centrally reviewed: CMR (Deauville 1–3) in 27% (n=8; 95% CI: 12%–46%), partial metabolic response (Deauville 4–5) in 60% (n=18). Three pts (10%) had an overall reduction in disease burden w/PEM, but new PET+ sites (progressive disease). Two pts were PET+ (Deauville 4) after AVD x 2 and escalated to BrECADD; 1 pt completed 4 cycles and 1 pt received 2 and transitioned to proton irradiation due to toxicity. CMR was 100% for all pts at end of treatment (n=28) who received AVD following PEM. At a median follow-up of 15.1 months (range 1.0–24.2), no pt has progressed or relapsed; median PFS has not been reached.

Conclusions: Induction with PEM x 6 was not more effective than PEM x 3 in producing CMR. Planned correlative analysis of ctDNA may provide a better assessment of disease status post-PEM, aiding in future de-escalation strategies. Upfront treatment with PEM was associated with significant toxicities, possibly reflecting the robust immune response in this young patient population.