Abstract P116

Extended Follow-up from an Intergroup Randomized Phase II Study of the Combinations of Ipilimumab, Nivolumab and Brentuximab Vedotin in Patients with Relapsed/Refractory Classic Hodgkin Lymphoma: A trial of the ECOG-ACRIN Research Group (E4412)

Background: The Phase 1/2 ECOG-ACRIN sponsored intergroup trial E4412 (NCT01896999) investigated brentuximab vedotin (BV) combined with the checkpoint inhibitors nivolumab (N) and ipilimumab (I) in patients with relapsed or refractory Hodgkin lymphoma (R/R HL); here we present the Phase 2 efficacy and safety data for the combined adult and pediatric patients with extended follow-up for the adult population.

Methods:
R/R HL patients were equally randomized between the doublet of BV/N and the triplet of BV/N/I. With 140 eligible & treated patients, there was 87% power to detect a 20% increase in complete response (CR) rate from 40% expected in BV/N to 60% in BV/N/I using a Fisher’s exact test with a 15% alpha (one-sided). Results: A total of 147 patients were randomized, 133 adults and 14 pediatric. Sixteen (12%) adult patients and no pediatric patients had prior BV.

Response: One hundred thirty-two patients are evaluable for response: 68 BV/N and 64 BV/N/I, with a CR rate of 64.7% (44 of 68) compared to 70.3% (45 of 64) (p=0.287); ORR was 89% in both arms. The median (Q1, Q3) survival follow-up is 38.0 months (32.6, 48.1). We compared PFS in response eligible adult patients by autologous stem cell transplant (SCT) status in a post-hoc comparison. Fifty-four of 118 (46%) patients received SCT; there is no difference in 36-month PFS: 93.0% for BV/N and 94.7% BV/N/I (p=0.1472). Baseline characteristics were balanced for the 64 adult patients that did not receive SCT (30 BV/N and 34 BV/N/I) in terms of prior BV and prior SCT. For patients who did not have SCT the 36-month PFS for BV/N was 48.1 months compared to 68.3 months for BV/N/I (p=0.0385).

Safety: 65 (BV/N) and 61 (BV/N/I) patients in the adult safety cohort received at least 1 cycle of therapy. The rate of treatment-related grade 3+ toxicities, excluding rash, is similar between both arms, 38.5% (25/65) BV/N and 39.3% (24/61) BV/N/I; there was a higher amount of grade 3 rash in BV/N/I 24.6% (15/61) vs 9.2% (6/65). There was no grade 4 rash. There were no grade 5 toxicities in either arm.

Conclusion: The experimental arm of BV/N/I did not significantly improve CR rate and led to a higher incidence of grade 3 rash. In a post-hoc analysis for patients receiving SCT 36-month PFS was > 90% in both arms, but for patients who did not undergo SCT there was a significant improvement 36-month PFS with BV/N/I.