Plasma-based tumor mutational burden (bTMB) as predictor for survival in phase III EAGLE study: Durvalumab (D) ± tremelimumab (T) versus chemotherapy (CT) in recurrent/metastatic head and neck squamous cell carcinoma (R/M HNSCC) after platinum failure.

Weimin Li,Sophie Wildsmith,Jiabu Ye,Han Si,Nassim Morsli,Philip He,Jagdish Shetty,Alejandro Javier Yovine,Nicholas Holoweckyj,Rajiv Raja,Katia Real,Paul Scorer,Todd Hembrough,Jill Walker,Qu Zhang,Ricard Mesia,Robert I. Haddad,Lisa F. Licitra,Robert L. Ferris
DOI: https://doi.org/10.1200/JCO.2020.38.15_suppl.6511
IF: 45.3
2020-05-26
Journal of Clinical Oncology
Abstract:6511 Background: In NSCLC, bTMB assessed from circulating tumor DNA shows promise as a predictive survival biomarker for immunotherapy, but its value in R/M HNSCC is uncertain. We evaluated bTMB as a predictor of survival in R/M HNSCC. Methods: EAGLE (NCT02369874) was a randomized, open-label, phase 3 trial evaluating D (anti-PD-L1), or D+T (anti-CTLA-4), vs CT in R/M HNSCC. Patients (pts) with disease progression after platinum-based CT were randomized (1:1:1) to D (10 mg/kg intravenous [IV] every 2 weeks [Q2W]), D (20 mg/kg IV Q4W) + T (1 mg/kg IV Q4W for up to 4 doses, followed by D at 10 mg/kg Q2W) or CT. bTMB was assessed in pretreatment plasma samples using the Guardant Health OMNI platform. Association of somatic loss of function mutations with OS was assessed. Results: 736 intent-to-treat pts were randomized; 247 were evaluable for bTMB (BEP). bTMB expression was not linked to HPV status, PD-L1 status, age, gender, tumor location, or ECOG PS. Smoking and progression within 6 months on multi-modality CT in localized disease trended with higher bTMB. OS and PFS HRs were significantly improved for D or D+T vs CT in pts with high bTMB (≥16 mut/Mb) vs low (<16 mut/Mb; Table). The benefit of D or D+T vs CT in pts with high bTMB generally improved with increasing cutoff. 74 pts (27 D, 20 D+T, 27 CT pts) were bTMB high. 18-month OS rates were higher for D+T (22%; 95% CI 7%–42%) and D (33%; 95% CI 17%–51%) vs CT (0%; 95% CI 0%–0%) in pts with high bTMB. Pts with mutations in KMT2D , a HNSCC tumor suppressor gene, showed improved OS for D+T vs CT (HR 0.39; 95% CI 0.18–0.85). A trend of improved OS for D+T vs CT (HR 0.19; 95% CI 0.03–1.03) was also seen in pts with ATM mutations. Conclusions: This is the first retrospective analysis of a phase 3 trial to show bTMB may be predictive of outcomes for checkpoint inhibitors in R/M HNSCC. In pts with high bTMB, D or D+T improved OS hazards by at least 60%, vs CT at cutoffs ≥16 mut/Mb. Validation of bTMB as a predictive biomarker is ongoing. Clinical trial information: NCT02369874 . OS HRs (95% CI) for high versus low bTMB. bTMB cutoff (mut/Mb) D vs CT D+T vs CT High Low High Low ≥8 0.63 (0.42–0.94) 1.04 (0.54–2.03) 0.68 (0.46–0.98) 1.34 (0.61–2.92) ≥12 0.65 (0.41–1.05) 0.75 (0.46–1.23) 0.61 (0.39–0.97) 0.98 (0.60–1.61) ≥16 0.39 (0.20–0.75) 0.91 (0.61–1.37) 0.40 (0.20–0.81) 0.92 (0.62–1.36) ≥20 0.40 (0.18–0.88) 0.81 (0.55–1.18) 0.41 (0.17–1.00) 0.84 (0.58–1.22) ≥24 0.26 (0.08–0.81) 0.82 (0.57–1.18) 0.29 (0.09–0.99) 0.83 (0.58–1.17)
oncology
What problem does this paper attempt to address?