Clearance of circulating tumor DNA (ctDNA) during treatment was associated with improved long-term overall survival (OS) among patients with human papillomavirus (HPV)–negative locally advanced head and neck squamous cell carcinoma (HNSCC) who received pembrolizumab (Keytruda) plus concurrent chemoradiotherapy, according to an exploratory ctDNA analysis of the phase 3 KEYNOTE-412 trial (NCT03040999) presented at the
In the HPV-negative subgroup, ctDNA clearance at cycle 3, day 1 of treatment occurred in 76.8% of evaluable patients in the pembrolizumab arm (n = 155) vs 66.3% of those in the placebo arm (n = 166).
Among patients who received pembrolizumab plus chemoradiotherapy, the 5-year OS rate was 72.9% for those who cleared ctDNA (n = 119) vs 38.9% for those who did not (n = 36; P < .001). Moreover, the 5-year event-free survival (EFS) rates with these respective groups were 63.1% vs 38.2%, respectively (P = .001). In the placebo arm 5-year EFS rates for patients who achieved ctDNA clearance (n = 110) vs those who did not (n = 56) was associated were 48.4% vs 35.5% (P = .011), respectively, with 5-year OS rates being 58.6% vs 48.2% (P = .075).
In the pembrolizumab arm of the HPV-negative subgroup, the 5-year OS rate was 70.5% for patients with a maximum somatic allele frequency (MSAF) below the median (n = 106) compared with 55.8% for those with MSAF at or above the median (n = 95; P = .009). The 5-year EFS rates were 58.9% vs 51.2% (P = .086) each of the respective groups.
Regarding the placebo arm, the 5-year EFS rates were 48.1% (n = 96) vs 38.1% (n = 108; P = .002) for patients below and above the MSAF median, respectively, and the 5-year OS rates were 58.3% vs 49.5% (P = .01).
Overall, investigators noted baseline ctDNA levels as prognostic for both EFS and OS with pembrolizumab benefits seen across ctDNA levels in HPV-negative HNSCC. For those with ctDNA clearance at cycle 3, day 1 and HPV-negative HNSCC, the clearance was predictive for long-term survival with pembrolizumab plus chemoradiotherapy regardless of other common prognostic markers.
"However, [these] results should be interpreted with caution as the HPV-positive group sample size was small,” said lead study author Jean-Pascal Machiels, MD, PhD, in a presentation of the data.
Machiels is the head of Adult and Pediatric Oncology at Institut Roi Albert II, Cliniques universitaires Saint-Luc, UCLouvain in Brusselsm Belgium.
The trial enrolled patients with evaluable tumor burden per RECIST 1.1 criteria and an ECOG performance status of 0 or 1, in addition to those who had not received prior surgical debulking for the head and neck.2
If patients received prior anti–PD-1, anti–PD-L1, anti–PD-L2 therapy, received a live vaccine within 30 days of their first dose, had active hepatitis B or C or had cancer outside oropharynx, larynx or oral cavity they were not enrolled in the trial.
KEYNOTE-412 randomly assigned 804 patients with unresected, high-risk locally advanced HNSCC to pembrolizumab or placebo plus chemotherapy. Patients received pembrolizumab at 200 mg intravenously every 3 weeks for 3 doses with 6 to 7 weeks of chemoradiotherapy, followed by 14 maintenance doses, or placebo on the same schedule. Randomization was stratified by radiotherapy regimen, tumor site/p16 status, and disease stage (III vs IV). EFS was the primary end point of the trial, compared with OS, which was the key secondary end point, and safety was the other secondary end point.
Approximately 95% of patients had detectable ctDNA at baseline regardless of HPV status or treatment arm, and baseline ctDNA levels demonstrated a moderate positive correlation with target tumor lesion size in HPV-negative (ρ = 0.494; n = 405) and HPV-positive (ρ = 0.341; n = 166) disease.
In the HPV-negative subgroup, EFS and OS numerically favored pembrolizumab plus chemoradiotherapy over placebo plus chemoradiotherapy at both baseline ctDNA levels. Among patients with MSAF at or above the median in the pembrolizumab arm compared with the placebo arm, the hazard ratio (HR) for EFS was 0.75 (95% CI, 0.52-1.07) and the HR for OS was 0.85 (95% CI, 0.57-1.25). Among those with MSAF below the median, the respective HRs were 0.76 (95% CI, 0.51-1.13) and 0.73 (95% CI, 0.46-1.14).
Notably, association between ctDNA clearance at cycle 3, day 1 and outcomes in HPV-negative disease was independent of ECOG performance status and other prognostic markers, including tumor size, disease stage, PD-L1 combined positive score, T-cell–inflamed gene expression profile, and tumor mutational burden.
In the HPV-positive subgroup, baseline ctDNA levels were not prognostic for EFS in the pembrolizumab or placebo arm, or for OS in either arm, and outcomes were similar between arms regardless of ctDNA level.
ctDNA clearance at cycle 3, day 1 occurred in 81.5% of HPV-positive patients in the pembrolizumab arm (n = 65) vs 64.6% of those in the placebo arm (n = 65). In the placebo arm, patients who cleared ctDNA (n = 42) had a 5-year EFS rate of 70.2% vs 49.0% for those who did not (n = 23; P = .028). Five-year OS rates for these respective groups were of 85.5% vs 65.2% (P = .013).
In the pembrolizumab arm, the 5-year EFS rates were 67.6% for patients who cleared ctDNA (n = 53) vs 41.7% for those who did not (n = 12; P = .095), and the 5-year OS rates were 78.8% vs 83.3% (P = .619). The investigators cautioned that these results should be interpreted carefully because of the small sample size.
Disclosures: Machiels reported receiving funding (to institution) from ALX Oncology, ANAVEON, Astellas, Bayer, BMS, Boehringer, Cue Biopharma, CureVac, eTheRNA, F-Star Therapeutics, Genmab, GSK, Incyte, Ipsen, iTeos, Merck Serono, Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA (MSD), Merus, Nektar, Novartis, Pfizer, Roche, and Seagen; serving on advisory boards (to institution) for ALX Oncology, ANAVEON, Astellas, Bayer, BMS, Boehringer, Cue Biopharma, CureVac, eTheRNA, F-Star Therapeutics, Genmab, GSK, Incyte, Ipsen, iTeos, Merck Serono, Merus, MSD, Nektar, Novartis, Pfizer, Roche, and Seagen; and receiving travel expenses (to institution) from Amgen, BMS, Gilead, Janssen, MSD, Pfizer, and Sanofi.
