Pembrolizumab Impresses in Heavily Pretreated Head and Neck Cancer

Pembrolizumab has shown durable, clinically significant responses in patients with locally advanced head and neck cancer who are refractory to platinum-based chemotherapy and cetuximab, according to results from the single-arm, phase II KEYNOTE-055 study.

Joshua M. Bauml, MD

Pembrolizumab (Keytruda) showed durable, clinically significant responses in patients with locally advanced head and neck cancer who were refractory to platinum-based chemotherapy and cetuximab (Erbitux), according to results from the single-arm, phase II KEYNOTE-055 study published online in the Journal of Clinical Oncology.1

Across the study, the overall response rate (ORR) was 16% (95% CI, 11-23) with the PD-1 inhibitor, including 1 complete response, 27 (16%) partial responses, and 33 (19%) patients with stable disease. Eighty-seven patients (51%) had progressive disease. Compared with baseline, target lesions shrank in half of evaluable patients.

Response rates were also similar regardless of HPV status. The ORR was 16% in HPV-positive patients and 15% in HPV-negative patients.

In an interview with OncLive, lead study author Joshua M. Bauml, MD, said that the prognosis is poor for patients with head and neck cancer who are refractory to both platinum-based chemotherapy and cetuximab. There are no FDA-approved treatments for this patient population and methotrexate, the most commonly used treatment, only has a response rate of around 5%.

“As such, a response rate of 16% is quite exciting,” said Bauml, an assistant professor of Hematology/Oncology and co-deputy director for the Lung/Head and Neck Medical Oncology Clinical Research Program in the Abramson Cancer Center at the University of Pennsylvania. “It is even more exciting when you consider the fact that this is a similar response rate to what was seen in KEYNOTE-012 and CHECKMATE-141, where patients were less heavily pretreated. For a select group of patients, we are seeing that pembrolizumab has activity, regardless of pretreatment status.”

KEYNOTE-012 showed an 18% response rate for pembrolizumab in patients with recurrent or metastatic squamous cell carcinoma of the head and neck; however, no specific prior treatments were required for enrollment in KEYNOTE-012.2 Based on the KEYNOTE-012 data, the FDA approved pembrolizumab in August 2016 as a treatment for patients with recurrent or metastatic head and neck squamous cell carcinoma following progression on a platinum-based chemotherapy.

In CHECKMATE-141, the PD-1 inhibitor nivolumab (Opdivo) improved overall survival (OS) by 30% (HR, 0.70; 97.73% CI, 0.51-0.96; P = 0.01) versus investigator’s choice standard of care in patients with recurrent squamous-cell carcinoma of the head and neck whose disease had progressed within 6 months after platinum-based chemotherapy.3 The ORR in the nivolumab arm was 13.3%. Based on these findings the FDA approved nivolumab in this setting in August 2016.

KEYNOTE-055 enrolled 228 patients between October 24, 2014, and September 23, 2015. Enrollees had been diagnosed with recurrent or metastatic head and neck cancer, and the majority had experienced disease progression within 6 months of treatment with platinum and cetuximab. Three-quarters of patients had received at least two prior lines of therapy in the metastatic setting.

Overall, 171 participants received at least 1 dose of pembrolizumab. The PD-1 inhibitor was administered at 200 mg every 3 weeks. Patients were assessed through imaging every 6 to 9 weeks. Patients received pembrolizumab for a median of 90 days (range, 1-401). The primary endpoints were ORR and safety.

Median progression-free survival (PFS) was 2.1 months (95% CI, 2.1-2.1) in all patients, regardless of HPV status. The overall 6-month PFS rate was 23%, including 25% in HPV-positive patients and 21% in HPV-negative patients.

Median OS was 8 months (95% CI, 6-11) in all patients. The overall 6-month OS rate was 59%, including 72% in HPV-positive patients and 55% in HPV-negative patients. The median duration of response was 8 months (range, 2+ to 12+);. At the April 2016 data cutoff 75% of responses were ongoing and 21% of patients were still receiving pembrolizumab treatment.

“It is particularly notable that patients in KEYNOTE-055 had prolonged responses, given that increasing lines of therapy are generally associated with worse outcomes in oncology,” the researchers wrote. “Pembrolizumab may therefore lead to significant improvements in outcomes for some patients, regardless of prior treatment with platinum and cetuximab.”

Researchers retrospectively evaluated the presence of PD-L1 using an investigational version of the PD-L1 IHC 22C3 pharmDx assay. Most patients (82%) were PD-L1 positive as determined by combined positive score (CPS) ≥1, and 28% of PD-L1-positive patients had a CPS ≥50%.

Response rates were similar regardless of PD-1 expression. Using the 1% CPS cutoff, the ORR was 18% (95% CI, 12-25) in PD-L1—positive patients and 12% (95% CI, 2-30) in PD-L1–negative patients. When the CPS was analyzed to 50% based on raw scores, the ORR was 27% (95% CI, 15-42) in patients with CPS ≥50% and 13% (95% CI, 7-20) in patients with CPS <50%. The 1 complete response recorded in the study occurred in a patient with a CPS of ≥50%.

Six-month PFS rates in PD-L1—positive patients were 24% (CPS ≥1%) and 31% (CPS ≥50%) compared with 20% for patients with either CPS <1% or CPS <50%. Six-month OS rates were 59% (CPS ≥1%) and 60% (CPS ≥50%) for PD-L1–positive patients compared with 56% (CPS <1%) and 58% (CPS <50%) for PD-L1–negative patients.

“Six- and 12-month progression-free and overall survival rates were relatively similar between PD-1 negative and positive patients, highlighting the clear limitation with this selection marker,” Amanda Psyrri, MD, PhD, chief of Medical Oncology and assistant professor of Medicine at National Kapodistrian University of Athens Medical School, said in an accompanying podcast. “Excluding such patients from potentially life-prolonging treatments on the basis of PD-1 expression remains a clinical and ethical dilemma.”

Bauml acknowledged that physicians need a better biomarker for this disease.

“The problem is that PD-L1, given its heterogeneous and dynamic nature, is simply not an ideal biomarker,” he said. “Current research efforts are underway to identify better ones, including mutational load and gene expression profiling.”

Two-thirds of patients reported treatment-related adverse events (AEs) of any grade. The most common all-grade AEs included fatigue (18%), hypothyroidism (9%), nausea (6%), AST increase (6%), and diarrhea (6%).


  1. Bauml JM, Seiwert TY, Pfister DG, et al. Pembrolizumab for platinum- and cetuximab-refractory head and neck cancer: results from a single-arm, phase II study. J Clin Oncol. doi: 10.1200/JCO.2016.70.1524.
  2. Seiwert TY, Burtness B, Mehra R, et al. Safety and clinical activity of pembrolizumab for treatment of recurrent or metastatic squamous cell carcinoma of the head and neck (KEYNOTE-012): an open-label, multicentre, phase 1b trial. Lancet Oncol. 2016;17:956-965.
  3. Ferris RL, Blumenschein G Jr, Fayette J, et al. Nivolumab for recurrent squamous-cell carcinoma of the head and neck. N Engl J Med. 2016;375:1856-1867.

Twenty-six patients (15%) experienced a grade 3 or higher AE. The only immune-mediated AEs of any grade or treatment attribution reported in 2% or more of patients were hypothyroidism (16%), pneumonitis (4%), and hyperthyroidism (2%). Seven patients (4%) left the study due to treatment-related AEs, and 1 patient died of treatment-related pneumonitis.