Few treatment options exist for non-curable recurrent or metastatic squamous cell carcinoma of the head and neck, and even fewer treatment options offer a survival benefit. The open-label, multicentre, phase 1b trial KEYNOTE-012 trial has now presented data on safety and efficacy from patients with head and neck cancer. In KEYNOTE-012, pembrolizumab was administrated every 2 weeks to patients with recurrent or metastatic squamous cell carcinoma of the head and neck that were PD-L1 positive (> 1%). Patients were heavily pre-treated with 70% of patients previously receiving two or more lines of treatment. 60 patients with PD-L1 positive squamous cell carcinoma of the head and neck were enrolled and treated.
Safety data showed that pembrolizumab was well tolerated with manageable toxicities and reported grade 1 - 2 adverse events in 38 (63%) patients, most commonly pruritus, fatigue or rash. Grade 3 - 5 adverse events occurred in 17% of patients. No treatment related deaths, or cases of pneumonitis and colitis requiring intervention were seen. Overall response was reported in 18% of patients, with a 25% response in HPV-positive patients and 14% in HPV-negative patients. The duration of response was reported at 12.2 months. Median progression-free survival (PFS) was 2 months and median overall survival (OS) was 13 months. Median overall survival was not reached for HPV-positive patients and was 8 months in HPV-negative patients. A gene expression composite score consisting of six interferon-γ-regulated genes provided a positive predictive value of 40% and a negative predictive value of 95%. An overall response rate of 18% is consistent with data from other trials investigating efficacy of immunotherapy in solid tumours 1.
These results are quite promising for the use of pembrolizumab in head and neck cancer. A median overall survival of 13 months in the recurrent or metastatic setting for squamous cell carcinoma of the head and neck, in a heavily pre-treated group of patients, must be considered substantial. But several issues should be addressed in this study and considered in future designs. Median progression-free survival was 2 months, corresponding to the time at the first evaluation scan. All though a 2 months PFS may be considered poor, it still translated into a median overall survival of 13 months. This could be due to a heavily selected group consisting of patients in performance status 0-1 and a large part of HPV positive patients (38%), recognised as having a more favourable prognosis than HPV negative patients.
A comparison between trials should always be done carefully. But a median overall survival of 13 months in the recurrent or metastatic setting for squamous cell carcinoma of the head and neck is comparable to first-line treatment in chemo-naive patients. In the EXTREME trial 2, the addition of cetuximab to fluorouracil and platin increased PFS from 3.3 to 5.6 months and increased the response rate from 20% to 36%. It also significantly improved median overall survival from 7.4 months to 10.1 months making cetuximab, platin and fluorouracil a standard choice for first line therapy. In the KEYNOTE-012 study, a PFS of only 2 months and a lower overall response of 18% provided an OS of 13 months.
When keeping in mind that 70% of patients in the KEYNOTE-012 had received 2 or more lines of systemic therapy, it suggests, a median overall survival of 13 months driven by relative few responders (18%). This is also evident from the long duration of response, reported at 12.2 months. Furthermore, the median overall survival was 8 months in HPV negative patients and was not reached in the HPV positive population. This also suggests a positive impact on the total median overall survival from the relative high proportion of HPV-positive tumours. So one may conclude that pembrolizumab for recurrent or metastatic squamous cell carcinoma of the head and neck is promising, but for the few. It is therefore crucial to be able to select patients with the greatest chance of responding.
The authors provide an interesting option, with a predictive algorithm consisting of six interferon-γ-regulated genes. Using this algorithm, a 40% positive predictive value was reported. More importantly though, with a negative predictive value of 95%, the algorithm could provide clinicians with a valuable tool that could assist in selecting patients who are not candidates to receive pembrolizumab. In conclusion, the KEYNOTE-012 provides valuable data, both for the future planning and considerations of trials and for adding treatment options to a group of patients with poor prognosis and limited efficient options. Trials investigating pembrolizumab, both as first-line treatment versus platin, fluorouracil and cetuximab as well as concomitant with radiotherapy in the intent to cure are warranted. Furthermore, the six gene predictive algorithm may provide a valuable tool in future treatment planning and should be verified in upcoming trials.
- Topalian SL, Hodi FS, Brahmer JR et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N. Engl. J. Med. 2012; 366(26):2443–54.
- Vermorken JB, Mesia R, Rivera F et al. Platinum-based chemotherapy plus cetuximab in head and neck cancer. N. Engl. J. Med. 2008; 359(11):1116–27.
- Is immunotherapy for the few or the many?
- How do we select patients for treatment with pembrolizumab and other immunotherapies?
The author has no actual, potential, real or apparent interest to declare and has no involvement that might raise the question of bias in the work reported or in the conclusions, implications, or opinions stated.