Following the approval of 2 immunotherapy agents, pembrolizumab and nivolumab for the treatment of patients with head and neck cancer over the last 6 months, immunotherapy is making its mark on the treatment paradigm for HNC.
Tanguy Seiwert, MD
Following the approval of 2 immunotherapy agents, pembrolizumab (Keytruda) and nivolumab (Opdivo) for the treatment of patients with head and neck cancer (HNC) over the last 6 months, immunotherapy is making its mark on the treatment paradigm for HNC.
Due to the responses seen with these 2 agents, immunotherapies are being investigated further in the treatment of HNC.
“Immunotherapy is a very potent treatment for some patients. In a way it shows you that we’re probably just scratching the surface with [immunotherapy treatment for HNC],” Tanguy Seiwert, MD, said during a presentation at the 1st Annual International Congress on Immunotherapies in CancerTM, hosted by the Physicians' Education Resource (PER).
Findings from the KEYNOTE-012 trial led to the approval of pembrolizumab in patients with recurrent head and neck squamous cell carcinoma (HNSCC). The overall response rate was 18% with only 1 patient experiencing a complete response.1However, about 50% of patients, both HPV-positive and HPV-negative, experienced a decrease in their target lesions.
“I would like to point out that response is a terrible, terrible outcome measure for immunotherapy. In the end, what we really care about with immunotherapy is overall survival [OS],” commented Seiwert, associate program director of the Head and Neck Cancer Program, and assistant professor of medicine, The University of Chicago Medicine. “Many patients have prolonged stable disease and that likely contributes signicantly to the strong OS signal that we oftentimes see.”
The phase III CheckMate 141 trial, which Seiwert said was “arguably the most important study in the field,” showed a difference in OS that is more revealing of outcome measures in immunotherapy. CheckMate 141 investigated nivolumab monotherapy in the second-line setting versus investigator’s choice of chemotherapy in patients with recurrent or metastatic HNSCC and demonstrated a median OS of 7.5 (95% CI, 5.5-9.1) versus 5.1 months (95% CI, 4.0-6.0) with standard therapy (P = .0101).2The 1-year OS rate was 36% with nivolumab versus 16.6% with standard therapy. Alternatively, the response rate was 13.3% with nivolumab compared with 5.8% in the standard therapy arm.
“The response rate wasn’t that impressive, but the overall survival data are stunning. And that’s again an example of how wonderfully these drugs work,” Seiwert said.
Following the responses seen in these 2 studies of PD-1 inhibitors, immunotherapy agents are being considered in the frontline, including in combination regimens, which Seiwert believes are promising. One such combination is durvalumab (MEDI4736), a PD-L1 inhibitor, and tremelimumab, an antiCTLA-4 agent, which was compared against durvalumab or the EXTREME trial regimen of cetuximab (Erbitux) and platinum-based chemotherapy in the phase III KESTREL trial.
Other first-line combination studies of interest in HNC include the KEYNOTE-048 study, which is looking at pembrolizumab and chemotherapy versus pembrolizumab monotherapy or the EXTREME regimen (NCT02358031); the CheckMate 651 study of ipilimumab (Yervoy) and nivolumab versus EXTREME (NCT02741570); and the CheckMate 714 study exploring ipilimumab and nivolumab versus nivolumab as a single agent (NCT02823574).
Preliminary results looking at the combination of lirilumab, an anti-KIR agent, and nivolumab in a phase I/II study were presented at the 2016 SITC Annual Meeting. The combination showed an objective response rate (ORR) of 24.1% versus an ORR of 13.3% seen with nivolumab monotherapy in the CheckMate 141 trial.2,3The OS at 1 year was 60% with the combination compared with 36% for nivolumab monotherapy. Among patients with PD-L1 expression in the tumor cells of ≥50%, the ORR was 57.1% with lirilumab and nivolumab versus 36.8% with nivolumab alone. Seiwert hypothesized that KIR was among a number of targets, also including CTLA-4, IDO, and OX40, that are more active in hot tumors.
In discussing which patients should receive immunotherapy treatment, Seiwert looked to various biomarkers currently under investigation for their predictive or prognostic association to immu- notherapy response. The KEYNOTE-024 trial looking at pembrolizumab versus chemotherapy in patients with nonsmall cell lung cancer changed the eld of PD-L1 testing, according to Seiwert. There was a significant difference in progression-free survival (PFS) and OS rates noted in patients with PD-L1 expression of ≥50% on the tumor cells.4 This can be translated into HNC, and notably, the KEYNOTE-048 trial of patients with recurrent or metastatic HNSCC will include a PD-L1–positive subgroup as part of its investigation.
“While I do have my doubts about how perfect PD-L1 testing is, I do believe it plays a role for enrichment,” Seiwert commented.
An interferon-gamma (IFN-γ) signature showed significant association with overall response (P = .005) and PFS (P <.001) in an analysis of PD-L1positive patients from the KEYNOTE-012 trial.5There was also a very high negative predictive value for patients with nonIFN-γ–inflamed tumors who did not receive benefit from pembrolizumab, which would prove useful in identifying which patients should not receive anti–PD-1 therapy. Of great interest are the patients with inflamed tumors who do not benefit from the treatment. Perhaps they could be converted into responders through combination therapies, Seiwert pondered.
“None of these biomarkers are perfect. I think we need a bit more time to fully understand this, but these are biomarkers that are potentially helpful and might outperform PD-L1 testing in the near future,” Seiwert said.