KRAS-Variant HNSCC Outcomes May Improve With Cetuximab

Joanne B. Weidhaas, MD, PhD

Patients with head and neck cancer whose disease is associated with KRAS variant had significantly better progression-free survival (PFS) and overall survival (OS) when treated with the monoclonal antibody cetuximab (Erbitux), according to findings of a retrospective analysis of a randomized trial.

During the first year of follow-up, cetuximab-treated patients with KRAS variant had almost a 70% reduction in the hazard for death or progression as compared with a KRAS-variant subgroup whose treatment did not include cetuximab. The hazard for OS during the first 2 years was improved by 80% in patients with KRAS-variant disease treated with cetuximab versus those who did not receive the agent.

The PFS benefit dropped out after the first year and the OS benefit after the second year. The reason for the rapid loss of benefit remained unclear but could reflect a short term of exposure to the drug (8 weeks), reported Joanne B. Weidhaas, MD, PhD, during the 2016 American Society for Radiation Oncology (ASTRO) Annual Meeting in Boston, Massachusetts.

The cetuximab benefit “may represent inherent resistance to cisplatin-radiation therapy or a weak immune response, overcome by cetuximab,” said Weidhaas, a professor of Radiation Oncology at the University of California, Los Angeles. “The KRAS variant and p16 status are both important in head and neck squamous cell carcinoma (HNSCC) in predictive response. Elevated TGFβ1 levels could perhaps explain the immunosuppression and elevated toxicity that we observed.”

The findings came from a new analysis of the RTOG (now NRG) 0522 trial—a phase III randomized comparison chemoradiation with or without cetuximab for patients with HNSCC. Identification of radiation biomarkers of response has proven challenging, and availability of tissue from the RTOG 0522 patient population afforded an opportunity to continue the exploration.

The trial had a primary endpoint of PFS and secondary endpoints that included OS. The overall trial results were negative, as the addition of cetuximab failed to improve PFS or OS.

“Clearly, there were some people who responded,” said Weidhaas. “We wanted to test for the KRAS variant in this population.”

The KRAS variant is a germline mutation in a microRNA-binding site in KRAS. The variation disrupts regulation of KRAS, occurs in a variety of tumor types, and is present in as many as 30% of patients with cancer patients, said Weidhaas. Previous studies had demonstrated an association between the variant and response to cancer therapy, including cetuximab.

In the current analysis, investigators evaluated the relationship of KRAS-variant status to cetuximab response, p16 status, and plasma levels of the immune protein TGFβ1.

The trial included 891 evaluable patients, 413 of whom had samples available for KRAS-variant testing, and 376 patients had plasma samples available for TGFβ1 measurement. KRAS assessment showed that 17% of the patients had the variant, evenly distributed in the treatment groups and not enriched in the p16-positive subgroup.

The results showed that patients with the KRAS variant treated with cetuximab had a hazard of 0.31 for progression or death during the first year, as compared with patients who had the variant and were not treated with the antibody (95% CI, 0.10-0.94; P =.04). After year 1, however, the hazard ratio for progression or death increased to 1.76 for cetuximab-treated patients.

The OS analysis showed a similar pattern of benefit during the first 2 years of follow-up: a hazard ratio of 0.19 for cetuximab-treated patients with the KRAS variant (95% CI, 0.04-0.86; P =.03), followed thereafter by a turn in the opposite direction, resulting in a hazard ratio of 2.34 (95% CI, 0.58-9.41; P =.23).

In a multivariate analysis of the PFS data, a favorable effect of cetuximab on distant metastasis appeared to drive the benefit (HR, 0.45), as compared with locoregional failure (HR, 0.84).

Searching for an explanation for the survival benefit, investigators explored cetuximab’s effect on immunity. Specifically, they examined interaction among KRAS variant, cetuximab, and p16, as p16-positive tumors are immunogenic at baseline. They also studied TGFβ1, a known marker of immunity found in the circulation.

The results demonstrated a significant three-way interaction (P = .02). Patients who had the KRAS variant but were p16 positive had much worse survival than with conventional chemotherapy and radiation therapy (HR, 2.48; P =.19). The addition of cetuximab appeared to reverse the unfavorable prognosis in that subset of patients (HR, 0.22; P = 0.14), said Weidhaas.

High levels of TGFβ1 are known to inhibit radiation-induced antitumor immunity, and cetuximab has been shown to overcome TGFβ1 inhibition of natural killer cells. Data from RTOG 0522 showed that KRAS-variant patients had elevated levels of TGFβ1. When treated without cetuximab, KRAS-variant patients with elevated TGFβ1 appeared to have an increased incidence of grade 3/4 toxicity (OR, 2.31), although the interaction between KRAS and TGFβ1 did not achieve statistical significance.

Reference

1. Weidhaas JB, Harris J, Schaue D, et al. The KRAS-variant is a biomarker of cetuximab response, potentially through altered immunity in head and neck cancer: re-analysis of NRG Oncology/RTOG 0522. Presented at: 2016 ASTRO Annual Meeting; Boston, Massachusetts, September 25-28, 2016.