Skip to main content
Premium Trial:

Request an Annual Quote

Tumor Mutational Burden in Blood May Predict Immunotherapy Response in Head, Neck Cancer

NEW YORK – An analysis of retrospective data from the Phase III EAGLE study suggests that tumor mutational burden (TMB), as assessed through cell-free DNA in patients' blood plasma, may predict which head and neck squamous cell carcinoma patients are likely to respond to checkpoint inhibitors.

The initial EAGLE study pitted the PD-L1 inhibitor durvalumab (AstraZeneca's Imfinzi), with or without the addition of the investigational CTLA4-inhibitor tremelimumab, against chemotherapy in patients with recurrent or metastatic head and neck squamous cell carcinoma who experienced disease progression after platinum-based chemotherapy. The EAGLE study did not meet its primary endpoint of overall survival, due to a higher early mortality rate among patients treated with durvalumab or durvalumab plus tremelimumab than among patients treated with chemotherapy.

Weimin Li of AstraZeneca presented results from the subsequent retrospective analysis of the data during the American Society of Clinical Oncology's virtual annual meeting this past week on behalf of the EAGLE investigators. The analysis evaluated the ability of TMB, detected from circulating tumor DNA (ctDNA), to predict survival after treatment with durvalumab or a combination of durvalumab and tremelimumab.

The investigators used Guardant Health's GuardantOmni assay to evaluate blood TMB, for which they set the cutoff for "high" TMB as greater than or equal to 16 mutations per megabase. Out of the 736 patients who were randomized in the original EAGLE study, 247 were evaluable for blood-based TMB, and 74 were found to be TMB high.

At the 18-month mark, overall survival rates were significantly higher among TMB-high patients who were treated with durvalumab and durvalumab-tremelimumab than those treated with chemotherapy. Specifically, 33 percent of TMB-high patients treated with durvalumab monotherapy and 22 percent of patients treated with the combination were still alive at 18 months. In contrast, none of the patients with high TMB who received chemotherapy alone were alive 18 months later.

Additionally, when looking at median overall survival by high and low TMB levels, there was a distinct advantage among high-TMB patients receiving single or combination immunotherapy over chemotherapy, but not among low-TMB patients. The median overall survival in the high-TMB group was 8.1 months with durvalumab, 7.6 months with durvalumab-tremelimumab, and 4 months with chemotherapy. For the low-TMB group, these numbers, in contrast, were 6.2 months, 6.8 months, and 8.6 months, respectively.

In a discussion of this data, Julie Bauman of the Arizona Cancer Center noted that the graphical presentation of the survival analysis shows that patients with high TMB sustained an overall survival benefit with the immunotherapy agents as the months on treatment progressed, whereas in the low-TMB group, patients in the chemotherapy arm appeared to be doing far better in the first year of treatment compared to those receiving immunotherapy.

Notably, Bauman pointed out that patients in the EAGLE study's initial intent-to-treat population all had tumors that expressed high levels of PD-L1. "Thus, high blood tumor mutation burden was predictive [of overall survival], even in a high PD-L1 setting," she said. Also of note, high blood TMB was independent of patients' HPV status.

The predictive ability of blood TMB will need further validation, Li said, acknowledging that this was a retrospective, exploratory analysis. Prospective investigations will be warranted to confirm blood TMB as a predictive biomarker for immunotherapy in recurrent, metastatic head and neck squamous cell carcinoma.

Nonetheless, this study demonstrated the potential of assessing a TMB biomarker through blood-based ctDNA as opposed to solid tissue samples, according to experts. "Although tumor TMB is a robust predictive biomarker, utilization is challenged by tissue availability, as well as the theoretical concern that tumor heterogeneity is missed by a single-site biopsy," Bauman said in her discussion. "Blood TMB may overcome these limitations."