SCOTTSDALE, Ariz. -- A test to determine mutant-allele tumor heterogeneity (MATH) accurately distinguished patients with high- and low-risk head and neck cancers, according to a study reported here.
A high MATH score more than doubled the hazard ratio for survival among patients with various types of squamous carcinoma of the head and neck. The relationship between MATH score and outcome held up regardless of a patient's human papillomavirus (HPV) status or other prognostic factors.
Results from the retrospective analysis suggest that MATH could influence treatment decisions for head and neck squamous carcinoma in the not-too-distant future, James W. Rocco, MD, PhD, of Massachusetts General Hospital in Boston, said at the Multidisciplinary Head and Neck Cancer Symposium (MHNCS). The MHNCS is co-sponsored by the American Society for Radiation Oncology, American Society of Clinical Oncology, and the American Head & Neck Society.
'MATH was still significantly related to outcome when HPV status was taken into account,' Rocco said a press briefing. 'A substantial portion of the relation of MATH to outcome does not arise from its relationship to HPV status.'
'MATH is particularly effective in predicting outcome in patients treated with chemoradiation,' he added. 'MATH, in conjunction with HPV status, is likely an effective prognostic marker of patient outcome.'
Investigators plan to continue the evaluation of MATH's prognostic value in head and neck cancer in a prospective study limited to patients with oropharyngeal cancer.
The evolution of next-generation sequencing technology and the wealth of data provided by The Cancer Genome Atlas (TGCA) has given rise to a wide spectrum of hypothesis-testing in oncology. Rocco and colleagues hypothesized that intratumor heterogeneity leads to development of discrete tumor cell populations that evade or resist therapy, allowing the cells to survive, proliferate, and spread.
The initial test of the hypothesis involved 74 tumor specimens from patients with head and neck squamous-cell carcinoma treated at a single center. The results showed that MATH score had a significant association with outcome in head and neck squamous-cell carcinoma.
To broaden the investigation of MATH, researchers evaluated 300 head and neck squamous-cell carcinoma specimens from TGCA. Using the median MATH score derived from the earlier study (32), they compared survival in patients with scores above and below the median value.
The results showed that a high MATH score prior to surgery was associated with significantly worse survival (HR 2.1, 95% CI 1.4-3.2, P=0.0002). Investigators also found that HPV-positive tumors had a significantly lower average MATH score compared with HPV-negative tumors (34.2 versus 39.8, P=0.007).
In a bivariate analysis, a high MATH score and HPV status (positive versus negative) had significant associations with survival (HR 1.85 and HR 0.36, P=0.0002).
Because the 300 specimens included only 35 HPV-positive tumors, the investigators had limited ability to compare outcomes between HPV-positive and HPV-negative tumors according to MATH score. Nonetheless, an analysis stratified by HPV status showed that a high MATH score significantly increased the hazard across multiple subpopulations that comprised favorable and unfavorable characteristics (negative surgical margins, HR 2.2; N0 status, HR 2.0; stage III-IV, HR 1.7).
If the prognostic accuracy of MATH holds up in the 200-patient study, a randomized trial would likely be the next step in the evaluation, Rocco told MedPage Today. Investigators also want to compare MATH scores before and after treatment and to delve deeper into the factors that contribute most to the MATH score.
'My suspicion -- and I have no data to back this up -- is that the MATH score reflects the discrete subpopulations of cells within a tumor, which direct tumor growth and progression,' he said.
The MATH test and underlying hypotheses reflect the growing interest among clinicians and researchers to know more about differences between HPV-positive and HPV-negative patients, said David Raben, MD, of the University of Colorado in Denver.
'We're all dealing with the issues of human papillomavirus-related head and neck cancers, primarily coming from the oropharynx,' said Raben, who monitored the press briefing. 'These cancers are showing some interesting metastatic patterns, they're showing some interesting failure patterns, and we want to learn what is the relevant biology of these types of patients. How do we stratify and select these patients out?
'We're now trying to dig down even deeper. We're culling out within the HPV-positive patients and trying to find which patients have specific mutations that we might be able to attack in a different way. Who are the patients within the HPV-positive population that we might be able to do dose de-intensification?'
HPV infection now accounts for a majority of new head and neck cancers and has transformed the patient population. Today's typical patient is in the 40s and has many years of life ahead. The predominant patient in past years was the older man with a history of smoking and heavy alcohol use.
'How do we manage the HPV-positive patients compared with the typical patient we saw in years gone by -- and still see, but much less frequently?' said Raben.
Rocco, co-investigators, and Raben disclosed no relevant relationships with industry.
Primary source: Multidisciplinary Head and Neck Cancer Symposium Source reference: Rocco JW, et al 'Mutant-allele tumor heterogeneity (MATH) adds to human papillomavirus (HPV) status in predicting outcome in head and neck squamous cell carcinoma' MHNCS 2014; Abstract 7.