Intratumor Heterogeneity of Head and Neck Cancers May Have Therapeutic Implications

Get Permission

A RETROSPECTIVE STUDY of The Cancer Genome Atlas may have therapy-specific implications for patients with head and neck squamous cell carcinoma, according to data presented at the 2018 Multidisciplinary Head and Neck Cancers Symposium.1 This first analysis of the relationship between intratumor genetic heterogeneity and overall survival showed a marked difference in outcomes depending on the therapy received. 

The results also suggest that patients with high–mutant-allele tumor heterogeneity might benefit from radiation, even when clinical considerations suggest adjuvant therapy can be omitted. More provocatively, the authors noted, patients with high–mutant-allele tumor heterogeneity might not benefit from the addition of chemotherapy to radiotherapy and may thus be spared the morbidity of complications from combination therapy. 

James W. Rocco, MD, PhD

James W. Rocco, MD, PhD

“Because tumor genetic diversity is associated with treatment resistance and consequently a poor prognosis, a quantitative measure of intratumor heterogeneity would likely provide clinically significant information that could predict outcome and potentially direct therapy,” said James W. Rocco, MD, PhD, Chairman, Department of Otolaryngology–Head and Neck Surgery; Professor and John and Mary Alford Chair of Head and Neck Surgery, The Ohio State University Wexner Medical Center. “Although based on retrospective analysis and statistical control of other variables, this report has provocative implications that deserve prospective study.” 

Study Details 

AS DR. ROCCO reported, high-resolution genome-wide studies have revealed significant genetic heterogeneity within individual tumors. Moreover, an initial study of 305 cases of head and neck squamous cell carcinoma from The Cancer Genome Atlas showed that high–mutant-allele tumor heterogeneity was strongly correlated to shorter overall survival,2 but data were insufficient to determine whether relation to overall survival depended on the type of therapy. Since the completion of that study, however, more complete clinical data now allow examination of this issue, revealed Dr. Rocco. 

Edmund Mroz, PhD

Edmund Mroz, PhD

For this analysis, Dr. Rocco, along with lead author Edmund Mroz, PhD, and colleagues used clinical and whole-exome sequencing data on 528 head and neck squamous cell carcinoma cases in The Cancer Genome Atlas, obtained from the National Cancer Institute Genome Data Center and the Broad Institute. Clinical data were reviewed to determine whether initial therapy included radiation therapy or chemotherapy as primary or adjuvant therapy. Intratumor genetic heterogeneity was assessed by a modification of the mutant-allele tumor heterogeneity measure, which improved handling of differing tumor-cell fractions among samples. Finally, Cox multiple regression of overall survival included patient age, year of diagnosis, smoking history, human papillomavirus status, clinicopathologic data, type of therapy, and mutant-allele tumor heterogeneity. 

Of the 528 cases analyzed, 438 had had sufficient data for Cox multiple regression of overall survival. As Dr. Rocco reported, the interaction of mutant-allele tumor heterogeneity with therapy was significantly related to overall survival (P = .016). Moreover, of the variables analyzed, only high–mutant-allele tumor heterogeneity (hazard ratio [HR] = 3.43; P = .0002) and the presence of extracapsular spread had high prognostic implications (HR = 3.65; P < .0001). With other clinical variables accounted for, the longest overall survival was seen for patients with tumors that had low– mutant-allele tumor heterogeneity who received chemoradiation (baseline for HR), whereas the shortest overall survival was observed in patients with tumors that had high–mutant-allele tumor heterogeneity not receiving adjuvant therapy (HR = 6.6; confidence interval, 2.8–20). The differences in response to therapy between tumors with high– and low–mutant-allele tumor heterogeneity, however, were most striking to the investigators. 

“In advanced-stage disease of the oral cavity, we saw low survival for patients with tumors that had high–mutant-allele tumor heterogeneity and received surgery alone. Although there were improvements with the addition of chemoradiation, the best survival came from adding radiation alone to surgery,” said Dr. Rocco. “With tumors that have low– mutant-allele tumor heterogeneity, on the other hand, the addition of chemoradiation to surgery significantly increased survival probability over the addition of radiation alone.” 

As Dr. Rocco explained, these data demonstrated completely different results for tumors that had high– vs low–mutant-allele tumor heterogeneity treated with surgery alone (HR = 3.4; P = .0002), suggesting adjuvant therapies are needed for tumors with high–mutant-allele tumor heterogeneity in advanced stage. Intriguingly, added Dr. Rocco, this difference is significantly reduced when radiation is added to surgery (HR = 1.2; P = .57). 

“What’s more provocative, however, is this difference doesn’t go away when chemoradiation is added (HR = 3.2; P = .009),” said Dr. Rocco, who noted these differences remained, even with the removal of all specimens with human papillomavirus from the analysis. “In the absence of high-risk pathologic features, tumors with high–mutant-allele tumor heterogeneity should be treated with radiation, and tumors with low–mutant-allele tumor heterogeneity should be observed in the early stage or treated with chemoradiation in the advanced stage,” Dr. Rocco observed. 

Further Study Needed 

ACCORDING TO THE AUTHORS, intratumor heterogeneity should be evaluated in controlled trials that compare observation, adjuvant radiation, and chemoradiation after surgery for head and neck cancer. 

“We struggle to know when chemotherapy works well with radiation and when it doesn’t, because we’re lumping tumors with high–mutant-allele tumor heterogeneity and low–mutant-allele tumor heterogeneity together in clinical trials and averaging the results,” Dr. Rocco concluded. “This approach might be a way to make both sides of the equation look better.” ■

DISCLOSURE: Dr. Rocco reported no conflicts of interest. 


1. Mroz EA, et al: 2018 Multidisciplinary Head and Neck Cancers Symposium. Abstract 4. Presented February 15, 2018. 

2. Mroz EA, et al: PLoS Med 12:e1001786, 2015. 

Related Articles

Expert Point of View: Young Jun Kim, MD, PhD

YOUNG JUN KIM, MD, PhD, Co-Leader of the Translational Research and Interventional Oncology Research Program at Vanderbilt-Ingram Cancer Center, in Nashville, called the research “very important work.” 

“This is very close to clinical relevance because no special test is required right now,” said...