This article is more than 5 years old. Information may no longer be current.
Biomarker may help better predict outcomes in HPV-positive head, neck cancer
Click Here to Manage Email Alerts
Researchers have identified a gene panel that could lead to less intensive treatment for certain patients with HPV-positive oropharyngeal squamous cell carcinoma, according to study results.
“This is very exciting because this may lead us to new ways to target HPV-driven tumors — or, more importantly, the HPV tumors associated with poor outcomes,” Curtis R. Pickering, PhD, assistant professor in the department of head and neck surgery at The University of Texas MD Anderson Cancer Center, told HemOnc Today.
May patients with HPV-oropharyngeal squamous cell carcinoma achieve long-term survival; however, they often experience radiotherapy-related morbidity. De-escalation protocols have been developed, but some patients do not respond and experience poor outcomes.
Pickering and colleagues pooled data on 80 oropharyngeal cancer cases in The Cancer Genome Atlas. The researchers examined all tumors based upon their level of HPV gene expression and hypothesized that variations in HPV biology lead to differences in treatment response and outcomes.
Results showed an initial panel of 582 genes classified by expression among three HPV subgroups — a high-HPV group, a low-HPV group and an HPV-negative group. Additional analyses identified a panel of 38 genes that distinguished between the two HPV-positive subgroups.
HemOnc Today spoke with Pickering about the study and the potential implications of the findings.
Question: What prompted this research?
Answer: HPV causes a large fraction of head and neck tumors — most commonly in the oropharynx subsite. Although patients with HPV-driven tumors generally do well, a subset still have poor outcomes, and a significant fraction have increased morbidity from the intense radiotherapy-based treatment. There are no good biomarkers to identify prognosis in HPV-driven head and neck tumors. We tried to use genomic tools to understand why certain patients do well and others do not, and to potentially use this information to identify which patients can receive less intensive therapy and who may need more intense therapy.
Q: How did you conduct the study?
A: We collected data from The Cancer Genome Atlas head and neck cancer project and identified tumors from the oropharynx subsite. Our hypothesis was essentially that because HPV is associated with good outcomes, perhaps the level of HPV function in a tumor could stratify very good or poor outcomes..
Q: What did you find?
A: We looked for human genes associated with the level of HPV expression and this led us to a gene expression signature that ended up being highly prognostic. We identified a subset of HPV-positive patients who did very well and another subset who did very poorly. We then validated the finding in an independent cohort from Johns Hopkins University and in patients with HPV-positive cervical cancer from The Cancer Genome Atlas. We found this 38-gene signature could effectively stratify patients into good- and poor-prognoses groups. This signature and stratification seem to also be associated with the level of HPV function in the tumor.
Q: What are the clinical implications of the findings?
A: This biomarker could be used to stratify patients with HPV-positive head and neck cancer by their risk, which can be used to personalize therapy. Patients with low-risk HPV can be treated with reduced-intensity therapy, lower radiation doses or other chemotherapy that may reduce their long-term morbidity. For those patients with high-risk disease, the biomarker could help decide which patients should receive more intense therapy or perhaps go on to a clinical trial. All of these findings need to be evaluated in larger cohorts..
Q: What is next for research?
A: All of these initial studies were conducted on RNA sequence data from frozen tumor samples. A next step is to transition this into an assay amendable to samples routinely used in the clinic. We are doing this now. We are also in the laboratory working on the biology that we identified associated with this biomarker and trying to target some of those differentially expressed pathways between the good and poor outcome groups. – by Jennifer Southall
Reference:
Pickering CR, et al. JCI Insight. 2019;doi:10.1172/jci.insight.124762.
For more information:
Curtis R. Pickering, PhD, can be reached at The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030; email: crpickering@mdanderson.org.
Disclosure: Pickering reports no relevant financial disclosures.