CHICAGO A prospective analysis of a phase II Eastern Cooperative Oncology Group study confirms what has up to now been reported only in retrospective, single-institution studies: head and neck squamous cell cancer patients infected with the human papilloma virus have significantly better survival than do their counterparts without the virus.
Human papilloma virus (HPV) positivity conferred a 79% lower risk of death in the multicenter ECOG 2399 study, Dr. Carole Fakhry reported at the annual meeting of the American Society of Clinical Oncology. She said HPV status should now be considered a biomarker for prognosis in head and neck squamous cell cancer (HNSCC). Moreover, these results may necessitate a reinterpretation of survival rates in previous trials to determine whether survival differences were in fact due to HPV status, rather than to the actual therapy that was used, according to Dr. Fakhry of the Johns Hopkins Medical Institutions in Baltimore.
The primary objective of ECOG 2399 was to assess organ preservation with taxane-based induction chemotherapy followed by taxane-based concurrent chemoradiation in resectable stage III and IV larynx and oropharyngeal cancer patients. The trial also sought to estimate disease-free survival and patterns of failure.
Dr. Fakhry and associates evaluated pathologic tissue samples from 96 study participants for the presence of HPV infection, then determined prognostic factors, treatment response, and survival outcomes in terms of HPV status.
HPV statusin particular HPV-16, the dominant viral isolate known to be responsible for a subset of HNSCCwas assessed with in situ hybridization, polymerase chain reaction, and line blot tests. These tests also screened for HPV-31, 33, and 35, which are the other isolates that have been linked to HNSCC.
A total of 40% of patients (38) were found to be HPV positive, and all had oropharyngeal tumors. They were more likely to have a better ECOG performance status and lower cumulative lifetime exposure to smoking than were HPV-negative patients. They were also more likely to be male, have less weight loss on presentation, and present with a stage T2 tumor, Dr. Fakhry reported.
HPV-positive patients had a higher response to induction and chemoradiation therapy. Response rates after induction chemotherapy were 81.6% for HPV-positive patients vs. 55.2% for HPV-negative patients (p=0.01). After chemoradiotherapy, they were 84.2% vs. 56.9%, respectively (p=0.07).
At a median follow-up of 39 months, progression risk was 72% lower and risk of death was 79% lower in HPV-positive patients, compared with HPV-negative patients. These figures were derived from a Cox proportional hazards model, Dr. Fakhry said.
Discussant Thomas F. Pajak, Ph.D., of the Radiation Therapy Oncology Group in Philadelphia said these results are impressive at first glance, but that he was troubled by the Cox analysis in the trial. He proposed that the investigators generate a new Cox model, restrict it only to oropharyngeal patients, and compare HPV status to no more than three other factors in order to obtain a new, and more reliable, estimate of the risk of death. "A Cox model with too many variables in it is unreliable," he said.
In another presentation that looked at HPV-associated HNSCC, Dr. Anil K. Chaturvedi of the National Cancer Institute reported that HPV-related HNSCC has increased in the United States during the last three decades, particularly among white men aged 40-59 years.
Using data from the Surveillance, Epidemiology, and End Results (SEER) registry for the period 1973-2003, Dr. Chaturvedi and co-investigators also found that HPV-related HNSCCs were being diagnosed at more advanced stages and at significantly younger ages. These trends became apparent in the early 1990s. Meanwhile, the incidence of cancers not related to HPV decreased in both men and women, especially in those over the age of 40.
He said the increasing incidence of HPV-linked HNSCC could be due to changes in sexual behavior, and that the decreasing incidence of cancers not related to HPV could be due to the decreased prevalence of smoking.