Oropharyngeal Cancer: HPV and Epidemiology

Hi Team,

Happy October! We're moving out of the reconstruction world and into oropharynx and robotic surgery for the month. As always, we meet Wednesday mornings @ 7AM, donuts and coffee for those who can join in person. If you can't make it, send me an email or grab any of us in the OR.

Here's the plan for the month: this week is HPV and epidemiology, next week is staging and systemic therapy (including the cooperative group trials that tried to take chemotherapy away), and the last two weeks are transoral robotic surgery, from its development here at Penn through the outcomes data and the PATHOS trial. This week is the foundation for everything else, so it's worth reading the primary papers rather than the summaries.

Where It Started - HPV as a Cause of Head and Neck Cancer

(Gillison JNCI 2000) is a retrospective study out of Johns Hopkins that tested tumors from 253 head and neck cancer patients for HPV using PCR, Southern blot, and in situ hybridization. TL;DR: HPV was found in 25% of tumors, HPV16 accounted for 90% of the positives, and an oropharyngeal primary increased the odds of HPV positivity about six-fold. The HPV-positive oropharyngeal cancers looked like a different disease: patients were less likely to be heavy drinkers or smokers, the tumors had basaloid morphology, TP53 mutations were rare, and the risk of death from cancer was reduced by 59% (HR 0.41). HPV had been established as an oncovirus in cervical cancer decades earlier, but this is the paper that made the case for HPV-positive oropharyngeal cancer as a distinct entity. It's an interesting read when you remember that most of us in practice today trained when HPV status wasn't on the pathology report at all (some of us more than others... although I still had to get the VA to purchase its own p16 IHC stain for oropharyngeal tumors in the late 2010s!).

Prognosis - Surgical and Chemoradiation Cohorts Agree

(Licitra JCO 2006) is a series of 90 consecutive oropharyngeal cancer patients treated with primary surgery at the Istituto Nazionale dei Tumori in Milan, Italy. Nineteen percent had integrated HPV16 DNA, and HPV positivity significantly improved overall survival and reduced both relapse and second primary tumors. It's one of the earliest surgical cohorts to show the survival advantage, which matters because the paper everyone quotes is a nonsurgical one.

(Ang NEJM 2010) is that paper. It is a retrospective analysis of the oropharyngeal patients enrolled on RTOG 0129, a randomized trial of accelerated versus standard fractionation radiation with cisplatin for stage III/IV head and neck cancer. Of 323 oropharyngeal patients, 63.8% were HPV-positive, and 3-year overall survival was 82.4% versus 57.1% for HPV-negative patients, a 58% reduction in the risk of death after adjustment (HR 0.42). Risk of death went up with every additional pack-year of smoking. The recursive partitioning analysis at the end sorts patients into low, intermediate, and high risk using four variables: HPV status, pack-years (with 10 pack-years as the cutpoint), T stage, and N stage. You will hear these risk groups quoted on rounds and at tumor board, and they have been the basis for trial eligibility for more than a decade. Keep in mind that this was a post-hoc analysis of a trial that was not designed to answer the HPV question, and the smoking cutpoint has been debated ever since.

(Maxwell Laryngoscope 2014) from Pittsburgh looked at University of Washington quality of life scores in 177 head and neck cancer patients by p16 status and primary treatment modality. p16-positive patients had better quality of life at baseline and after treatment, with better speech, chewing, and swallowing scores beyond one year. Among p16-positive patients, primary surgery versus nonsurgical treatment made no difference in overall quality of life or any of the 12 domains. Here's a question to sit with before our TORS weeks: if survival is excellent and quality of life looks similar regardless of modality, what should actually drive the choice between primary surgery and primary chemoradiation for a given patient? Many of the studies that discuss quality of life have a lot of methodological noise within them that makes it hard to actually guide a patient (such as the ORATOR trial, which we'll get to in our TORS weeks).

p16 Versus HPV Testing - Not Quite the Same Thing

The curriculum asks you to compare p16 immunohistochemistry with HPV-specific testing, but its reading list doesn't include a paper on this, so I searched PubMed for one. (Mehanna Lancet Oncol 2023; HNCIG-EPIC-OPC) is an individual patient data analysis of 7,654 oropharyngeal cancer patients from 13 cohorts in Europe and Canada. Of the 3,805 p16-positive patients, 10.9% were HPV-negative on DNA or RNA testing, and that discordance was much more common outside the tonsil and base of tongue (29.7% vs 9.0%), a great example of pre-test probability driving the predictive value of a test! The survival differences are large: 5-year overall survival was 81.1% for p16+/HPV+, 54.7% for p16+/HPV-, 53.2% for p16-/HPV+, and 40.4% for p16-/HPV-. In other words, a p16+/HPV- tumor behaves much more like an HPV-negative tumor than its p16 stain suggests. The authors recommend HPV testing for all trial patients and wherever HPV status would change care, especially in regions with a lower HPV-attributable fraction. Note that there were too few discordant North American patients to draw firm conclusions here, and that p16 alone remains the surrogate used for AJCC 8th edition staging, which is where we'll start next week.

p16 is a cheap immunohistochemical stain, whereas HPV in situ hybridization is more costly. At Penn, any p16-positive SCCa of the sinonasal tract or an unknown primary gets reflex HPV testing. In the oropharynx, nonkeratinizing SCCa gets p16 staining, and if the staining isn't strong and diffuse (>70%), it gets reflex HPV testing. Keratinizing SCCa of the oropharynx also gets reflex HPV testing. Dr. Jalaly (cc'ed) will also chase a p16-negative case if it clinically appears to be an HPV-associated tumor.

Projections and Prevention - Vaccination Works, but Slowly

(Xu Head Neck 2019) is a SEER analysis out of MD Anderson projecting oropharyngeal cancer incidence. Incidence rose by 1.94% per year from 2000 to 2015 and is projected to keep rising over the next 30 years, with more than half of new cases in non-Hispanic white men aged 55 to 69. By 2045, oropharyngeal cancer is projected to be the third most common cancer in that group.

So does vaccination change this? The proof that it works on the upper aerodigestive tract comes from HPV types 6 and 11 rather than 16. (Novakovic J Infect Dis 2018) tracked juvenile-onset recurrent respiratory papillomatosis through the Australian Paediatric Surveillance Unit after Australia's national quadrivalent vaccination program for women aged 12 to 26 began in 2007. Incidence fell from 0.16 to 0.02 per 100,000 between 2012 and 2016, and none of the mothers of affected children had been vaccinated before pregnancy. Juvenile RRP is now close to disappearing in Australia, and it fell quickly because it is acquired at birth.

In the US, (Chaturvedi JCO 2017) used national survey data from 2011 to 2014 and found that vaccine-type oral HPV infections were reduced by about 88% in vaccinated versus unvaccinated young adults. However, because so few people were vaccinated (only 6.9% of men at the time), the population-level reduction was 17% overall and 6.9% in men. Uptake is better now but has stalled: in 2024, 78.2% of US adolescents had at least one dose and 62.9% were up to date (Pingali MMWR 2025), and coverage did not increase in 2025 for the fourth straight year, with lower coverage in rural areas (Pingali MMWR 2026). Oropharyngeal cancer shows up decades after the oral infection, so we should expect a delay of roughly 30 years before vaccination affects incidence, and with US coverage where it is, the decline will likely be slower here than in Australia. If you like this stuff, check out the Population Health Service in the AHNS!

Thanks for taking the time to read this and looking forward to discussing on Wednesday!

Have a great week and happy fall!

Rob