Dear Editors,

We read with great interest the recent article by Hoenle et al., ‘HPV status fully mediates sex-related survival differences in oropharyngeal cancer: a population-based analysis with formal interaction testing’.Reference Hoenle, Rickert, Steinhart and Lorenz¹ The authors should be commended for addressing an important clinical question in a large contemporary population-based cohort. Their rationale is compelling, and the attempt to explore formally the relationship between sex, p16 and/or human papillomavirus (HPV) status and survival represents a valuable contribution to the literature.

We would, however, urge caution in concluding that HPV status ‘fully and mechanistically’ explains apparent sex-related differences in survival, particularly when considering overall survival. The principal limitation is inherent to the Surveillance, Epidemiology, and End Results (SEER) dataset. As the authors acknowledge, smoking history, performance status and co-morbidity data are unavailable.Reference Hoenle, Rickert, Steinhart and Lorenz¹ These are crucial prognostic determinants in oropharyngeal squamous cell carcinoma (SCC); their omission limits the extent to which differences in overall survival can be attributed independently to sex or HPV status.

The importance of tobacco exposure is particularly well established. Platek et al. demonstrated that, among patients with oropharyngeal SCC treated with chemoradiotherapy, current smoking was associated with substantially increased mortality in both HPV-positive and HPV-negative disease.Reference Platek, Jayaprakash, Merzianu, Platek, Cohan and WL² Furthermore, Ang et al. found smoking to be an independent and significant determinant of overall survival, with tobacco exposure contributing to their risk stratification of patients with HPV-positive disease.Reference Ang, Harris, Wheeler, Weber, Rosenthal and Nguyen-Tân³ More recently, a meta-analysis of 15 studies of HPV-related oropharyngeal SCC found that ever-smokers had more than twice the mortality of never-smokers (hazard ratio 2.4), while current smoking was associated with a hazard ratio for overall mortality of 3.2.Reference Ference, Liao, Gao and Mehta⁴ Smoking therefore represents more than simply an aetiological risk factor; it is an important prognostic variable even after HPV status has been established.

Co-morbidity and baseline functional status are similarly relevant. Habbous et al. demonstrated that increasing Charlson Comorbidity Index was independently associated with poorer overall survival in oropharyngeal cancer, including after adjustment for HPV status.Reference Habbous, Harland, La Delfa, Fadhel, Xu and Liu⁵ Skillington et al. subsequently found 5-year overall survival of 71 per cent in patients with none-to-mild co-morbidity compared with 49 per cent in those with moderate-to-severe co-morbidity, and demonstrated particular prognostic importance of co-morbidity in p16-negative disease.Reference Skillington, Kallogjeri, JS and Piccirillo⁶ Contemporary oropharyngeal SCC series have also demonstrated the importance of performance status. In a 2023 cohort, Embring et al. found both impaired performance status (hazard ratio 1.51) and current smoking (hazard ratio 2.57) to be independent adverse prognostic factors for overall survival alongside HPV status.Reference Embring, Onjukka, Mercke, Lax, Berglund and Friesland⁷ Klasen et al. similarly identified Eastern Cooperative Oncology Group performance status as a significant predictor of overall survival on multivariable analysis in their sex-specific oropharyngeal SCC cohort.Reference Klasen, Wuerdemann, Rothbart, Prinz, Eckel and Suchan⁸

Alcohol exposure may provide an additional source of residual confounding. Koyama et al., studying 2626 patients with oral and pharyngeal malignancies, reported increased all-cause mortality among heavy alcohol consumers after adjustment for smoking, stage, treatment and other co-variates; importantly, this association was observed in both men and women.Reference Koyama, Tabuchi, Morishima and Miyashiro⁹ Alcohol consumption has also been shown to differ according to sex in oropharyngeal SCC cohorts.Reference Klasen, Wuerdemann, Rothbart, Prinz, Eckel and Suchan⁸

Importantly, socioeconomic status has not been considered. Human papillomavirus-positive oropharyngeal SCC patients had higher rates of smoking and deprivation than the general Scottish population in a recent study, challenging the concept that HPV-positive oropharyngeal SCC is not associated with these factors.Reference Donachie, Horackiewicz, Philp, Slim, Hurley and Douglas¹⁰ We also know that socioeconomic deprivation is associated with incomplete radiotherapy treatment in head and neck cancer, potentially contributing to inequity in outcomes.Reference Brooker, Parvulescu, Jones, Meysner, Moore and Dailey¹¹ Treatment interruptions may themselves negatively affect locoregional control and overall survival.Reference Gharzai, Morris, Schipper, Kidwell, Nguyen-Tân and Rosenthal¹²

Although the reported findings cannot be considered ‘unambiguous’, we strongly support the authors’ calls for sex-disaggregated reporting in future de-escalation trials. We would further encourage studies of sex-related outcomes in oropharyngeal SCC to consider the patient as a whole, rather than attributing observed differences primarily to tumour biology alone. Sex, HPV status, smoking, alcohol exposure, co-morbidity, performance status and socioeconomic circumstances may all influence treatment, including treatment completion and survival. Mixed-methods attempts to incorporate these behavioural, clinical and social determinants alongside tumour-specific characteristics may help optimise our patients’ treatment.