Discussion
This is the first population-based study that confirms worse HPV-associated OPSCC survival outcomes in black patients across a range of adjustment sets incorporating demographic, clinical, and individual-level SES variables. It has previously been documented that racial differences in SES and access to care are significant determinants of head and neck cancer disparities.(11,12,16) In OPSCC, racial differences in HPV tumor-status have been shown to contribute to the survival disparity.(17) However, the relative contribution of tobacco use, alcohol use, treatment, and SES to racial disparities in HPV-associated OPSCC has not been determined. Our analysis with sequential adjustment sets demonstrates that racial differences in SES contributes to some but not all of the disparity.
Our findings build upon the growing evidence in current literature for the presence of racial disparities in OPSCC. In a large database analysis of OPSCC stratified by HPV-status, Faraji et al. found that there was a weak but non-significant trend towards worse overall survival (OS) among black patients with HPV-associated OPSCC after adjusting for age, sex, race, year of diagnosis, insurance status, income, education, rural residence, facility region, TNM stage, Charlson‐Deyo score, and treatment.(18) In an analysis of the SEER database among patients with HPV-associated OPSCC, black patients had significantly worse cancer-specific mortality than white patients even after adjusting for county-level indicators of SES.(19) In another study, no racial difference in OS was found when controlling for treatment received in the Veteran’s Affairs medical system.(20) It is possible that the more homogenous care afforded by the VA system could have mitigated some of the racial disparity attributable to differences in SES and access to care.
There may be several explanations for the persistence of the racial disparity in OS after adjustment for clinical and socioeconomic factors. First, unmeasured confounders in socioeconomic factors, physician and system factors, and access to care may exist. We found that black patients were more likely to have a low income compared to white patients, but our sample lacked the power to adjust for other variables such as frequency of primary care visits or routine dental visits. Studies have found that patients lacking routine dental visits are diagnosed at more advanced stages of head and neck cancer,(21–23) although it is unknown if this pattern varies by race. In the past, racial disparities were largely attributed to genetic differences, but it is now recognized that race is a complex social construct and genetic factors are not likely to integrate with socially-defined racial groups as previously thought.(24)
Our study has several limitations. First, our study population had a small number of black patients with HPV-associated OPSCC. It should be noted that this is a limitation nationally, thereby limiting precise estimation of differences in this study population. Also, p16 testing was not routinely performed at the time of data collection for this study so our sample likely underestimates the true number of HPV-associated OPSCC patients in [Blinded for peer review]. Given the small sample size of our study, it is important to recognize that some of the non-significant findings may be due to lack of statistical power rather than a true non-effect. Efforts are being undertaken to combine databases to increase the sample size of black patients with HPV-associated OPSCC for future validation studies. Another limitation is that the [Blinded for peer review] database did not collect data on patient comorbidities, which may confound the relationship between race and OS. Disease specific mortality was not included in the analysis due to a lack of data on the specific cause of death for many of the [Blinded for peer review] patients.
Our findings may have implications for cancer treatment and future research directions. There is strong evidence to support that patients with HPV-associated OPSCC have improved treatment response and survival outcomes compared to patients with HPV-negative disease.(25) These findings have led to efforts to de-intensify treatment with the goal of alleviating therapy-related morbidity and mortality. Our results suggest that black patients with HPV-associated OPSCC may have worse outcomes despite their HPV-positive tumor status, and therefore, may not receive the same benefits from treatment deintensification. It is important to note that in our sample, we found no differences by race with regards to treatment type or number of treatments modalities received. Future de-intensification trials should closely monitor survival outcomes by race to ensure that this disparity is not being propagated.