Introduction

While cervical cancer incidence rates (IR) dropped in the last 20 years, non-cervical human papillomavirus (HPV)-related cancers increased. Many people in Texas (TX) live in medically underserved areas and have higher risk of developing HPV-related cancers. Since previous studies of these regions focused on cervical cancer, we included other HPV-related cancers in our analysis of IR in East TX and TX Mexico Border compared to other TX regions.

Methods

Cancer data from 2006-2019 were obtained from the TX Cancer Registry. Cases of HPV-related cancer are defined by ICD-O-3 codes for sites: cervix, vagina, vulva, penis, anus, and oropharynx. Sites were refined by histology codes likely to be HPV-related (cervical: 8010-8671 and 8940-8941, other sites: 8050-8084 and 8120-8131). This included patient level data on age, sex, diagnosis year, race/ethnicity, county, and poverty. To calculate IR per 100,000, we used five-year population estimates (2010, 2015, 2019) from the American Community Survey. We drew heat maps to visualize cancer IR by county. To control potential confounders, we added county level risk factors: rates for smoking, excessive drinking, obesity, STIs, primary care provider availability and dentist, from the County Health Rankings and Roadmaps program. We reported IRs by region (Border, East TX, rest of TX) and period (2006-2010, 2011-2015, 2016-2019), and used zero inflated Poisson regression models to estimate unadjusted and adjusted risk ratio (RR) for association of each type of cancer and region. Last, we created adjusted models for each cancer by period to see time trends of regional differences.

Results

From 2006-2019, IRs of HPV-related cancers had varying trends by region. Anal cancer rates decreased at the Border from IR: 1.2 [95% CI: 1.0-1.5] to 1.1 [0.9-1.3] but increased in East TX (2.5 [2.1-2.9] to 2.7 [2.4-3.1]) and the rest of TX (2.0 [1.9-2.1] to 2.1 [2.0-2.2]). Vaginal cancer IRs decreased at the Border (0.7 [0.5-1.0] to 0.5 [0.3-0.7]) and the rest of TX (0.7 [0.6-0.8]) to 0.5 [0.4-0.5]) but were stable in East TX (0.6 [0.4-0.9] to 0.6 [0.4-0.9]). Vulvar cancer IRs increased at the Border (1.3 [1.0-1.7] to 1.4 [1.1-1.8]) but decreased in East TX (2.9 [2.4-3.6] to 2.6 [2.2-3.2]) and the rest of TX (2.1 [2.0-2.3] to 2.0 [1.8-2.1]). Cervical cancer IRs decreased less at the Border (13.3 [12.2-14.4] to 12.1 [11.1-13.1]) than in East TX (12.6 [11.4-13.8] to 9.9 [8.9-11.0]) and the rest of TX (12.0 [11.7-12.4] to 9.2 [8.9-9.5]). Penile cancer IRs in East TX declined most (1.4 [1.1-1.9] to 0.8 [0.6-1.2]) vs. the Border (1.9 [1.5-2.4] to 1.8 [1.4-2.2]) and the rest of TX (1.0 [0.9-1.1] to 0.9 [0.8-1.0]).

In unadjusted models for the Border, risk of cervical (RR: 1.3 [95% CI: 1.2-1.3]) and penile (1.8 [1.5-2.1]) cancers was higher than the rest of TX. Risk of anal (0.6 [0.6-0.7]) and oropharyngeal (0.5 [0.5-0.6]) cancers was lower than the rest of TX. After adjustment, risk of cervical cancer became lower than the rest of TX (0.8 [0.7-0.9]), and results for penile cancer were not statistically significant. In East TX unadjusted models, risk of anal (1.5 [1.4-1.7]), oropharyngeal (1.6 [1.5-1.7]), penile (1.6 [1.3-2.0]), and vulvar (1.6 [1.4-1.9]) cancers were higher than the rest of TX. In the adjusted model, risk of oropharyngeal cancer (0.9 [0.9-1.0]) decreased to less than the rest of TX, and East TX region no longer increased risk of other cancers.

Conclusion

IRs of HPV-related cancers differed based on TX region. Anal cancer IR increased in East TX and the rest of TX but decreased at the Border. Conversely, vulvar cancer IR decreased in East TX and the rest of TX but increased at the Border. Difference of HPV cancer IR across regions did not change much over time. Patient demographics, risk factors, and access to care were attributed to some observed differences on cancer IR across regions. This indicates that targeted prevention efforts towards these regions, especially in poor communities, may benefit future generations.