Disparities in surgery for early-stage cancer: the impact of refusal Abstract Background For early-stage cancer surgery is often curative, yet refusal of recommended surgical interventions may be contributing to disparities in patient treatment. This study aims to assess predictors of early-stage cancers surgery refusal, and the impact on survival. Methods Patients recommended surgery with primary stage I and II lung, prostate, breast, and colon cancers, diagnosed between 2007–2014, were identified in the Surveillance, Epidemiology and End Results database (n = 498,927). Surgery refusal was reported for 5,757 (1.2%) patients. Associations between sociodemographic variables and surgery refusal by cancer type were assessed in adjusted multivariable logistic regression models. The impact of refusal on survival was investigated using adjusted Cox-Proportional Hazard regression in a propensity score-matched cohort. Results Increasing age (p < 0.0001 for all four cancer types), non-Hispanic Black race/ethnicity (ORadjBREAST 2.00, 95% CI 1.68–2.39; ORadjCOLON 3.04, 95% CI 2.17–4.26; ORadjLUNG 2.19, 95% CI 1.77–2.71; ORadjPROSTATE 2.02, 95% CI 1.86–2.20; vs non-Hispanic White), insurance status (uninsured: ORadjBREAST 2.75, 95% CI 1.89–3.99; ORadjPROSTATE 2.10, 95% CI 1.72–2.56; vs insured), marital status (ORadjBREAST 2.16, 95% CI 1.85–2.51; ORadjCOLON 1.56, 95% CI 1.16–2.10; ORadjLUNG 2.11, 95% CI 1.80–2.47; ORadjPROSTATE 1.94, 95% CI 1.81–2.09), and stage (ORadjBREAST 1.94, 95% CI 1.70–2.22; ORadjCOLON 0.13, 95% CI 0.09–0.18; ORadjLUNG 0.71, 95% CI 0.52–0.96) were all associated with refusal; patients refusing surgery were at increased risk of death compared to patients who underwent surgery. Conclusions More vulnerable patients are at higher risk of refusing recommended surgery, and this decision negatively impacts their survival.

Changes in use of opioid therapy after colon cancer diagnosis: a population-based study Abstract Purpose To describe patterns of opioid use in cancer survivors. Methods In a cohort study of colon cancer patients diagnosed during 1995–2014 and enrolled at two Kaiser Permanente regions, we constructed quarterly measures of opioid use from 1 year before cancer diagnosis through 5 years after diagnosis to examine changes in use. Measures included any use, incident use, regular use (use ≥ 45 days in a 91-day quarter), and average daily dose (converted to morphine milligram equivalent, MME). We also assessed temporal trends of opioid use. Results Of 2,039 colon cancer patients, 11–15% received opioids in the four pre-diagnosis quarters, 68% in the first quarter after diagnosis, and 15–17% in each subsequent 19 quarters. Regular opioid use increased from 3 to 5% pre-diagnosis to 5–7% post diagnosis. Average dose increased from 15 to 17 MME/day pre-diagnosis to 14–22 MME/day post diagnosis (excluding the quarter in which cancer was diagnosed). Among post-diagnosis opioid users, 73–95% were on a low dose (< 20 MME/day). Over years, regular use of opioids increased in survivorship with no change in dosage. Conclusion Opioid use slightly increased following a colon cancer diagnosis, but high-dose use was rare. Research is needed to differentiate under- versus over-treatment of cancer pain.

Racial differences in brain cancer characteristics and survival: an analysis of SEER data Abstract Purpose Racial disparity with shorter survival for Blacks than Whites is well known for many cancers. However, for brain cancer, some national cancer registry studies have shown better survival among Blacks compared to Whites. This study aimed to systematically investigate whether Blacks and Whites differ in survival and also in tumor characteristics and treatment for neuroepithelial brain tumors. Methods The National Cancer Institute’s Surveillance Epidemiology and End Results (SEER) database was used to identify non-Hispanic White and Black patients diagnosed with malignant, histologically confirmed neuroepithelial brain cancer from 2004 through 2015. Racial differences in brain cancer survival were compared using Kaplan–Meier curve and Cox proportional hazard models. The associations of race with tumor and treatment characteristics (location, size, grade, surgical type) were examined using multinomial logistic regression. Results After adjusting for demographic, tumor, and treatment factors, there were no significant differences in survival for non-Hispanic Blacks compared to non-Hispanic Whites [hazard ratio (HR) 1.05, 95% confidence interval (CI) 0.99–1.10]. Non-Hispanic Blacks had higher odds of being diagnosed with tumors of unknown grade [odds ratio (OR) 1.16, 95% CI 1.05–1.29], unknown size (OR 1.14, 95% CI 1.01–1.29), infratentorial (OR 1.12, 95% CI 1.01–1.24) or overlapping area (OR 1.39, 95% CI 1.14–1.70), and lower odds of having a total surgical resection (OR 0.83, 95% CI 0.74–0.93). Conclusion Non-Hispanic Blacks do not exhibit longer brain cancer-specific survival than non-Hispanic Whites. They were more likely to have tumors of unknown size or grade and less likely to receive total surgical resection, which may result from racial differences in access to and use of healthcare.

Body size and dietary risk factors for aggressive prostate cancer: a case–control study Abstract Purpose Diet and body size may affect the risk of aggressive prostate cancer (APC), but current evidence is inconclusive. Methods A case–control study was conducted in men under 75 years of age recruited from urology practices in Victoria, Australia; 1,254 with APC and 818 controls for whom the presence of prostate cancer had been excluded by biopsy. Dietary intakes were assessed using a validated food frequency questionnaire. Multivariable unconditional logistic regression estimated odds ratios and confidence intervals for hypothesized risk factors, adjusting for age, family history of prostate cancer, country of birth, socioeconomic status, smoking, and other dietary factors. Results Positive associations with APC (odds ratio, 95% confidence intervals, highest vs. lowest category or quintile) were observed for body mass index (1.34, 1.02–1.78, Ptrend = 0.04), and trouser size (1.54, 1.17–2.04, Ptrend = 0.001). Intakes of milk and all dairy products were inversely associated with APC risk (0.71, 9.53–0.96, Ptrend = 0.05, and 0.64, 0.48–0.87, Ptrend = 0.012, respectively), but there was little evidence of an association with other dietary variables (Ptrend > 0.05). Conclusions We confirmed previous evidence for a positive association between body size and risk of APC, and suggest that consumption of dairy products, and milk more specifically, is inversely associated with risk.

Correlates of poor adherence to a healthy lifestyle among a diverse group of colorectal cancer survivors Abstract Purpose Lifestyle factors may have a synergistic effect on health. We evaluated the correlates of poor adherence to a healthy lifestyle among a diverse sample of colorectal cancer (CRC) survivors to inform future lifestyle promotion programs. Methods Lifestyle questions from a cross-sectional survey were completed by 283 CRC survivors (41% Hispanic, 40% rural, 33% low income). Adherence to recommendations (yes/no) for physical activity, fruit and vegetable servings/day, avoiding tobacco, and healthy weight was summed to create an overall lifestyle quality score. Polytomous logistic regression was used to evaluate correlates of good (reference group), moderate, and poor overall lifestyle quality. Potential correlates included sociodemographic characteristics, cancer-related factors, and indicators of health and well-being. Results CRC survivors with poor adherence were 2- to 3.4-fold significantly more likely to report multiple comorbidities, poor physical functioning, fatigue, anxiety/depressive symptoms, and poor social participation. In multivariable analyses, poor physical functioning was the only significant correlate of poor adherence to lifestyle recommendations, compared to good adherence [OR (95% CI) 3.4 (1.8–6.4)]. The majority of survivors, 71% and 78%, indicated interest in receiving information on exercise and eating a healthy diet, respectively. Conclusion Future lifestyle promotion programs for CRC survivors should carefully consider indicators of physical and psychosocial health and well-being, especially poor physical functioning, in the design, recruitment, and implementation of these health programs.

Changing trends in liver cancer incidence by race/ethnicity and sex in the US: 1992–2016 Abstract Purpose Liver cancer incidence continues to increase while incidence of most other cancers is decreasing. We analyze recent and long-term trends of US liver cancer incidence by race/ethnicity and sex to best understand where to focus preventive efforts. Methods Liver cancer incidence rates from 1992 to 2016 were obtained from the Surveillance, Epidemiology, and End Results registry. Delay-adjusted age-standardized incidence trends by race/ethnicity and sex were analyzed using joinpoint regression. Age-specific incidence was analyzed using age-period-cohort models. Hepatitis C seroprevalence by cohort was calculated using National Health and Nutrition Examination Survey data. Results Liver cancer incidence has peaked in males and Asian or Pacific Islanders. Hispanic males, a high-incidence population, are experiencing a decrease in incidence, although not yet statistically significant. In contrast, incidence continues to increase in females, although at lower rates than in the 1990s, and American Indian/Alaska Natives (AI/ANs). Liver cancer incidence continues to be higher in males. Non-Hispanic Whites have the lowest incidence among racial/ethnic groups. Trends largely reflect differences in incidence by birth-cohort, which increased considerably, particularly in males, for those born around the 1950s, and continues to increase in females and AI/ANs. The patterns in males are likely driven by cohort variations in Hepatitis C infection. Conclusions Liver cancer incidence appears to have peaked among males. However, important differences in liver cancer trends by race/ethnicity and sex remain, highlighting the need for monitoring trends across different groups. Preventive interventions should focus on existing liver cancer disparities, targeting AI/ANs, females, and high-incidence groups.

Racial disparities of liver cancer mortality in Wisconsin Abstract Purpose To calculate tract-level estimates of liver cancer mortality in Wisconsin and identify relationships with racial and socioeconomic variables. Methods County-level standardized mortality ratios (SMRs) of liver cancer in Wisconsin were calculated using traditional indirect adjustment methods for cases from 2003 to 2012. Tract-level SMRs were calculated using adaptive spatial filtering (ASF). The tract-level SMRs were checked for correlations to a socioeconomic advantage index (SEA) and percent racial composition. Non-spatial and spatial regression analyses with tract-level SMR as the outcome were conducted. Results County-level SMR estimates were shown to mask much of the variance within counties across their tracts. Liver cancer mortality was strongly correlated with the percent of Black residents in a census tract and moderately associated with SEA. In the multivariate spatially-adjusted regression analysis, only Percent Black composition remained significantly associated with an increased liver cancer SMR. Conclusions Using ASF, we developed a high-resolution map of liver cancer mortality in Wisconsin. This map provided details on the distribution of liver cancer that were inaccessible in the county-level map. These tract-level estimates were associated with several racial and socioeconomic variables.

Economics of public health programs for underserved populations: a review of economic analysis of the National Breast and Cervical Cancer Early Detection Program Abstract Purpose The purpose of this paper is to provide a brief overview of economic analysis methods used in estimating the costs and benefits of public health programs and systematically review the application of these methods to the National Breast and Cervical Cancer Early Detection Program (NBCCEDP). Methods Published literature on economic analyses of the NBCCEDP was systematically reviewed. The Consensus on Health Economic Criteria checklist was used to assess methodological quality of the included studies. Results Methods available for economic analysis of public health programs include program cost, cost-effectiveness, cost–utility, cost–benefit analysis, and budget impact analysis. Of these, program cost analysis, cost-effectiveness analysis, and cost–utility analysis have been applied to the NBCCEDP in previously published literature. Conclusion While there have been multiple program cost analyses, there are relatively fewer cost-effectiveness and cost–utility studies and no cost–benefit and budget impact analysis studies to evaluate the NBCCEDP. Addressing these gaps will inform implementation of effective public health programs with equitable resource allocation to all population subgroups.

Polymorphisms in oxidative stress pathway genes and prostate cancer risk Abstract Purpose Age-related factors including oxidative stress play an important role in prostate carcinogenesis. We hypothesize that germline single-nucleotide polymorphisms (SNPs) in oxidative stress pathway are associated with prostate cancer (PCa) risk. In this study, we aim to examine which of these SNPs is associated with PCa. Methods Participants included in this analyses came from the “Genetic Susceptibility, Environment and Prostate Cancer Risk Study” conducted at the Veterans Affairs Portland Health Care System. After applying exclusion criteria, 231 PCa cases and 382 prostate biopsy-negative controls who had genotyping data on twenty-two single-nucleotide polymorphisms (SNPs) in six genes (MAPK14, NRF2, CAT, GPX1, GSTP1, SOD2, and XDH) associated with oxidative stress pathway were included in the analyses. The genotyping of SNPs was conducted by the Illumina BeadXpress VeraCode platform. We investigated these SNPs in relation to overall and aggressive PCa risk using logistic regression models controlling for relevant covariates. Results One SNP in the MAPK14 (rs851023) was significantly associated with incident PCa risk. Compared to men carrying two copies of allele A, the presence of one or two copies of the G allele was associated with decreased risk of PCa [OR (95% CI) 0.19 (0.06–0.51)]. There was no statistically significant association between other SNPs in the NRF2, CAT, GPX1, GSTP1, SOD2, and XDH genes and PCa risk. Conclusions The MAPK14 gene SNP rs851023 was associated with PCa and aggressive PCa risk after multiple comparison adjustment. Further studies in other populations or functional studies are needed to validate the finding.

The association between clinically determined periodontal disease and prostate-specific antigen concentration in men without prostate cancer: the 2009–2010 National Health and Nutrition Examination Survey Abstract Purpose We evaluated the association between clinically assessed periodontal disease and serum prostate-specific antigen (PSA) concentration in men without a prostate cancer diagnosis in a US nationally representative sample of non-institutionalized men. Methods Included were 1263 men aged ≥ 40 years who participated in the National Health and Nutrition Examination Survey in 2009–2010. Measurements of periodontal health and tooth count were used to define periodontal disease severity (no, mild, moderate, severe) and edentulism. Linear and logistic regressions were used to estimate the association of periodontal disease severity and edentulism with PSA concentration and elevated PSA, respectively. Results Adjusting for age and other factors including race, body mass index, and education, the natural logarithm of PSA concentration did not change with increasing severity (mild − 0.20, 95% confidence interval [CI] − 0.34 to − 0.05; moderate − 0.12, 95% CI − 0.26 to 0.01; severe − 0.16, 95% CI − 0.43 to 0.12; edentulism − 0.16, 95% CI − 0.35 to 0.04; P-trend 0.13) compared with dentate men without periodontal disease. Although the multivariable-adjusted ORs of elevated PSA were not statistically significant, participants with more severe periodontal disease were less likely to have PSA > 2.0 and > 2.5 ng/mL, but more likely to have PSA > 4.0 ng/mL, compared to dentate men without periodontal disease. Similar non-significant associations with PSA were observed when comparing edentulous men to dentate men without periodontal disease. Conclusions In this US nationally representative sample, men with periodontal disease did not have higher serum PSA and were not more likely to have clinically elevated PSA after taking into account age and other factors, contrary to the hypothesis. This study suggests that periodontal disease does not notably affect the specificity of PSA for prostate cancer screening.