Background: Increased body mass index (BMI) in midlife is a recognized risk factor for renal cell carcinoma (RCC). Limited data exists on how lifetime BMI patterns, particularly BMI reduction, are associated with RCC overall and its histologic subtypes.

Methods: Using the NIH-AARP Diet and Health Study (n=204,364, 1425 incident RCCs), we evaluated lifetime body weight patterns using 1) BMI at ages 18, 35, 50 years and at baseline (mean: 62 years); 2) BMI trajectory across adulthood; 3) cumulative exposure to excess body weight across adulthood, measured by weighted number of years living overweight/obese (WYO); and 4) BMI change between specific ages. We conducted Cox models to yield hazard ratios (HR) and 95% confidence intervals (CI) for the association of each anthropometric metric with the incidence of overall RCC, aggressive RCC (n=583), fatal RCC (n=339), and histologic subtypes, including clear cell (ccRCC, n=541), papillary (pRCC, n=146), and chromophobe RCC (chRCC, n=64).

Results: A higher BMI at every specific age was associated with a greater hazard of overall RCC and all subtypes (HR=1.10-1.40 per 5-unit increase), except for chRCC (HR=0.80-0.98). Similar patterns were observed for BMI trajectories indicating weight gain during adulthood to overweight/obesity, compared to individuals who maintained normal BMI. Higher WYO (per standard deviation increase) was associated with an elevated hazard of overall RCC (HR=1.17, 95%CI:1.12-2.22), aggressive RCC (HR=1.21, 95%CI:1.13-1.29), fatal RCC (HR=1.16, 95%CI:1.06-1.27), ccRCC (HR=1.20, 95%CI:1.13-1.30), and pRCC (HR=1.13, 95%CI:0.97-1.32), but not with chRCC (HR=0.92, 95%CI:0.68,1.25). BMI reduction of ≥10%, particularly after age 50 years (HR=0.72, 95% CI: 0.52-0.99), was associated with a reduced hazard of RCC, compared with retaining a stable BMI.

Conclusions: Higher cumulative exposure to excess body weight and weight gain during adulthood were associated with increased risk of ccRCC and pRCC, but not with chRCC. Weight loss during adulthood was associated with lower RCC risk.

BACKGROUND: In recent years the U.S. healthcare system has witnessed a substantial increase in telehealth use; however, post-pandemic CMMS will be discontinuing reimbursement for telehealth. Given the enhancement of healthcare access and quality and potential reduction in costs, the VA has invested in teleoncology. The shift from in-person to telehealth care delivery may differentially improve cancer care access and quality in certain clinical settings and for specific patient populations, while potentially exacerbating disparities in care. Our National Cancer Institute-funded center, called Telehealth Research and Innovation for Veterans with Cancer (THRIVE), focused on telehealth-delivered cancer care equity. We seek to understand how social determinants of telehealth (SDTH) – particularly race, poverty, and rurality – affect the use of telehealth.

METHODS: THRIVE draws from the Health Disparities Research Framework and the Consolidated Framework for Implementation Research. THRIVE consists of multiple cores that work synergistically to assess and understand health equity for telehealth-delivered cancer care. These include the Administrative Core, Research and Methods Core, Clinical Practice Network, and Pragmatic Trial.

RESULTS: As of October 2023, we have identified and began training 5 THRIVE Scholars, junior faculty beginning a research career. We have reviewed 20 potential pilot studies, funding six. Additionally, in communication with our funders and advisory boards, we have adjusted our study design and analytic approach, ensuring feasibility while addressing our operational partners’ needs.

CONCLUSIONS: THRIVE has several key strengths. First, the VA healthcare system is large and diverse regarding both healthcare setting type and patient population. Second, we have access to longitudinal data, predating the COVID-19 pandemic, about telehealth use. Finally, equitable access to high quality care for all Veterans is a major tenet of the VA healthcare mission. As a result of these advantages, THRIVE can focus on isolating and evaluating the impact of SDTH on equity in cancer care.

Background: Treatment patterns and survival outcomes in advanced non-small cell lung cancer (NSCLC) are crucial for understanding and improving patient care.

Methods: A patient registry (ConcertAI) was used for this retrospective study. Adult patients who initiated lung cancer treatment following diagnosis of advanced NSCLC (stage IIIB/IIIC/IV) from 01/01/2015-12/31/2021 were identified; clinical characteristics were compared among patient subgroups using Student’s t-test and Chi-square test. Overall survival (OS), progression free survival (PFS), time to treatment (TTT), time to next treatment (TTnT), and time to treatment discontinuation (TTD) were assessed using Kaplan-Meier and Cox Proportional Hazards models. Analyses were performed using SAS Studio (SAS Institute); statistical significance was p-value <0.05.

Results: 8,811 patients were identified (1,370 stage IIIB/IIIC; 7,441 stage IV). Mean age was 66.6 (SD=±10.0) years; 52.7% were male; 65.7% having non-squamous cell carcinoma; 22.5% having squamous cell carcinoma; 11.8% unknown. Median OS was 21.5 [19.2, 23.4] months for stage IIIB/IIIC patients, and 12.4 [11.8, 13.0] months for stage IV patients. Patients with non-squamous cell carcinoma had longer OS in both stages. Median PFS for stage IIIB/IIIC patients was 15.7 [14.4, 17.4] months. Median TTT was 1.2 [0.8, 1.9] months for overall population. Median duration of first line (1L) therapy ranged from 1.4 months for chemotherapy (CT) to 9 months for targeted therapy. Median TTnT from 1L to second line (2L) for CT, immunotherapy (IO), CT plus IO, targeted therapy, and CT plus targeted therapy were 4.3, 21.7, 12.4, 15.8, and 5.9 months, respectively.

Conclusion: Our findings underscore variability in treatment durations and survival outcomes among advanced NSCLC patient subgroups. Particularly, traditional CT had the shortest treatment duration and TTnT, while targeted therapies and IO had extended TTD and TTnT, suggesting that individualized treatment approaches, such as targeted therapies and IO, can be more effective and tolerable treatment options for patients.

Background

Research demonstrates that social factors such as poverty, absent social supports and lower levels of education have significant consequences on health outcomes for vulnerable populations. When resources are lacking, individuals are far less likely to seek cancer prevention and screening services or seek care for existing health problems.

Methods

The Levine Cancer Institute Disparities and Outreach team established a “one stop shop” for care by conducting free, community-based screenings in five cancer domains, with full assessment of barriers to care and navigation to resources. The programs targeted economically disadvantaged individuals living across 24 counties in North and South Carolina. All individuals screened were uninsured/ underinsured. The programs targeted diverse populations including African Americans, Hispanics/Latinos and Native Americans who are at high-risk of being diagnosed with cancer and experiencing worse outcomes. Utilizing a detailed navigation assessment, RN program coordinators used motivational interview-style communication to uncover socioeconomic barriers to care. Patients were then directly referred to partners within the their community.

Results

A total of 2,883 patients were screened for cancer and 40 cancers were diagnosed. All positive findings were navigated for follow-up treatment. Of the individuals who were screened, 62% (n=1,792) were identified with at least one barrier to care. Referrals included: 935 individuals to a community partner for short- or long-term care; 448 individuals scheduled for another cancer screening; 131 for food insecurity; 323 for tobacco cessation; 54 for primary care and 82 for transportation. This method of providing whole human care also resulted in a robust network of 177 community partners and improved trust and participation in health events.

Conclusion

Delivering comprehensive, whole human care where most-at-risk individuals live created access points. Identifying gaps in resources and navigating to solutions prior to a cancer diagnosis provided opportunity to build a bridge before the disparity gap could widen.

BACKGROUND

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related deaths worldwide. Despite the role of clinical trials in advancing HCC treatments, there is limited understanding of clinical trial participation among this patient population. Therefore, this study analyzed clinical trial participation among Medicare beneficiaries diagnosed with HCC.

METHODS

The Medicare 100% Standard Analytical Files were used to identify patients aged 65+ with an encounter for HCC between January 1, 2016 and December 31, 2022. Patients were stratified by clinical trial participation, identified by the presence of International Classification of Diseases, Tenth Revision diagnosis code Z00.6 (encounter for examination for normal comparison and control in clinical research program).

RESULTS

A total of 105,415 HCC patients were identified. The mean age was 73.5 and the mean Charlson comorbidity index (CCI) score was 8.0. Most patients were male (66.8%), White (76.4%), and located in the Southern region of the United States (37.3%). Among this cohort, 4,773 patients (4.5%) participated in clinical trials. Compared to non-participants, clinical trial participants were younger (73.0 vs. 73.5, p<0.0001) and had a higher mean CCI score (9.0 vs. 8.0, p<0.0001). Race, sex, and census region were also significantly associated with clinical trial participation. Compared to non-participants, a higher proportion of participants were White (82.4% vs. 76.2%, p<0.0001), male (68.9% vs. 66.7%, p<0.05), and located in either the Northeast (25.4% vs. 19.0%, p<0.0001) or the West (25.3% vs. 23.5%, p<0.05).

CONCLUSION

This study provides insights into clinical trial participation among Medicare beneficiaries with HCC, revealing demographic disparities in enrollment. Relative to the general population of HCC patients, White males from the Northeast and West regions of the United States were over-represented in clinical trial participation. Addressing barriers to access and promoting inclusivity in clinical trial recruitment strategies are crucial to ensure study findings are representative of the intended population.

Background

Leukemia, a form of cancer affecting the blood and bone marrow, hinders the normal production of healthy blood cells, leading to health complications.

Methods

In exploring mathematical hypotheses for leukemia, three distinct approaches are proposed. Firstly, an over-dispersed Poisson leukemia regression model is suggested, with the consideration of outliers being addressed through the application of a negative binomial model featuring a non-homogenously distributed mean. Secondly, an eigenfunction, eigendecomposition spatial filter algorithm is introduced, aiming to identify potential leukemia clusters based on hyper/hypo-endemic aggregation/non-aggregation orientations. Lastly, a Bayesian hierarchical model is advocated for determining causation covariates within a non-frequentistic model. This research examined the spatial aggregation of leukemia cases by utilizing sociodemographic data at the zip code level in Hillsborough County, Florida. The investigation involved spatial autocorrelation and Bayesian analyses to pinpoint the covariates linked to the risk of leukemia. The Poissonian regression model revealed a nondispersed paradigm. Hence, we did not need to utilize the negative binomial regression to treat the outliers.

Results

We conducted a second-order eigenfunction eigendecomposition which revealed multiple non-zero autocorrelated clusters throughout various zip codes in Hillsborough County. The hot spots were in 33647, 33578, and 33511 and the cold spots were in 33621, 33503, and 33530. Our proposed approach identifies leukemia hotspots among whites and Asians aged 65+. Urban residential communities in 33647 were most vulnerable to leukemia.

Conclusion

The most common landscape variable associated with leukemia was Urban Residential. By leveraging social media and fostering collaboration between public health and technology sectors, we can work towards achievable leukemia prevention for at-risk populations. Future research should explore the method's applicability at the state level and promote routine blood tests among people.

Background:

Social determinants of health (SDOH) play a critical role in predicting cancer outcomes. Composite indicators, such as the Health Opportunity Index (HOI), have been used to understand health disparities by combining SDOH such as affordability, income, education, and food access. However, traditional methods of computing these indicators, such as Principal Component Analysis (PCA), may not fully capture the complex interactions between SDOH, and may not be able to predict cancer outcomes accurately. Therefore, there is a need for more advanced techniques, such as machine learning, to develop more accurate and comprehensive health indicators that can better capture the relationship between SDOH and cancer outcomes.

Objective:

This study aims to develop a novel composite health indicator using machine learning (ML) and geospatial tools to predict cancer outcomes, with a focus on the role of SDOH.

Method:

Area-level SDOH variables were extracted at the census tract from the Virginia cancer registry and publicly available data sources, such as the American Community Survey, the National Health Interview Survey, the Behavioral Risk Factor Surveillance System. Machine learning techniques were used to identify the most important variables for predicting colorectal and breast cancer patient survival.

Results:

The ML techniques are expected to identify new and important SDOH that have not been captured by the traditional PCA methods. The most important variables in predicting cancer outcomes will be identified and used to compute a novel composite score. Further analyses will be completed by the end of May 2024.

Conclusion:

This investigation can help find novel SDOH that predict cancer outcomes and improve public health policies and interventions. The study will also show how machine learning may improve health indicators and show the limitations of PCA. The insights will assist public health policymakers and stakeholders rethink and create new policy and community intervention initiatives.

Background: Mississippi has the highest mortality rate of breast cancer in the United States. Such extreme mortality rates could be attributed to racial disparity between Black and White populations. This study analyzed the difference in rates of and trends in breast cancer mortality by race in Mississippi.

Methods: This was a retrospective trend analysis of the age-adjusted breast cancer mortality rates for Black and White Mississippians. Data for the 1991-2020 period were obtained from the National Cancer Institute. I conducted analyses using Excel's LINEST function for building linear regression models.

Results: The latest 5-year (2016-2020) age-adjusted mortality rate for White females with breast cancer was 20.0 deaths per 100,000 population (CI, 18.8, 21.1), while the rate for Black females was 30.9 (CI, 28.9, 33.0) in Mississippi. Between 1991 and 2020, the cancer mortality rates decreased by 25.2% for White Mississippians but only by 8.2% for Black Mississippians. The linear regression suggested a statistically significant downtrend in breast cancer mortality for the two study groups. Black women, however, demonstrated a flatter line with a higher deviation from the predicted regression line. This was evident by the smaller slope presented in the beta coefficient among Black women (β = -0.198; F = 10.48; p = 0.003) compared to White women (β = -0.287; F = 109.63; p = <0.00). Additionally, the model explained 80% of the variance in the trend for White women but only 24% for Black women.

Conclusion: The most recent 5-year mortality rate revealed a significant disparity between the black and white populations of Mississippi, a gap that has been growing over the last few decades. This was observed by the flattened line with a high variance experienced by the Black women compared to the steeper slope with a lower variance shown by the White women.

Structural barriers to health equity can significantly impact the incidences and mortalities of screenable cancers due to inequitable access to screening services. This study applied geospatial analytics to examine the social determinants of health (SDOH) and the outcomes (incidences and mortalities) of five screenable cancers, including cervical, colorectal, metastatic breast, prostate, and lung, associated with present-day census tracts in Miami-Dade County that were designated grades C (definitely declining) and D (hazardous) by the Home Owners' Loan Corporation (HOLC) in the 1930s.

Geocoded data for historic redline zoning in Miami were downloaded from the County’s website. Spatial-join was performed to identify present-day census tracts at historically redlined locations. Age-adjusted incidences and mortality rates for the five screenable cancers were calculated with 2010-2020 Florida cancer registry data for all census tracts in the County. SDOH data for all census tracts were obtained from the American Community Survey. Average cancer incidences, mortality rates, and SDOH variables for the redlined census tracts were compared with those for the non-redlined to identify specific disparities associated with neighborhoods subject to the historic practice of redline zoning.

Census tracts at redlined neighborhoods are currently occupied by mostly non-Hispanic black and have significantly higher average age-adjusted incidences and mortality rates for the five screenable cancers than those not redlined. Striking SDOH disparities in the redlined neighborhoods were identified as the average median household income for the redlined is only $26,000 with a 34% poverty rate, compared to $55,000 and 17% for those not redlined.

Lack of access to health insurance is clearly a factor for the higher incidences and mortalities of screenable cancers in the redlined tracts that has a 33% percent uninsured rate, compared to 23% in the counterpart tracts. Providing equitable access to screening services for these cancers is urgently needed for addressing cancer outcome disparities.

Background: African Americans/Blacks face significantly higher lung cancer rates than other racial/ethnic minority groups, yet effective interventions are lacking. Engaging stakeholders is crucial for identifying impactful strategies. This study aims to collaborate with stakeholders to identify and prioritize interventions that hold promise for reducing lung cancer disparities, employing deliberative democracy principles to ensure community involvement and empowerment in the decision-making process.

Method: A one-day 6-hour workshop convened 17 key stakeholders to develop interventions promoting lung cancer screening in Broward County, Florida African-American/Black communities. Utilizing the Nominal Group Technique (NGT), stakeholders engaged in silent idea generation, round-robin idea sharing, discussion to consolidate overlapping interventions, and voting on top ideas. Selected interventions were further evaluated for effectiveness, feasibility, cost, sustainability, political will, and unintended consequences using a decision matrix model. The highest-ranked idea was further detailed through solution mapping, outlining goals, implementation process, and key stakeholders.

Results: The 25 interventions identified were categorized into three themes: coordination of care, access to care, and educational awareness. The top interventions favored by stakeholders were: training community health workers (CHWs) on lung cancer screening navigation, initiating a lung cancer educational campaign, hosting a community health resource fair, and advocating for policy and environmental changes regarding eligibility criteria for lung cancer screening. Utilizing the decision matrix model, training CHWs for lung cancer screening navigation emerged as the top intervention across all six evaluation categories, with CHWs broadly defined as community leaders providing health services or education.

Conclusion: Our findings demonstrate the effectiveness of the NGT in eliciting valuable input and consensus among stakeholders, highlighting the feasibility and value of this approach in addressing complex public health challenges. Future research should focus on the development and evaluation of these interventions to ensure their effectiveness and sustainability in addressing health inequities in lung cancer.