Why location matters: Availability of cancer clinical trials

The majority of patients seek care at community oncology sites ¹ ; however, most trials are available at National Cancer Institute (NCI)-designated sites. While the NCI National Cancer Oncology Research Program (NCORP) was designed to address this problem ² , little is known about the county-level demographics of NCORP sites locations. Questions remain about where NCORP and NCI sites are located, including area-level environmental, societal, logistical, and financial factors that may facilitate individuals’ ability to access these sites. NCORP sites are located in the US (including Guam and Puerto Rico) with a total of seven research bases, 46 Community Sites (14 are Minority/Underserved Sites), and 1029 participating hospitals (which may include multiple clinics within a hospital). There are 63 NCI-designated cancer centers within the US, which conduct the majority of clinical trials. ³ 

The objectives of our study were to understand the associations between county-level demographics and availability of both NCORP and NCI sites. GIS is appropriate for this question in order to visualize where these sites are located on a map and in relation to other sites. This Story Map is a supplement to the following publication: Why location matters: Associations between county-level demographics and availability of National Cancer Institute Community Oncology Research Program and National Cancer Institute Sites by Nicole E. Caston, MPH; Courtney P. Williams, DrPH; Emily B. Levitan, ScD; Russell Griffin, PhD; Andres Azuero, PhD, MBA; Stephanie B. Wheeler, PhD, MPH; Gabrielle B. Rocque, MD, MSPH. The manuscript is in preparation to be submitted to Journal of Clinical Oncology.

Previous studies have assessed availability of specific clinical trials within specific cancer types; however, our study evaluated access to sites that have the resources (i.e., financial, staffing, medical equipment) to conduct a clinical trial via their NCORP or NCI status. Wang et al found that US counties with higher proportions of African Americans are less likely to have access to any prostate cancer clinical trials. ⁴  Additionally, Grant et al assessed the association between Social Vulnerability Index themes and availability of multiple myeloma trials within North Carolina and found similar results to Wang et al. ⁵ 

This cross-sectional study utilized publicly available data of location of NCORP and NCI sites and county-level data for the US. Inclusion criteria included counties within the 50 states and the District of Columbia.

Availability of an NCORP and NCI sites: Individual counties were considered having access to an NCORP or NCI site if a site location was in the corresponding county. NCORP site locations were taken from the NCORP website, which lists all NCORP sites and their addresses. ³  NCI sites were taken from the NCI site county ShapeFile dataset. ⁶ 

County-level demographics and characteristics were abstracted from the Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry’s (CDC/ATSDR) Social Vulnerability Index (SVI). ⁷  The SVI uses 2020 US Census data to determine the communities’ social vulnerability: factors that could weaken a community’s ability to prevent loss, of both human and financial, following a disaster. Sixteen variables make up four themes: 1) socioeconomic status; 2) household characteristics; 3) racial & ethnic minority status; and 4) housing type & transportation. Finally, the overall vulnerability theme combines all four themes. All themes are scored from 0 to 1, which are percentile rankings. For each theme, higher scores represent higher vulnerability.

Overall vulnerability theme: According to the CDC/ATSDR documentation, those counties with scores 0.90-1.0 are considered the highest vulnerable counties; 90% of counties are less vulnerable and 10% are more vulnerable. We used this logic to dichotomize counties. The following variables are included for each theme: 

• Socioeconomic status theme: Below 150% poverty, unemployed, housing cost burden, no high school diploma, and no health insurance.
• Household characteristics theme: Aged 65 & older, aged 17 & younger, civilians with a disability, single-parent households, and English language proficiency.
• Racial & ethnic minority status theme: Hispanic or Latino (of any race); Not Hispanic or Latino for Black and African American, American Indian and Alaska Native, Asian, Native Hawaiian and Other Pacific Islander, two or more races, other races.
• Housing type & transportation theme: Multi-unit structures, mobile homes, crowding, no vehicle, and group quarters.  

Rural-Urban Continuum Codes (RUCC): We included 2013 RUCC from the Area Health Resource File dataset for each county. RUCC scores range from 1-9 with scores 1-3 representing metropolitan areas, 4-6 representing suburban areas, and scores 7-9 representing rural areas. Additional information on household with a broadband internet subscription was abstracted from the SVI dataset.

First, we geocoded NCORP and NCI sites to the map. Next, we included information from the Social Vulnerability Index dataset. The overall theme was dichotomized into the highest vulnerable counties (upper 10% vs other 90%). The subthemes were split into four classes each. However, for the models, the subthemes were kept continuous and the model also contained RUCC. Geocoding was performed using ArcGIS software. We also ran modified Poisson regression models with robust standard errors estimating risk ratios (RR) and 95% confidence intervals (CI) to assess the relationship between SVI themes and availability of both NCORP and NCI sites. These analyses were performed using SAS© software, version 9.4 (SAS Institute, Cary, NC).

We found that the majority of counties do not have access to sites that have the resources and infrastructure to offer cancer clinical trials. NCORP and NCI sites appear to be appropriately located in singular counties that serve racially and ethnically diverse populations. For NCORP sites, this could do be due to NCORP's Minority/Underserved Community Sites. However, as counties become more vulnerable according to the socioeconomic and household characteristics themes, there was a lower likelihood of NCORP and NCI site availability. Socioeconomic status is associated with place, race, insurance status, education, income, and employment which affect cancer outcomes. ⁸  Additionally, individuals living in lower SES locations present to clinic with more advanced cancers which is associated with poorer cancer outcomes. ^9-11 

Results from our study point to many issues so many programs face: that we are not reaching the truly vulnerable places. We call for the NCORP network to expand their inclusion to specific sites. According to Carlos et al, 46 NCORP sites are independent community practices, health system-affiliated practices, and safety-net hospitals, ¹²  therefore, the ability to work with academic medical centers with satellite sites is feasible.

This study should be considered in light of limitations. As this uses county-level data, there may be ecological fallacies. Additionally, county-level data may not be generalizable to other geographical levels nor the individual-level experiences. As this is an exploratory analysis, we were unable to ascertain a causal link between county-level data and inclusion of NCORP or NCI sites. Further information would need to be abstracted to understand when sites were included in NCORP or were designated NCI Cancer Centers, which is outside of the scope of this research. Also, we were unable to determine how the counties with access multiple NCORP and/or NCI sites differ from counties with one site and if there is a “dose” effect.

1.            Petrelli NJ. A community cancer center program: getting to the next level. J Am Coll Surg. Mar 2010;210(3):261-70. doi:10.1016/j.jamcollsurg.2009.11.015

2.            National Cancer Institute: Community Oncology Research Program (NCORP). Sites by State. Accessed on February 28, 2023.  https://ncorp.cancer.gov/findasite/components.php . .

3.            DelNero PF, Buller ID, Jones RR, et al. A National Map of NCI-Designated Cancer Center Catchment Areas on the 50th Anniversary of the Cancer Centers Program. Cancer Epidemiol Biomarkers Prev. May 4 2022;31(5):965-971. doi:10.1158/1055-9965.EPI-21-1230

4.          Wang WJ, Ramsey SD, Bennette CS, Bansal A. Racial Disparities in Access to Prostate Cancer Clinical Trials: A County-Level Analysis. JNCI Cancer Spectr. Feb 2022;6(1)doi:10.1093/jncics/pkab093

5.          Grant SJ, Jansen M, Kuo TM, et al. Cross-Sectional Analysis of Clinical Trial Availability and North Carolina Neighborhood Social Vulnerability. JCO Oncol Pract. Dec 6 2022:OP2200325. doi:10.1200/OP.22.00325

6.          National Cancer Institute: GIS Portal for Cancer Research. Catchment Areas of NCI-Designated Cancer Centers. Assessed on Feburary 28, 2023.  https://gis.cancer.gov/ncicatchment/ .

7.          Centers for Disease Control and Prevention/ Agency for Toxic Substances and Disease Registry/ Geospatial Research, Analysis, and Services Program. CDC/ATSDR Social Vulnerability Index 2020 Database US.  https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html . Accessed on January 17, 2023.

8.          Coughlin SS. Social determinants of breast cancer risk, stage, and survival. Breast Cancer Res Treat. Oct 2019;177(3):537-548. doi:10.1007/s10549-019-05340-7

9.          Henry KA, Boscoe FP, Johnson CJ, Goldberg DW, Sherman R, Cockburn M. Breast cancer stage at diagnosis: is travel time important? J Community Health. Dec 2011;36(6):933-42. doi:10.1007/s10900-011-9392-4

10.          Nguyen-Pham S, Leung J, McLaughlin D. Disparities in breast cancer stage at diagnosis in urban and rural adult women: a systematic review and meta-analysis. Ann Epidemiol. Mar 2014;24(3):228-35. doi:10.1016/j.annepidem.2013.12.002

11.          Ayanian JZ, Kohler BA, Abe T, Epstein AM. The relation between health insurance coverage and clinical outcomes among women with breast cancer. New England Journal of Medicine. 1993;329(5):326-331.

12.          Carr E. Access to Care. Clin J Oncol Nurs. Oct 1 2018;22(5):475. doi:10.1188/18.Cjon.475