Introduction

Purpose of the study

To understand safety and efficacy associated with novel cancer therapies by:

  1. comparing adequate reporting between premarketing and postmarketing studies, and

  2. characterizing representation in the FDA-approved studies by comparing with cancer population.

Methods

Overall methods

  • Cross-sectional

  • Novel cancer therapeutics studies approved by FDA between 2012-2016.

  • Study demographic information were obtained from publicly available sources.

  • U. S. Cancer Population demographics were obtained from U.S. Cancer Statistics.

  • Calculate representation using participation to prevalence ratio (PPR)

Identify therapeutics and studies

Manual review/abstraction of yearly published Center for Drug Evaluation and Research New Molecular Entity Drug and Original Biologic Approvals list (two parts):

  • Drugs: Drug Approval Reports by Month

Identify therapeutics and studies (continued)

  • Biologics: Annual Biological License Application Approvals

Manual abstraction of study demographics

  1. Medical review documents from FDA (drug-only)

  2. Indexed publications from the clinical trial entry on ClinicalTrials.gov

  3. Sponsors’ websites (clinical trials synopsis)

  4. Scopus database for publications describing the study

Source of population data

2012 to 2016 U.S. Cancer Statistics data set (two parts):

  1. Cancer registry data from the National Program of Cancer Registries

  2. National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) data set

Data not reported in the paper. But, can be accessed through SEER*Stats:

Comparison between study and population (representation)

Representation of age, gender, racial/ethnic minority group

Participation to prevalence ratio (PPR) calculated to compare demographic composition of patients in trials relative to the disease population

(E.g., percentage of female among lung cancer trials / percentage lung cancer patients in the US that are female)

  • Adequate representation: PPR = [0.8 – 1.2]

  • Under-representation: PPR < 0.8

  • Over-representation Over: PPR > 1.2

Results

Availability of demographic data from the identified studies

Among the 77 premarketing studies and 56 postmarketing studies:

Fig.1

Some demographics data are hard to obtain, at least by using their methods.

Representation older populations in the studies

Across cancer sites, older patients underpresented with mean PPR [95% CI] = 0.73 [0.72-0.74] in premarkeing studies and 0.75 [0.75-0.76] in post marketing studies:

Fig.3

Representation Black populations in the studies

Across cancer sites, black patients underrepresented in premarketing (0.32 [0.31-0.32]) and postmarketing (0.21 [0.20-0.22]) studies.

Fig.3b

Limitations

  • Missing demographcis from several studies

  • Small sample size

  • Did not consider genetic factors that may contribute to the racial composition of the trial participants (e,g., EGFR mutations being more prevalent among Asian)

  • Did not exclude trials outside of US

References

  • Varma, T., Wallach, J. D., Miller, J. E., Schnabel, D., Skydel, J. J., Zhang, A. D., Dinan, M. A., Ross, J. S., & Gross, C. P. (2021). Reporting of Study Participant Demographic Characteristics and Demographic Representation in Premarketing and Postmarketing Studies of Novel Cancer Therapeutics. JAMA network open, 4(4), e217063. https://doi.org/10.1001/jamanetworkopen.2021.7063

  • United States cancer statistics. US Centers for Disease Control and Prevention. Accessed July 15, 2020. https:// www.cdc.gov/cancer/uscs/index.htm

  • Centers for Disease Control and Prevention. Cancer stat facts: soft tissue including heart cancer. National Cancer Institute Surveillance, Epidemiology, and End Results Program. Accessed July 21, 2020. https://seer.cancer. gov/statfacts/html/soft.html