TL;DR
This study analyzed over two decades of lung cancer clinical trial enrollment data, revealing significant improvements in the representation of elderly patients with non-small-cell lung cancer (NSCLC) and women. However, disparities persist for elderly patients with small-cell lung cancer (SCLC) and racial/ethnic minorities, highlighting the need for targeted interventions to ensure trial results are generalizable to the broader lung cancer patient population.
What Problem Does This Paper Address?
Clinical trials are crucial for advancing cancer treatment, but their results are only truly valuable if they reflect the diversity of the patient population. Historically, certain groups, including the elderly, women, and racial/ethnic minorities, have been under-represented in these trials. This under-representation can lead to treatments that are less effective or have different side effect profiles in these populations, limiting the generalizability of research findings. This paper specifically addresses the long-term trends and persistent disparities in the enrollment of these demographic groups in National Cancer Institute (NCI)-sponsored lung cancer clinical trials from 1990 to 2012, comparing trial participants to the broader US lung cancer population.
Methods
The researchers employed a longitudinal, comparative analysis. They utilized data from two primary sources: NCI-cooperative group lung cancer trials and the Surveillance, Epidemiology, and End Results (SEER) registry. To quantify enrollment disparities, they calculated two metrics: the Enrollment Disparity Difference (EDD), which is the absolute difference in subgroup proportion between the trial population and the US population, and the Enrollment Disparity Ratio (EDR), which is the ratio of these proportions. They then analyzed trends over time using joinpoint regression to calculate the Annual Percentage Change (APC) for subgroup proportions in both datasets and to test for significant changes in these trends. This approach allowed them to assess not only the existence of disparities but also whether they were improving or worsening over the 22-year study period.
Datasets
- National Cancer Institute (NCI) Cooperative Group Lung Cancer Trials: Data from 23,006 patients enrolled in 131 NCI-sponsored lung cancer treatment trials activated between 1990 and 2010, with accruals through December 2012.
- Surveillance, Epidemiology, and End Results (SEER) Registry: Data from 578,476 patients diagnosed with lung cancer in the US during the period 1990 to 2012, used as a benchmark for the general US lung cancer population.
Key Insights and Ideas
- The NIH Revitalization Act of 1993 had a tangible impact: The study suggests that the mandate to include women and minorities in NIH-funded research likely contributed to the observed improvements in the enrollment of women in lung cancer trials.
- Not all elderly patients are the same: The improvement in elderly enrollment was specific to NSCLC. Elderly patients with SCLC showed no improvement, indicating that disease subtype is a critical factor in enrollment disparities.
- Distant disease enrollment is catching up for women: The improved representation of women in trials was particularly driven by those with distant NSCLC, suggesting that newer therapies or trial designs might be more appealing or accessible to this group.
- Racial/ethnic minorities remain a significant challenge: Despite overall improvements in some areas, the under-representation of Black, Asian/Pacific Islander, and Hispanic patients in lung cancer trials persisted throughout the study period.
- The choice of metrics matters: Using both EDD and EDR provides a more nuanced understanding of disparities, capturing both absolute differences and relative over/under-representation.
Key Results
- Elderly patients with NSCLC: Enrollment disparity improved significantly (test of parallelism, P = .020). By 2010-2012, the EDD was 0.22 and EDR was 1.65, indicating continued, though reduced, under-representation.
- Elderly patients with SCLC: No improvement in enrollment disparity was observed (APC = 0.20, P = .714), despite an increase in the proportion of elderly SCLC patients in the US population. Disparity remained high in 2010-2012 (EDD = 0.35, EDR = 3.82).
- Women with lung cancer: Overall enrollment disparity closed by 2010-2012 (EDD = 0.03, EDR = 1.07), showing significant improvement.
- Racial/ethnic minorities: Enrollment disparities persisted for Black, Asian/Pacific Islander, and Hispanic patients without significant improvement over the study period.
- Targeted Agents: In trials with targeted agents (2001-2012), elderly patient enrollment was similar to non-targeted trials, but women were more represented in erlotinib trials (55.6%) compared to non-erlotinib trials (43.4%).
Limitations and Future Work
The authors acknowledge several limitations:
- Restriction to NCI-cooperative group trials: This excludes trials sponsored by industry, non-profit organizations, or other NCI-funded non-cooperative groups, potentially limiting the generalizability of findings to all lung cancer trials.
- Small sample sizes for certain minority groups: This limited the power to detect enrollment disparities for groups like American Indians/Alaska Natives.
- Reliance on SEER data: While widely used, SEER data represents a sample of all lung cancer cases.
Future work should focus on:
- Understanding barriers: Deeper investigation into why disparities persist for elderly SCLC patients and racial/ethnic minorities is crucial. This includes exploring barriers to trial awareness, access, and participation.
- Targeted interventions: Developing and implementing strategies to specifically address the under-representation of these groups, such as culturally competent patient navigation and community-based outreach.
- Subgroup analysis: Continuing to examine enrollment trends in more refined subgroups (e.g., by specific cancer subtypes, genetic mutations, or extent of disease) to inform personalized medicine approaches and trial design.
- Elderly-specific trials: Continuing to design and conduct trials tailored for older adults, potentially incorporating comprehensive geriatric assessments.
Summary updated