Developing a model to predict accrual to cancer clinical trials: Data from an NCI designated cancer center

Successful completion of large, low-cost randomized cancer trials at an academic cancer center

BACKGROUND

Concerns about low accrual have long been a standard part of the discourse on cancer clinical trials, reaching even as far as the news media. Indeed, so many trials are closed before completing accrual that a cottage industry has recently developed creating statistical models to predict trial failure. We previously proposed four methodologic fixes for the current crisis in clinical trials: (1) dramatically reducing the number of eligibility criteria, (2) using data routinely collected in clinical practice for trial endpoints; then lowering barriers to accrual by (3) cluster randomization or (4) staged consent.

METHODS

We report our practical experience of applying these fixes to randomized trials at Memorial Sloan Kettering Cancer Center.

RESULTS

We have completed seven single-center randomized trials, with several more underway and accruing rapidly, with a total accrual approaching 10,000. Many of the trials have compared surgical interventions, an area where trials have traditionally been hard to complete. Only one of these trials was externally funded. While low costs were possible due to the existing research infrastructure at our institution, such infrastructure is common at major cancer centers.

CONCLUSIONS

Further research on innovative clinical trial designs is warranted, particularly in higher-stakes settings, and in trials of medical and radiotherapy interventions.

Uncovering key clinical trial features influencing recruitment

Background

Randomized clinical trials (RCT) are the foundation for medical advances, but participant recruitment remains a persistent barrier to their success. This retrospective data analysis aims to (1) identify clinical trial features associated with successful participant recruitment measured by accrual percentage and (2) compare the characteristics of the RCTs by assessing the most and least successful recruitment, which are indicated by varying thresholds of accrual percentage such as ≥ 90% vs ≤ 10%, ≥ 80% vs ≤ 20%, and ≥ 70% vs ≤ 30%.

Methods

Data from the internal research registry at Columbia University Irving Medical Center and Aggregated Analysis of ClinicalTrials.gov were collected for 393 randomized interventional treatment studies closed to further enrollment. We compared two regularized linear regression and six tree-based machine learning models for accrual percentage (i.e., reported accrual to date divided by the target accrual) prediction. The outperforming model and Tree SHapley Additive exPlanations were used for feature importance analysis for participant recruitment. The identified features were compared between the two subgroups.

Results

CatBoost regressor outperformed the others. Key features positively associated with recruitment success, as measured by accrual percentage, include government funding and compensation. Meanwhile, cancer research and non-conventional recruitment methods (e.g., websites) are negatively associated with recruitment success. Statistically significant subgroup differences (corrected p-value < .05) were found in 15 of the top 30 most important features.

Conclusion

This multi-source retrospective study highlighted key features influencing RCT participant recruitment, offering actionable steps for improvement, including flexible recruitment infrastructure and appropriate participant compensation.

Overcoming racial disparities in cancer clinical trial enrollment of Asians and Native Hawaiians

Background

Asians and Native Hawaiians are two of the fastest growing minority populations in the United States, however these racial minority groups are severely underrepresented in clinical trials. This study looks at cancer clinical trial accrual among Asians and Native Hawaiians in a community-based network with a mission of increasing minority accrual to studies.

Methods

The University of Hawaii Cancer Center (UHCC) network enrolls patients to treatment and non-treatment cancer studies. Enrollment on studies opened between 2009 and 2013 were obtained from UHCC's clinical trial management system. Incidence of cancer by race was acquired from the Hawaii Tumor Registry. Enrollment fractions were compared for the most common races in the state: White, Asian (specifically Chinese, Filipino, Japanese), and Native Hawaiian.

Results

Whites comprised the largest proportion of cancer patients and participants in trials. Asians and Native Hawaiians were enrolled into cancer clinical trials at the same or higher enrollment fraction compared to Whites. Chinese, Japanese, and Native Hawaiian patients participated in treatment trials significantly more often than Whites (p < 0.05). Similarly, Chinese and Native Hawaiians enrolled in non-treatment trials at a significantly higher rate compared to Whites (p < 0.05).

Conclusions

The UHCC network has instituted many strategies to increase minority accrual that have likely led to Asian and Native Hawaiian patients participating in studies at least as often as White patients. The strategies implemented at UHCC may benefit similar communities with a high number of minority cancer patients.

A Web-Based Intervention to Increase Smokers' Intentions to Participate in a Cessation Study Offered at the Point of Lung Screening: Factorial Randomized Trial

BACKGROUND

Screen ASSIST is a cessation trial offered to current smokers at the point of lung cancer screening. Because of the unique position of promoting a prevention behavior (smoking cessation) within the context of a detection behavior (lung cancer screening), this study employed prospect theory to design and formatively evaluate a targeted recruitment video prior to trial launch.

OBJECTIVE

The aim of this study was to identify which message frames were most effective at promoting intent to participate in a smoking cessation study.

METHODS

Participants were recruited from a proprietary opt-in online panel company and randomized to a 2 (benefits of quitting vs risks of continuing to smoke at the time of lung screening; BvR) × 2 (gains of participating vs losses of not participating in a cessation study; GvL) message design experiment (N=314). The primary outcome was self-assessed intent to participate in a smoking cessation study. Message effectiveness and lung cancer risk perception measures were also collected. Analysis of variance examined the main effect of the 2 message factors and a least absolute shrinkage and selection operator (LASSO) approach identified predictors of intent to participate in a multivariable model. A mediation analysis was conducted to determine the direct and indirect effects of message factors on intent to participate in a cessation study.

RESULTS

A total of 296 participants completed the intervention. There were no significant differences in intent to participate in a smoking cessation study between message frames (P=.12 and P=.61). In the multivariable model, quit importance (P<.001), perceived message relevance (P<.001), and affective risk response (ie, worry about developing lung cancer; P<.001) were significant predictors of intent to participate. The benefits of quitting frame significantly increased affective risk response (Meanbenefits 2.60 vs Meanrisk 2.40; P=.03), which mediated the relationship between message frame and intent to participate (b=0.24; 95% CI 0.01-0.47; P=.03).

CONCLUSIONS

This study provides theoretical and practical guidance on how to design and evaluate proactive recruitment messages for a cessation trial. Based on our findings, we conclude that heavy smokers are more responsive to recruitment messages that frame the benefits of quitting as it increased affective risk response, which predicted greater intention to participate in a smoking cessation study.

Eligibility in Cancer Clinical Research: The Intersection of Discovery, Generalizability, Beneficence, and Justice

Eligibility criteria in clinical trials limit the study population for safety and scientific purposes. The American Society of Clinical Oncology and The Friends of Cancer Research collaboration reconsidered common eligibility criteria in cancer trials and found many to be unnecessarily restrictive. The current recommendations further their efforts to facilitate accrual and improve the generalizability of research results to practice.See related articles, p. 2394, 2400, 2408, 2416, 2424, and 2430.