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.

How Have Experimental Cancer Interventions Evolved over Time?

We performed a trends analysis of experimental cancer interventions. The complete records of 32,623 interventional neoplasm clinical trials involving 454 types of neoplasms from 2000 to 2017 were downloaded from the AACT database. The conditions and drug concepts were normalized using MetaMap. The normalized frequencies (NF) for each type of intervention were calculated and compared. Among 95,440 interventions, 77.4% were drugs, 5.3% were radiation, 6.6% were surgery and 10.6% were other therapies. Among 47,754 arms, 82.8% were mono-type interventions and 17.2% were multi-type interventions. Among 73,889 drug interventions, immunologic factor drugs increased rapidly over the last five years. Both breast cancer and pancreatic cancer have been testing new drugs in clinical trials; however, more drugs have been tested in phase 3 or 4 trials and employed in comparator arms for breast cancer compared to pancreatic cancer. Breast cancer trials showed a more even drug NF distribution than pancreatic cancer trials. The JS Distance among three periods (2000-05 vs. 2006-11 vs. 2012-17) showed unidirectional research progress trend for breast cancer, but reverse trend for pancreatic cancer. This study contributes a large-scale landscape overview of the trends in cancer experimental interventions and a methodology for using public clinical trial summaries for understanding the evolving cancer research.