P1.16-14 Effects of an Artificial Intelligence (AI) System on Clinical Trial Enrollment in Lung Cancer

K Leventakos,A Schwecke,J Molina,T Bleeker,C Hocum,E Deering,A Adjei,J Helgeson,M Rammage,T Halfdanarson,R Marks,A Preininger,G Purcell Jackson,S Coverdill,K Parikh,K Pomerleau,A Mansfield,T Haddad

doi:10.1016/j.jtho.2019.08.1240

K Leventakos, A Schwecke + Show 16 more

Open Access

PDF Available

https://doi.org/10.1016/j.jtho.2019.08.1240

Copy DOI

Export

Save

Cite

Abstract
Full-Text PDF
Similar Papers

Abstract

Listen

Clinical trials offer cancer patients access to the latest promising therapies and improve the overall quality and safety of cancer care. However, few patients participate, due in part to the time and effort associated with traditional manual ad hoc screening methods. IBM Watson Health’s Clinical Trials Matching (CTM) is an artificial intelligence (AI) system that employs natural language processing to abstract patient and trial data from unstructured sources and machine learning to match patients to trials. This study compared the clinical trial enrollment rates of lung cancer patients before and after deployment of CTM. CTM was trained for lung cancer and adopted in an academic outpatient oncology clinic using a phased implementation approach beginning July 2018. Clinical trials included ∼42 therapeutic, supportive care, and observational trials. Clinical research coordinators validated Watson-derived clinical trial matches on the day prior to patient clinic visits. Oncologists were provided with a list of potentially eligible trials for each patient to facilitate evaluation at point of care. The average monthly enrollment rates for therapeutic trials were compared 6 months before and after CTM deployment. Average enrollment rates per active clinical trial were reported. Clinical trial matches were validated and delivered to lung oncology providers in 69% (1818/2637) of patients’ visits during the 6-month phased implementation. Enrollment of patients in lung cancer therapeutic clinical trials occurred at a rate of 3.83 patients/month after CTM deployment, as compared to 1.83 patients/month prior to CTM deployment; a 109% enrollment increase. When adjusted for the average number of active clinical trials before (30) and after (39) CTM implementation, the enrollment was 0.097 patients/trial using CTM, compared to 0.061 patients/trial using traditional methods; a 58.4% enrollment increase. Use of IBM Watson Health’s Clinical Trials Matching (CTM) system with screening coordinators facilitated an increase in clinical trial enrollment and promoted awareness of clinical trial opportunities within the lung oncology practice.

Full Text