Triple blinded prospective study assessing the impact of genomics: EndoPredict and artificial intelligence Watson for Oncology (WFO) on MDT’s decision of adjuvant systemic therapy for hormone receptor positive early breast carcinoma.

Authors

S.P. Somashekhar

S.P. Somashekhar

Manipal Comprehensive Cancer Center, Bangalore, India

S.P. Somashekhar , Ramya Yethadka , Rohit Kumar C , Ashwin K Rajgopal , Amit Rauthan , Poonam Patil

Organizations

Manipal Comprehensive Cancer Center, Bangalore, India, Manipal Comphrensive Cancer Center, Bangalore, India, Manipal Hospital, Bangalore, India

Research Funding

Other

Background: : Decision on adjuvant systemic therapy in hormone positive early breast carcinoma is the only grey area in breast carcinoma management. This study was done to investigate the concordance between the results of genomic test, artificial intelligence and tumor board decision and implications of the same in clinical practice. Methods: This was a triple blinded, prospective study. Decision regarding the adjuvant systemic therapy was done by the multidisciplinary tumor board (MDT)after reviewing the pathology reports & the results correlated with Endopredict test reports & artificial intelligence(Watson for Oncology). Results: Total of 42 patients included. Mean age was 58.3 years, 71.4% were post-menopausal. Breast conservation was done in 47.6%. 64.2% were T1-2N0 stage. Infiltrating ducal carcinoma was major type (83.3%). Decision by MDT to give adjuvant chemotherapy was for 25 patients (59.5%) & hormonal therapy for rest. Recommendation by Watson for oncology was to give adjuvant chemotherapy in 50%. Endopredict score (EPclin) resulted in a low-risk group of 22 patients (52.3%), while 15(47.6%) had a high risk EPclin score. Discordance between the endopredict test, Watson & tumor board was for 11 patients (26.1%): 3 patients had high risk score, but the tumor board decision was to give hormonal therapy due to the age factor. 8 patients had low risk score, but tumor board decision was to give adjuvant chemotherapy. Extremes of age, premenopausal status, intermediate grade & high Ki 67% values were the factors associated with discordance. The treatment decision changed for 4 patients (4/11, 36%) after reviewing the endopredict test and Watson recommendation. Conclusions: Tumor board decision can be more scientific & evidence based with the help of genomics & a learnt colleague in the form of Watson for Oncology. Even though the clinical experience is the important determinant of adjuvant therapy, genomic test with artificial intelligence, which includes the scientific evidence, will guide in decision making. Long term follow up is needed for the validation in our clinical setting.

Disclaimer

This material on this page is ©2024 American Society of Clinical Oncology, all rights reserved. Licensing available upon request. For more information, please contact licensing@asco.org

Abstract Details

Meeting

2019 ASCO Annual Meeting

Session Type

Publication Only

Session Title

Health Services Research, Clinical Informatics, and Quality of Care: Publication Only

Track

Quality Care/Health Services Research

Sub Track

Care Delivery/Models of Care

Citation

J Clin Oncol 37, 2019 (suppl; abstr e18013)

DOI

10.1200/JCO.2019.37.15_suppl.e18013

Abstract #

e18013

Abstract Disclosures