Chiara Corti1, Marisa Cobanaj2, Federica Marian3, Edward C Dee4, Maxwell R Lloyd5, Sara Marcu6, Andra Dombrovschi7, Giorgio P Biondetti8, Felipe Batalini9, Leo A Celi10, Giuseppe Curigliano11. 1. Division of New Drugs and Early Drug Development for Innovative Therapies, European Institute of Oncology, IRCCS, Milan, Italy; Department of Oncology and Haematology (DIPO), University of Milan, Milan, Italy. Electronic address: chiara.corti@ieo.it. 2. Department of Electronics Informatics and Bioengineering, Polytechnic University of Milan, Milan, Italy. 3. DaVinci Healthcare, Milan, Italy. 4. Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, USA. 5. Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA 02215, USA. 6. S.A.T.E. Systems and Advanced Technologies Engineering, Venice, Italy. 7. synbrAIn, Milan, Italy. 8. OM1, Inc., Boston, MA, USA. 9. Women's Cancer Program, Mayo Clinic Cancer Center, Phoenix, AZ, USA. 10. Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA 02215, USA; Laboratory for Computational Physiology, Massachusetts Institute of Technology, Cambridge, MA 02139, USA; Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA 02115, USA. 11. Division of New Drugs and Early Drug Development for Innovative Therapies, European Institute of Oncology, IRCCS, Milan, Italy; Department of Oncology and Haematology (DIPO), University of Milan, Milan, Italy.
Abstract
BACKGROUND: Artificial intelligence (AI) has the potential to personalize treatment strategies for patients with cancer. However, current methodological weaknesses could limit clinical impact. We identified common limitations and suggested potential solutions to facilitate translation of AI to breast cancer management. METHODS: A systematic review was conducted in MEDLINE, Embase, SCOPUS, Google Scholar and PubMed Central in July 2021. Studies investigating the performance of AI to predict outcomes among patients undergoing treatment for breast cancer were included. Algorithm design and adherence to reporting standards were assessed following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement. Risk of bias was assessed by using the Prediction model Risk Of Bias Assessment Tool (PROBAST), and correspondence with authors to assess data and code availability. RESULTS: Our search identified 1,124 studies, of which 64 were included: 58 had a retrospective study design, with 6 studies with a prospective design. Access to datasets and code was severely limited (unavailable in 77% and 88% of studies, respectively). On request, data and code were made available in 28% and 18% of cases, respectively. Ethnicity was often under-reported (not reported in 52 of 64, 81%), as was model calibration (63/64, 99%). The risk of bias was high in 72% (46/64) of the studies, especially because of analysis bias. CONCLUSION: Development of AI algorithms should involve external and prospective validation, with improved code and data availability to enhance reliability and translation of this promising approach. Protocol registration number: PROSPERO - CRD42022292495.
BACKGROUND: Artificial intelligence (AI) has the potential to personalize treatment strategies for patients with cancer. However, current methodological weaknesses could limit clinical impact. We identified common limitations and suggested potential solutions to facilitate translation of AI to breast cancer management. METHODS: A systematic review was conducted in MEDLINE, Embase, SCOPUS, Google Scholar and PubMed Central in July 2021. Studies investigating the performance of AI to predict outcomes among patients undergoing treatment for breast cancer were included. Algorithm design and adherence to reporting standards were assessed following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement. Risk of bias was assessed by using the Prediction model Risk Of Bias Assessment Tool (PROBAST), and correspondence with authors to assess data and code availability. RESULTS: Our search identified 1,124 studies, of which 64 were included: 58 had a retrospective study design, with 6 studies with a prospective design. Access to datasets and code was severely limited (unavailable in 77% and 88% of studies, respectively). On request, data and code were made available in 28% and 18% of cases, respectively. Ethnicity was often under-reported (not reported in 52 of 64, 81%), as was model calibration (63/64, 99%). The risk of bias was high in 72% (46/64) of the studies, especially because of analysis bias. CONCLUSION: Development of AI algorithms should involve external and prospective validation, with improved code and data availability to enhance reliability and translation of this promising approach. Protocol registration number: PROSPERO - CRD42022292495.