Hava Izci1, Tim Tambuyzer2, Krizia Tuand3, Victoria Depoorter1, Annouschka Laenen4, Hans Wildiers1,5, Ignace Vergote1,6, Liesbet Van Eycken2, Harlinde De Schutter2, Freija Verdoodt2, Patrick Neven1,6. 1. Department of Oncology, KU Leuven - University of Leuven, Leuven, Belgium. 2. Research Department, Belgian Cancer Registry, Brussels, Belgium. 3. KU Leuven Libraries - 2Bergen - Learning Centre Désiré Collen, Leuven, Belgium. 4. Interuniversity Centre for Biostatistics and Statistical Bioinformatics, Leuven, Belgium. 5. Department of General Medical Oncology, University Hospitals Leuven, Leuven, Belgium. 6. Department of Gynaecological Oncology, University Hospitals Leuven, Leuven, Belgium.
Abstract
BACKGROUND: Exact numbers of breast cancer recurrences are currently unknown at the population level, because they are challenging to actively collect. Previously, real-world data such as administrative claims have been used within expert- or data-driven (machine learning) algorithms for estimating cancer recurrence. We present the first systematic review and meta-analysis, to our knowledge, of publications estimating breast cancer recurrence at the population level using algorithms based on administrative data. METHODS: The systematic literature search followed Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. We evaluated and compared sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of algorithms. A random-effects meta-analysis was performed using a generalized linear mixed model to obtain a pooled estimate of accuracy. RESULTS: Seventeen articles met the inclusion criteria. Most articles used information from medical files as the gold standard, defined as any recurrence. Two studies included bone metastases only in the definition of recurrence. Fewer studies used a model-based approach (decision trees or logistic regression) (41.2%) compared with studies using detection rules without specified model (58.8%). The generalized linear mixed model for all recurrence types reported an accuracy of 92.2% (95% confidence interval = 88.4% to 94.8%). CONCLUSIONS: Publications reporting algorithms for detecting breast cancer recurrence are limited in number and heterogeneous. A thorough analysis of the existing algorithms demonstrated the need for more standardization and validation. The meta-analysis reported a high accuracy overall, which indicates algorithms as promising tools to identify breast cancer recurrence at the population level. The rule-based approach combined with emerging machine learning algorithms could be interesting to explore in the future.
BACKGROUND: Exact numbers of breast cancer recurrences are currently unknown at the population level, because they are challenging to actively collect. Previously, real-world data such as administrative claims have been used within expert- or data-driven (machine learning) algorithms for estimating cancer recurrence. We present the first systematic review and meta-analysis, to our knowledge, of publications estimating breast cancer recurrence at the population level using algorithms based on administrative data. METHODS: The systematic literature search followed Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. We evaluated and compared sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of algorithms. A random-effects meta-analysis was performed using a generalized linear mixed model to obtain a pooled estimate of accuracy. RESULTS: Seventeen articles met the inclusion criteria. Most articles used information from medical files as the gold standard, defined as any recurrence. Two studies included bone metastases only in the definition of recurrence. Fewer studies used a model-based approach (decision trees or logistic regression) (41.2%) compared with studies using detection rules without specified model (58.8%). The generalized linear mixed model for all recurrence types reported an accuracy of 92.2% (95% confidence interval = 88.4% to 94.8%). CONCLUSIONS: Publications reporting algorithms for detecting breast cancer recurrence are limited in number and heterogeneous. A thorough analysis of the existing algorithms demonstrated the need for more standardization and validation. The meta-analysis reported a high accuracy overall, which indicates algorithms as promising tools to identify breast cancer recurrence at the population level. The rule-based approach combined with emerging machine learning algorithms could be interesting to explore in the future.
Authors: Reina Haque; Jiaxiao Shi; Joanne E Schottinger; Syed Ajaz Ahmed; Joanie Chung; Chantal Avila; Valerie S Lee; Thomas Craig Cheetham; Laurel A Habel; Suzanne W Fletcher; Marilyn L Kwan Journal: Med Care Date: 2015-04 Impact factor: 2.983
Authors: Haejin In; Karl Y Bilimoria; Andrew K Stewart; Kristen E Wroblewski; Mitchell C Posner; Mark S Talamonti; David P Winchester Journal: Ann Surg Oncol Date: 2014-02-07 Impact factor: 5.344
Authors: Daina B Esposito; Leo Russo; Dina Oksen; Ruihua Yin; Vibha C A Desai; Jennifer G Lyons; Patrice Verpillat; Jose L Peñalvo; Francois-Xavier Lamy; Stephan Lanes Journal: Cancer Epidemiol Date: 2019-05-22 Impact factor: 2.984
Authors: Simu K Thomas; Sandra E Brooks; C Daniel Mullins; Claudia R Baquet; Sanjay Merchant Journal: Pharmacoepidemiol Drug Saf Date: 2002-12 Impact factor: 2.890
Authors: P McGale; C Taylor; C Correa; D Cutter; F Duane; M Ewertz; R Gray; G Mannu; R Peto; T Whelan; Y Wang; Z Wang; S Darby Journal: Lancet Date: 2014-03-19 Impact factor: 79.321