Kate R Pawloski1, Mithat Gonen2, Hannah Y Wen3, Audree B Tadros1, Donna Thompson3, Kelly Abbate1, Monica Morrow1, Mahmoud El-Tamer4. 1. Breast Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA. 2. Biostatistics Service, Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY, USA. 3. Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, NY, USA. 4. Breast Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY, USA. eltamerm@mskcc.org.
Abstract
PURPOSE: Routine use of the oncotype DX recurrence score (RS) in patients with early-stage, estrogen receptor-positive, HER2-negative (ER+/HER2-) breast cancer is limited internationally by cost and availability. We created a supervised machine learning model using clinicopathologic variables to predict RS risk category in patients aged over 50 years. METHODS: From January 2012 to December 2018, we identified patients aged over 50 years with T1-2, ER+/HER2-, node-negative tumors. Clinicopathologic data and RS results were randomly split into training and validation cohorts. A random forest model with 500 trees was developed on the training cohort, using age, pathologic tumor size, histology, progesterone receptor (PR) expression, lymphovascular invasion (LVI), and grade as predictors. We predicted risk category (low: RS ≤ 25, high: RS > 25) using the validation cohort. RESULTS: Of the 3880 tumors identified, 1293 tumors comprised the validation cohort in patients of median (IQR) age 62 years (56-68) with median (IQR) tumor size 1.2 cm (0.8-1.7). Most tumors were invasive ductal (80.3%) of low-intermediate grade (80.5%) without LVI (80.9%). PR expression was ≤ 20% in 27.3% of tumors. Specificity for identifying RS ≤ 25 was 96.3% (95% CI 95.0-97.4) and the negative predictive value was 92.9% (95% CI 91.2-94.4). Sensitivity and positive predictive value for predicting RS > 25 was lower (48.3 and 65.1%, respectively). CONCLUSION: Our model was highly specific for identifying eligible patients aged over 50 years for whom chemotherapy can be omitted. Following external validation, it may be used to triage patients for RS testing, if predicted to be high risk, in resource-limited settings.
PURPOSE: Routine use of the oncotype DX recurrence score (RS) in patients with early-stage, estrogen receptor-positive, HER2-negative (ER+/HER2-) breast cancer is limited internationally by cost and availability. We created a supervised machine learning model using clinicopathologic variables to predict RS risk category in patients aged over 50 years. METHODS: From January 2012 to December 2018, we identified patients aged over 50 years with T1-2, ER+/HER2-, node-negative tumors. Clinicopathologic data and RS results were randomly split into training and validation cohorts. A random forest model with 500 trees was developed on the training cohort, using age, pathologic tumor size, histology, progesterone receptor (PR) expression, lymphovascular invasion (LVI), and grade as predictors. We predicted risk category (low: RS ≤ 25, high: RS > 25) using the validation cohort. RESULTS: Of the 3880 tumors identified, 1293 tumors comprised the validation cohort in patients of median (IQR) age 62 years (56-68) with median (IQR) tumor size 1.2 cm (0.8-1.7). Most tumors were invasive ductal (80.3%) of low-intermediate grade (80.5%) without LVI (80.9%). PR expression was ≤ 20% in 27.3% of tumors. Specificity for identifying RS ≤ 25 was 96.3% (95% CI 95.0-97.4) and the negative predictive value was 92.9% (95% CI 91.2-94.4). Sensitivity and positive predictive value for predicting RS > 25 was lower (48.3 and 65.1%, respectively). CONCLUSION: Our model was highly specific for identifying eligible patients aged over 50 years for whom chemotherapy can be omitted. Following external validation, it may be used to triage patients for RS testing, if predicted to be high risk, in resource-limited settings.
Authors: Soonmyung Paik; Gong Tang; Steven Shak; Chungyeul Kim; Joffre Baker; Wanseop Kim; Maureen Cronin; Frederick L Baehner; Drew Watson; John Bryant; Joseph P Costantino; Charles E Geyer; D Lawrence Wickerham; Norman Wolmark Journal: J Clin Oncol Date: 2006-05-23 Impact factor: 44.544
Authors: Austin D Williams; Sylvia A Reyes; Renee L Arlow; Julia Tchou; Lucy M De La Cruz Journal: Ann Surg Oncol Date: 2018-06-29 Impact factor: 5.344
Authors: Soonmyung Paik; Steven Shak; Gong Tang; Chungyeul Kim; Joffre Baker; Maureen Cronin; Frederick L Baehner; Michael G Walker; Drew Watson; Taesung Park; William Hiller; Edwin R Fisher; D Lawrence Wickerham; John Bryant; Norman Wolmark Journal: N Engl J Med Date: 2004-12-10 Impact factor: 91.245
Authors: Fabrice Andre; Nofisat Ismaila; N Lynn Henry; Mark R Somerfield; Robert C Bast; William Barlow; Deborah E Collyar; M Elizabeth Hammond; Nicole M Kuderer; Minetta C Liu; Catherine Van Poznak; Antonio C Wolff; Vered Stearns Journal: J Clin Oncol Date: 2019-05-31 Impact factor: 44.544
Authors: Lyndsay N Harris; Nofisat Ismaila; Lisa M McShane; Fabrice Andre; Deborah E Collyar; Ana M Gonzalez-Angulo; Elizabeth H Hammond; Nicole M Kuderer; Minetta C Liu; Robert G Mennel; Catherine Van Poznak; Robert C Bast; Daniel F Hayes Journal: J Clin Oncol Date: 2016-02-08 Impact factor: 44.544
Authors: B Fisher; J Dignam; N Wolmark; A DeCillis; B Emir; D L Wickerham; J Bryant; N V Dimitrov; N Abramson; J N Atkins; H Shibata; L Deschenes; R G Margolese Journal: J Natl Cancer Inst Date: 1997-11-19 Impact factor: 13.506
Authors: L Chin-Lenn; R H De Boer; E Segelov; G M Marx; T M Hughes; N J McCarthy; S C White; S S Foo; J J Rutovitz; S Della-Fiorentina; R Jennens; Y C Antill; D Tsoi; M F Cronk; J M Lombard; B E Kiely; J H Chirgwin; A Gorelik; G B Mann Journal: Asia Pac J Clin Oncol Date: 2018-09-30 Impact factor: 2.601
Authors: Joseph A Sparano; Robert J Gray; Della F Makower; Kathleen I Pritchard; Kathy S Albain; Daniel F Hayes; Charles E Geyer; Elizabeth C Dees; Matthew P Goetz; John A Olson; Tracy Lively; Sunil S Badve; Thomas J Saphner; Lynne I Wagner; Timothy J Whelan; Matthew J Ellis; Soonmyung Paik; William C Wood; Peter M Ravdin; Maccon M Keane; Henry L Gomez Moreno; Pavan S Reddy; Timothy F Goggins; Ingrid A Mayer; Adam M Brufsky; Deborah L Toppmeyer; Virginia G Kaklamani; Jeffrey L Berenberg; Jeffrey Abrams; George W Sledge Journal: N Engl J Med Date: 2018-06-03 Impact factor: 91.245
Authors: Mei-Yin C Polley; Samuel C Y Leung; Lisa M McShane; Dongxia Gao; Judith C Hugh; Mauro G Mastropasqua; Giuseppe Viale; Lila A Zabaglo; Frédérique Penault-Llorca; John M S Bartlett; Allen M Gown; W Fraser Symmans; Tammy Piper; Erika Mehl; Rebecca A Enos; Daniel F Hayes; Mitch Dowsett; Torsten O Nielsen Journal: J Natl Cancer Inst Date: 2013-11-07 Impact factor: 13.506
Authors: Austin D Williams; Kate R Pawloski; Hannah Y Wen; Varadan Sevilimedu; Donna Thompson; Monica Morrow; Mahmoud El-Tamer Journal: Breast Cancer Res Treat Date: 2022-10-21 Impact factor: 4.624