Philip M Grimley1, Zhenqiu Liu2, Kathleen M Darcy3, Matthew T Hueman4, Huan Wang5, Li Sheng6, Donald E Henson7, Dechang Chen7. 1. Department of Pathology, F. Edward Hébert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, MD, USA. 2. Department of Public Health Sciences, Penn State Cancer Institute, Hershey, PA, USA. 3. Department of Obstetrics & Gynecology, Uniformed Services University of the Health Sciences, Bethesda, MD, USA. 4. Department of Surgical Oncology, John P. Murtha Cancer Center, Walter Reed National Military Medical Center, Bethesda, MD, USA. 5. Department of Biostatistics, George Washington University, Washington, DC, USA. 6. Department of Mathematics, Drexel University, Philadelphia, PA, USA. 7. Department of Preventive Medicine & Biostatistics, F. Edward Hébert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, MD, USA.
Abstract
INTRODUCTION: Integrating additional factors into the International Federation of Gynecology and Obstetrics (FIGO) staging system is needed for accurate patient classification and survival prediction. In this study, we tested machine learning as a novel tool for incorporating additional prognostic parameters into the conventional FIGO staging system for stratifying patients with epithelial ovarian carcinomas and evaluating their survival. MATERIAL AND METHODS: Cancer-specific survival data for epithelial ovarian carcinomas were extracted from the Surveillance, Epidemiology, and End Results (SEER) program. Two datasets were constructed based upon the year of diagnosis. Dataset 1 (39 514 cases) was limited to primary tumor (T), regional lymph nodes (N) and distant metastasis (M). Dataset 2 (25 291 cases) included additional parameters of age at diagnosis (A) and histologic type and grade (H). The Ensemble Algorithm for Clustering Cancer Data (EACCD) was applied to generate prognostic groups with depiction in dendrograms. C-indices provided dendrogram cutoffs and comparisons of prediction accuracy. RESULTS: Dataset 1 was stratified into nine epithelial ovarian carcinoma prognostic groups, contrasting with 10 groups from FIGO methodology. The EACCD grouping had a slightly higher accuracy in survival prediction than FIGO staging (C-index = 0.7391 vs 0.7371, increase in C-index = 0.0020, 95% confidence interval [CI] 0.0012-0.0027, p = 1.8 × 10-7 ). Nevertheless, there remained a strong inter-system association between EACCD and FIGO (rank correlation = 0.9480, p = 6.1 × 10-15 ). Analysis of Dataset 2 demonstrated that A and H could be smoothly integrated with the T, N and M criteria. Survival data were stratified into nine prognostic groups with an even higher prediction accuracy (C-index = 0.7605) than when using only T, N and M. CONCLUSIONS: EACCD was successfully applied to integrate A and H with T, N and M for stratification and survival prediction of epithelial ovarian carcinoma patients. Additional factors could be advantageously incorporated to test the prognostic impact of emerging diagnostic or therapeutic advances.
INTRODUCTION: Integrating additional factors into the International Federation of Gynecology and Obstetrics (FIGO) staging system is needed for accurate patient classification and survival prediction. In this study, we tested machine learning as a novel tool for incorporating additional prognostic parameters into the conventional FIGO staging system for stratifying patients with epithelial ovarian carcinomas and evaluating their survival. MATERIAL AND METHODS: Cancer-specific survival data for epithelial ovarian carcinomas were extracted from the Surveillance, Epidemiology, and End Results (SEER) program. Two datasets were constructed based upon the year of diagnosis. Dataset 1 (39 514 cases) was limited to primary tumor (T), regional lymph nodes (N) and distant metastasis (M). Dataset 2 (25 291 cases) included additional parameters of age at diagnosis (A) and histologic type and grade (H). The Ensemble Algorithm for Clustering Cancer Data (EACCD) was applied to generate prognostic groups with depiction in dendrograms. C-indices provided dendrogram cutoffs and comparisons of prediction accuracy. RESULTS: Dataset 1 was stratified into nine epithelial ovarian carcinoma prognostic groups, contrasting with 10 groups from FIGO methodology. The EACCD grouping had a slightly higher accuracy in survival prediction than FIGO staging (C-index = 0.7391 vs 0.7371, increase in C-index = 0.0020, 95% confidence interval [CI] 0.0012-0.0027, p = 1.8 × 10-7 ). Nevertheless, there remained a strong inter-system association between EACCD and FIGO (rank correlation = 0.9480, p = 6.1 × 10-15 ). Analysis of Dataset 2 demonstrated that A and H could be smoothly integrated with the T, N and M criteria. Survival data were stratified into nine prognostic groups with an even higher prediction accuracy (C-index = 0.7605) than when using only T, N and M. CONCLUSIONS: EACCD was successfully applied to integrate A and H with T, N and M for stratification and survival prediction of epithelial ovarian carcinoma patients. Additional factors could be advantageously incorporated to test the prognostic impact of emerging diagnostic or therapeutic advances.
Authors: Martin Köbel; Steve E Kalloger; Patricia M Baker; Carol A Ewanowich; Jocelyne Arseneau; Viktor Zherebitskiy; Soran Abdulkarim; Samuel Leung; Máire A Duggan; Dan Fontaine; Robin Parker; David G Huntsman; C Blake Gilks Journal: Am J Surg Pathol Date: 2010-07 Impact factor: 6.394
Authors: Anthony N Karnezis; Kathleen R Cho; C Blake Gilks; Celeste Leigh Pearce; David G Huntsman Journal: Nat Rev Cancer Date: 2016-11-25 Impact factor: 60.716
Authors: Lindsey A Torre; Britton Trabert; Carol E DeSantis; Kimberly D Miller; Goli Samimi; Carolyn D Runowicz; Mia M Gaudet; Ahmedin Jemal; Rebecca L Siegel Journal: CA Cancer J Clin Date: 2018-05-29 Impact factor: 508.702
Authors: Jeffrey D Seidman; Russell Vang; Brigitte M Ronnett; Anna Yemelyanova; Jonathan A Cosin Journal: Gynecol Oncol Date: 2014-12-17 Impact factor: 5.482
Authors: Dechang Chen; Huan Wang; Li Sheng; Matthew T Hueman; Donald E Henson; Arnold M Schwartz; Jigar A Patel Journal: J Med Syst Date: 2016-05-17 Impact factor: 4.460
Authors: Mathew T Hueman; Huan Wang; Charles Q Yang; Li Sheng; Donald E Henson; Arnold M Schwartz; Dechang Chen Journal: Cancer Med Date: 2018-07-02 Impact factor: 4.452