Riccardo Lemini1, Kristopher Attwood2, Stacey Pecenka3, Juliet Grego3, Aaron C Spaulding3, Steven Nurkin4, Dorin T Colibaseanu1, Emmanuel Gabriel5. 1. Department of Surgery, Division of Colon and Rectal Surgery, Mayo Clinic, Jacksonville, FL, USA. 2. Department of Biostatistics, Roswell Park Comprehensive Cancer Center, Buffalo, NY, USA. 3. Department of Health Sciences Research, Mayo Clinic, Jacksonville, FL, USA. 4. Department of Surgical Oncology, Roswell Park Comprehensive Cancer Center, Buffalo, NY, USA. 5. Department of Surgery, Division of Surgical Oncology, Mayo Clinic, Jacksonville, FL, USA.
Abstract
BACKGROUND: Individualized postoperative survival calculators for patients with cancer can be an aid for predicting prognosis and clinical decision making, such as the use of adjuvant chemotherapy. The aim of this study was to compare existing survival calculators for colon cancer and determine their performance using an independent cohort of patients. METHODS: A retrospective analysis of a multi-site institutional experience was performed on patients diagnosed with stage II-III colon cancer between January 2012 and March 2013. Patient survival rates were estimated using Roswell Park Comprehensive Cancer Center (RPCCC), Memorial Sloan Kettering Cancer Center (MSKCC), and MD Anderson Cancer Center (MDACC) calculators. These calculators vary in the number and breadth of variables that are included. The agreement between selected models was obtained through a scatter plot matrix and related intra-class correlation coefficient (ICC). Calculators' performances were compared using time-dependent receiver operating characteristic (ROC) curves and corresponding area under the curve (AUC) values. RESULTS: After the application of inclusion and exclusion criteria, a total of 97 patients were included in the analysis. Survival data were available for all patients. Median follow-up was 57.6 months, and the overall 5-year survival rate was 0.74 (95% CI: 0.64-0.82). Overall, the different calculators tended to predict survival similarly (ICC =0.017). However, there was variation among calculator performance with the RPCCC calculator showing the highest performance (AUC =0.913), followed by the MSKCC calculator (AUC =0.803), and the MDACC calculator (AUC =0.644). CONCLUSIONS: Prognostic models incorporating a more comprehensive amount of patient and tumor specific variables may provide a more accurate estimate of individual patient survival rates. These tools can be an actual aid in the clinical practice, allowing physicians to personalize treatment and follow-up for patients with colon cancer.
BACKGROUND: Individualized postoperative survival calculators for patients with cancer can be an aid for predicting prognosis and clinical decision making, such as the use of adjuvant chemotherapy. The aim of this study was to compare existing survival calculators for colon cancer and determine their performance using an independent cohort of patients. METHODS: A retrospective analysis of a multi-site institutional experience was performed on patients diagnosed with stage II-III colon cancer between January 2012 and March 2013. Patient survival rates were estimated using Roswell Park Comprehensive Cancer Center (RPCCC), Memorial Sloan Kettering Cancer Center (MSKCC), and MD Anderson Cancer Center (MDACC) calculators. These calculators vary in the number and breadth of variables that are included. The agreement between selected models was obtained through a scatter plot matrix and related intra-class correlation coefficient (ICC). Calculators' performances were compared using time-dependent receiver operating characteristic (ROC) curves and corresponding area under the curve (AUC) values. RESULTS: After the application of inclusion and exclusion criteria, a total of 97 patients were included in the analysis. Survival data were available for all patients. Median follow-up was 57.6 months, and the overall 5-year survival rate was 0.74 (95% CI: 0.64-0.82). Overall, the different calculators tended to predict survival similarly (ICC =0.017). However, there was variation among calculator performance with the RPCCC calculator showing the highest performance (AUC =0.913), followed by the MSKCC calculator (AUC =0.803), and the MDACC calculator (AUC =0.644). CONCLUSIONS: Prognostic models incorporating a more comprehensive amount of patient and tumor specific variables may provide a more accurate estimate of individual patient survival rates. These tools can be an actual aid in the clinical practice, allowing physicians to personalize treatment and follow-up for patients with colon cancer.
Authors: Sharlene Gill; Charles Loprinzi; Hagen Kennecke; Axel Grothey; Garth Nelson; Ryan Woods; Caroline Speers; Steven R Alberts; Aditya Bardia; Michael J O'Connell; Daniel J Sargent Journal: Cancer Date: 2011-03-01 Impact factor: 6.860
Authors: Danielle M Hari; Anna M Leung; Ji-Hey Lee; Myung-Shin Sim; Brooke Vuong; Connie G Chiu; Anton J Bilchik Journal: J Am Coll Surg Date: 2013-06-12 Impact factor: 6.113
Authors: Sharlene Gill; Charles L Loprinzi; Daniel J Sargent; Stephan D Thomé; Steven R Alberts; Daniel G Haller; Jacqueline Benedetti; Guido Francini; Lois E Shepherd; Jean Francois Seitz; Roberto Labianca; Wei Chen; Stephen S Cha; Michael P Heldebrant; Richard M Goldberg Journal: J Clin Oncol Date: 2004-04-05 Impact factor: 44.544
Authors: George J Chang; Chung-Yuan Hu; Cathy Eng; John M Skibber; Miguel A Rodriguez-Bigas Journal: J Clin Oncol Date: 2009-10-05 Impact factor: 44.544
Authors: Victor E Pricolo; Jon Steingrimsson; Tracey J McDuffie; Joshua M McHale; Brian McMillen; Mark Shparber Journal: Am J Clin Oncol Date: 2020-02 Impact factor: 2.787