Predicting survival after curative colectomy for cancer: individualizing colon cancer staging

Abstract

Purpose: Cancer staging determines extent of disease, facilitating prognostication and treatment decision making. The American Joint Committee on Cancer (AJCC) TNM classification system is the most commonly used staging algorithm for colon cancer, categorizing patients on the basis of only these three variables (tumor, node, and metastasis). The purpose of this study was to extend the seventh edition of the AJCC staging system for colon cancer to incorporate additional information available from tumor registries, thereby improving prognostic accuracy.

Methods: Records from 128,853 patients with primary colon cancer reported to the Surveillance, Epidemiology and End Results Program from 1994 to 2005 were used to construct and validate three survival models for patients with primary curative-intent surgery. Independent training/test data sets were used to develop and test alternative models. The seventh edition TNM staging system was compared with models supplementing TNM staging with additional demographic and tumor variables available from the registry by calculating a concordance index, performing calibration, and identifying the area under receiver operating characteristic (ROC) curves.

Results: Inclusion of additional registry covariates improved prognostic estimates. The concordance index rose from 0.60 (95% CI, 0.59 to 0.61) for the AJCC model, with T- and N-stage variables, to 0.68 (95% CI, 0.67 to 0.68) for the model including tumor grade, number of collected metastatic lymph nodes, age, and sex. ROC curves for the extended model had higher sensitivity, at all values of specificity, than the TNM system; calibration curves indicated no deviation from the reference line.

Conclusion: Prognostic models incorporating readily available data elements outperform the current AJCC system. These models can assist in personalizing treatment and follow-up for patients with colon cancer.

Fig 1.

Flow chart for creation of the Surveillance, Epidemiology, and End Results (SEER) patient data set. AJCC, American Joint Committee on Cancer; ICD, International Classification of Diseases.

Fig 2.

Kaplan-Meier overall survival on the basis of the seventh edition of the American Joint Committee on Cancer Staging Manual.

Fig 3.

Nomograms convey the results of prognostic models estimating overall survival after complete resection of nonmetastatic colon cancer: (A) Model including T- and N-staging elements; (B) model including T-staging elements and number of collected and metastatic lymph nodes; (C) model included in B, with the additions of tumor differentiation, patient age, and sex. Instructions for users: Locate the patient's age on the age axis. Draw a straight line up to the points axis to determine how many points toward survival the patient should receive. Repeat this process for each of the remaining axes, drawing a straight line each time to the points axis. Sum the points received from each prognostic variable, and locate this number on the total points axis. Draw a straight line down from the total points to the 5-year survival axis to ascertain the patient's specific risk of remaining alive 5 years after surgery. An electronic version of the nomograms, which provide 5-year survival estimates with 95% CIs, is also available.

Fig 4.

(A) Calibration curve for 5-year survival using the nomogram with T-stage element, number of collected and metastatic lymph nodes, tumor differentiation, patient age, and sex. The x-axis is nomogram-predicted probability of survival, and the y-axis is observed survival. The reference line is 45 degrees and indicates perfect calibration. (B) Receiver operating characteristic curve predicting 5-year survival after colectomy for cancer, using the nomogram that includes T-stage element, number of collected and metastatic lymph nodes, tumor differentiation, patient age, and sex. Axes are the sensitivity (true-positive rate) and 1 – specificity (false-positive rate) for each particular threshold. Each point on the curve corresponds to a threshold on the predicted probability. The marked points represent potential thresholds to create high- and low-risk cohorts. For example: Using a 0.30 probability of death at 5 years as a cutoff, patients are segregated into high- and low-risk groups with a sensitivity of 0.72 and specificity of 0.64. Decreasing the cutoff threshold to 0.25 increases the sensitivity to 0.81 but lowers the specificity to 0.52. Increasing the threshold to 0.40 reduces sensitivity to 0.53 but raises specificity to 0.80.