Establishment and validation of a prognostic nomogram for patients with resectable perihilar cholangiocarcinoma

Abstract

As the conventional staging systems have poor prognosis prediction ability for patients with perihilar cholangiocarcinoma (pCCA), we established and validated an effective prognostic nomogram for pCCA patients based on their personal and tumor characteristics. A total of 235 patients who received curative intent resections at the Eastern Hepatobiliary Surgery Hospital from 2000 to 2009 were recruited as the primary training cohort. Age, preoperative CA19-9 levels, portal vein involvement, hepatic artery invasion, lymph node metastases, and surgical treatment outcomes (R0 or R1/2) were independent prognostic factors for pCCA patients in the primary cohort as suggested by the multivariate analyses and these were included in the established nomogram. The calibration curve showed good agreement between overall survival probability of pCCA patients for the nomogram predictions and the actual observations and the concordance index (C-index) was 0.68 (95% CI, 0.61-0.71). The C-index values and time-dependent ROC tests suggested that the nomogram is superior to the conventional staging systems including the Bismuth-Corlette, Gazzaniga, Memorial Sloan Kettering Cancer Center (MSKCC), American Joint Committee on Cancer (AJCC) TNM 7th edition, and Mayo Clinic. The nomogram also performed better than the traditional staging system in the internal cohort with 93 pCCA patients from the same institution and an external validation cohort including 84 pCCA patients from another institution in predicting the overall survival of the pCCA patients as suggested by the C-index values and the time-dependent ROC tests. In summary, the proposed nomogram has superior predictive accuracy of prognosis for resectable pCCA patients.

Keywords: nomogram; overall survival; perihilar cholangiocarcinoma.

Figure 1. The survival nomogram for the pCCA patients

(To use the nomogram, the individual patient's value is located on each variable axis, and a line is drawn upward to determine the risk score for each variant. The sum of these scores is located on the total points axis, and a line is drawn downward to the survival axes to determine the probability of 3- or 5-year survival).

Figure 2. Calibration curves for predicting patient survival at

A. 3 years and B. 5 years in the primary cohort and at 3 years in the internal validation cohort C. or external validation cohort D. Nomogram-predicted probability of OS is plotted on the x-axis; actual OS is plotted on the y-axis.

Figure 3. Comparison of the staging systems in the prediction of the OS for pCCA patients in primary training cohort

(A. Bismuth-Corlette; B. Gazzaniga; C. MSKCC; D. AJCC TNM (seventh edition); E. Mayo Clinic; F. Nomogram risk score).