A Simple And Effective Prognostic Staging System Based On Clinicopathologic Features Of Intrahepatic Cholangiocarcinoma

AimsThis study aimed to establish a simple and effective clinicopathologic prognosticstaging system applicable for intrahepatic cholangiocarcinoma (ICC) after partialhepatectomy.MethodsRetrospective univariate and multivariate survival analyses were conducted with atotal of370patients who underwent partial hepatectomy for ICC that resulted inmacroscopic curative resections (R0and R1) at the Eastern Hepatobiliary Surgery Hospita lbetween January2005and December2009. A simple clinicopathologic prognostic stagingsystem based on the independent predictors was developed. The validity of the prognosticstaging system was prospectively assessed in115patients who underwent partialhepatectomy on between January2010and December2010at the same institution andcompared with the American Joint Committee on Cancer (AJCC)6thand7theditionstaging systems. The prognostic power was quantified using a linear trend χ2test and–2log likelihood.ResultsAll patients underwent anatomical resection for ICC that resulted in macroscopiccurative resections (R0and R1). Multivariate analysis of the primary cohort revealed thatseven variables were significantly independent predictors for overall survival (OS): serumpre-albumin (hazard ratio:1.447), carbohydrate antigen19-9(hazard ratio:1.438),carcinoembryonic antigen (hazard ratio:1.732), tumor number (hazard ratio:1.781),vascular invasion (hazard ratio:1.784), regional lymphatic metastasis (hazard ratio:2.003)and local extrahepatic metastasis (hazard ratio:1.506). Using the seven independentpredictors for OS, a new staging system was devised: stage I disease was de fined asmeeting none of the seven predictors, stage II disease was defined as meeting one of theseven predictors, stage III disease was defined as meeting two or three of the seven predictors, stage IV disease was defined as meeting four or more of the seven predictors.For the four staging patients (stage I to stage IV) according to the new staging system, the1-year survival rates were, respectively,97.1%,69.9%,42.1%, and19.3%; the3-yearsurvival rates were, respectively,70.6%,37.6%,14.5%, and1.8%. Compared with the6thand7thAJCC staging systems, the new staging system in the primary cohort had a higherpredictive accuracy for OS in terms of homogeneity and discriminatory ability (value oflinear trend χ2test and–2log likelihood in6thAJCC,7thAJCC and new staging system:48.773and2962.378,43.884and2956.343and66.412and2914.590, respectively). In thevalidation cohort, the new staging system discrimination was likewise superior to the twoother staging systems (value of linear trend χ2test and–2log likelihood in6thAJCC,7thAJCC and new staging system:10.056and720.708,13.340and712.812and22.217and704.804, respectively).ConclusionsThe new staging system based on clinicopathologic features may providemore-accurate prognostic prediction for patients with ICC after tumor resection. But moreimportantly, the new staging system can been easy enforced in clinical practice.