|
|
Back to 2016 Annual Meeting
Neuroendocrine Liver Metastasis: A Novel Nomogram Predicting the Prognosis of Patients after Liver Resection
Andrea Ruzzenente1, Fabio Bagante*1, Francesca Bertuzzo1, Luca Aldrighetti6, Giorgio Ercolani3, Alessandro Ferrero5, Felice Giuliante4, Guido Torzilli2, Gian Luca grazi7, Alfredo Guglielmi1, Calogero Iacono1
1G. B. Rossi Hospital, University of Verona, Verona, Italy; 2Istituto Clinico Humanitas, Milan, Italy; 3Ospedale Sant'Orsola University of Bologna, Bologna, Italy; 4Catholic University of the Sacred Heart, Rome, Italy; 5Ospedale Mauriziano Umberto I, Turin, Italy; 6Ospedale San Raffaele, Milano, Italy; 7Regina Elena National Cancer Institute, Rome, Italy
Background: Despite surgery remains the only potentially curative option for patients with neuroendocrine liver metastases (NELMs), the factors determining the prognosis of patients following hepatectomy are poorly understood. We sought to identify the predictors of survival after liver resection for NELMs and develop a clinical tool to predict the prognosis of patients following surgery.
Methods: A multicentric database including 7 tertiary referral center Hepato-biliary-pancreatic centers was used to identify patients who underwent hepatectomy for NELMs between January, 1990 and December, 2014. To predict the patient’s prognosis, we developed a nomogram including all variables resulted statistically significant (p<0.05) in both univariate and multivariable survival analysis. Herrell's concordance index (c-index) was used to internally validate the nomogram.
Results: The median age of the 238 patients included in the study was 61.9 years (IQR: 51.5-70.1) and 55.5% (n=132) of patients were male. The majority of patients had a non-functional (n=160, 67.3%) neuroendocrine tumor (NET) and no underlying genetic syndrome (n=223, 93.9%). In 95 (39.9%) patients the primary NET was located in the pancreas-duodenum region, 169 (71.0%) patients had NELMs at diagnosis while only 16 (6.7%) had extrahepatic NET metastasis. The NET grade resulted G1-G2 in 191 (80.2%) and median Ki-67 was 9% (IQR, 2-15). The median number of NELMs was 3 (IQR, 2-7) while the median size resulted 45 mm (IQR, 24-75). A major hepatectomy was performed on 152 (63.9%) patients. The 5- and 10-year overall survival (OS) were 67.1% (95%CI, 60.3-74.8) and 50.8% (95%CI, 42.5-58.5), respectively. Number of NELMs (HR=1.06, 95%CI 1.02-1.11, p=0.007), tumor size (HR=1.01, 95%CI 1.00-1.01, p<0.001), and Ki-67 (HR=1.06, 95%CI 1.04-1.09, p<0.001) resulted predictors of OS in both univariate and multivariable analysis and were included in the nomogram predicting patient’s prognosis (Figure 1a). The stratification into three risk classes based on the predictions of our nomogram demonstrated a good prognostic discrimination (c-index=0.72). Particularly, 42 (17.6%) patients were in low-risk class (>80% predicted 5-year OS), 140 (58.8%) in medium-risk class (80%-40% predicted 5-year OS), and 56 (23.5%) in high-risk class (<40% predicted 5-year OS). The 5- and 10-year OS resulted 97.0% and 97.0% in low-risk class, 71.0% and 55.9% in medium-risk class and 41.1% and 20.0% in high-risk class (p<0.001; Figure 1b).
Conclusions: We developed a nomogram including both clinical (size and number of NELMs) and biological (Ki-67) variables able to predict patient’s prognosis with a c-index >70%. Moreover, the classification based on the prediction of our nomogram demonstrated a good ability to discriminate patients with favorable and poor prognosis.
 
Back to 2016 Annual Meeting
|
