Prognostic Scoring Index for Patients with Metastatic Pancreatic Adenocarcinoma.
- Author:
Hyung Soon PARK
1
;
Hye Sun LEE
;
Ji Soo PARK
;
Joon Seong PARK
;
Dong Ki LEE
;
Se Joon LEE
;
Dong Sup YOON
;
Min Goo LEE
;
Hei Cheul JEUNG
Author Information
- Publication Type:Original Article
- Keywords: Pancreatic neoplasms; Prognosis; Research design; Survival
- MeSH: Adenocarcinoma*; Carcinoembryonic Antigen; Drug Therapy; Humans; Methods; Multivariate Analysis; Pancreatic Neoplasms; Prognosis; Proportional Hazards Models; Reference Values; Research Design; Retrospective Studies
- From:Cancer Research and Treatment 2016;48(4):1253-1263
- CountryRepublic of Korea
- Language:English
- Abstract: PURPOSE: This study focused on implementation of a prognostic scoring index based on clinico-laboratory parameters measured routinely on admission in metastatic pancreatic cancer patients. MATERIALS AND METHODS: Records from 403 patients of metastatic disease were analyzed retrospectively. Continuous variables were dichotomized according to the normal range or the best cut-off values statistically determined by Contal and O’Quigley method, and then analyzed in association with prognosis—overall survival (OS), using Cox's proportional hazard model. Scores were calculated by summing the rounded chi-square scores for the factors that emerged in the multivariate analysis. RESULTS: Performance status, hemoglobin, leucocyte count, neutrophil-lymphocyte ratio, and carcinoembryonic antigen were independent factors for OS. When patients were divided into three risk groups according to these factors, median survival was 11.7, 6.2, and 1.3 months for the low, intermediate, and high-risk groups, respectively (p < 0.001). Palliative chemotherapy has a significant survival benefit for low and intermediate-risk patients (median OS; 12.5 months vs. 5.9 months, p < 0.001 and 8.0 months vs. 2.0 months, p < 0.001, respectively). CONCLUSION: We advocate the use of a multivariable approach with continuous variables for prognostic modeling. Our index is helpful in accurate patient risk stratification and may aid in treatment selection.
