Prediction of overall survival for metastatic pancreatic cancer: Development and validation of a prognostic nomogram with data from open clinical trial and real‐world study. (1st June 2018)

• Record Type: Journal Article
• Authors: Hang, Junjie; Wu, Lixia; Zhu, Lina; Sun, Zhiqiang; Wang, Ge; Pan, Jingjing; Zheng, Suhua; Xu, Kequn; Du, Jiadi; Jiang, Hua
• Is Part Of: Cancer medicine. Volume 7:Number 7(2018:Jul.)
• Page Start: 2974
• Page End: 2984
• Publication Date: 2018-06-01
• Subjects: metastatic pancreatic cancer -- open clinical trial -- overall survival -- prognostic nomogram -- real‐world study; 616.994005
• DOI: 10.1002/cam4.1573
• Languages: English
• ISSNs: 2045-7634

Abstract

Abstract: It is necessary to develop prognostic tools of metastatic pancreatic cancer (MPC) for optimizing therapeutic strategies. Thus, we tried to develop and validate a prognostic nomogram of MPC. Data from 3 clinical trials (NCT00844649, NCT01124786, and NCT00574275) and 133 Chinese MPC patients were used for analysis. The former 2 trials were taken as the training cohort while NCT00574275 was used as the validation cohort. In addition, 133 MPC patients treated in China were taken as the testing cohort. Cox regression model was used to investigate prognostic factors in the training cohort. With these factors, we established a nomogram and verified it by Harrell's concordance index ( C ‐index) and calibration plots. Furthermore, the nomogram was externally validated in the validation cohort and testing cohort. In the training cohort (n = 445), performance status, liver metastasis, Carbohydrate antigen 19‐9 (CA19‐9) log‐value, absolute neutrophil count (ANC), and albumin were independent prognostic factors for overall survival (OS). A nomogram was established with these factors to predict OS and survival probabilities. The nomogram showed an acceptable discrimination ability ( C ‐index: .683) and good calibration, and was further externally validated in the validation cohort (n = 273, C ‐index: .699) and testing cohort (n = 133, C ‐index: .653).The nomogram total points (NTP) had the potential to stratify patients into 3‐risk groups with median OS of 11.7, 7.0 and 3.7 months ( P < .001), respectively. In conclusion, the prognostic nomogram with NTP can predict OS for patients with MPC with considerable accuracy. Abstract : We use the data of 718 patients from 3 open clinical trials (NCT00844649, NCT01124786, and NCT00574275) in the Project Data Sphere (PDS) and 133 patients from real‐world study to establish and validate a prognostic nomogram based on clinicopathology characteristics of independent prognostic value. The findings in this study provide compelling evidence that the prognostic nomogram with nomogram total points (NTP) can predict overall survival (OS) for patients with metastatic pancreatic cancer (MPC). The model shows an acceptable discernibility (revealed as C ‐index) and good calibration (demonstrated as calibration curve) in both training cohort and validation cohort.