|
|
Back to 2018 Posters
PREDICTORS FOR DISTANT METASTASIS IN PATIENTS WITH POTENTIALLY RESECTABLE PANCREATIC CANCER BASED ON MDCT
Jishu Wei*, Xinchun Liu, Yue Fu, Qiuyang Chen, Zipeng Lu, Junli Wu, Wentao Gao, Kuirong Jiang, Yi Miao Pancreas Center, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China Purpose: To investigate the predictors of distant metastasis in patients with potentially resectable pancreatic cancer. Methods: 235 cases were identified through medical records reviewing and included in this study. Perioperative data and surgical outcomes were retrospectively analyzed. Results: Comprising 149 men (62.9%) and 88 women (37.1%), with a mean age of 61.3 ± 10.0 years old. Distant metastasis was found in 62 patients (26%, 62/235). Patients with distal pancreatic cancer were more likely found distant metastasis than patients with pancreatic head cancer (22/61 vs 40/174, p = 0.046). Patients with metastasis had a larger primary tumor (5.5 ± 2.0 vs 3.8 ± 1.6, p < 0.001), and lower ALT level (75.9 ± 146.1 vs 142.2 ± 179.1, p = 0.010), and a higher CA199 level (525.2 ± 422.0 vs 335.7 ± 350.2, p < 0.001) than patients without metastasis. On multivariate logistic regression analysis, it showed that age ≤ 62 years old (OR = 3.97, p<0.001), male sex (OR = 2.79, P = 0.011), a tumor size ≥ 4.0 cm (OR = 16.02, P < 0.001), an ALT level < 125 U/L (OR = 6.19, P < 0.001), and a CA19-9 level ≥ 385 U/ml (OR = 3.53, P < 0.001) were independent risk factors of occult distant metastasis of pancreatic cancer. These five factors were used to develop a predictive model, which when tested, fitted the data well, with an area under the receiver operating characteristic curve of 0.813. Conclusions: In patients with a potentially resectable pancreatic cancer based on MDCT, younger age, male, large tumor size, ALT<150 U/L and CA19-9 ≥ 385 U/ml are key risk factors of micro distant metastasis at exploration. Patients with more than 3 factors should undergo further staging before laparotomy.
Back to 2018 Posters
|