A nomogram of preoperative predictors for occult metastasis in patients with PDAC during laparoscopic exploration

Abstract

Background: Surgical resection is the only potentially curative treatment for pancreatic ductal adenocarcinoma (PDAC). However, most of patients lose the chance of surgery due to the unresectable disease at the time of diagnosis. Despite the improvement of radiological imaging, a portion of patients intended for radical resection were proven to be unresectable at surgical exploration due to occult metastasis.

Methods: Patients who were aimed to undergo radical pancreatectomy for PDAC from 2010 to 2019 were reviewed retrospectively. All patients included underwent diagnostic laparoscopic exploration. Patients were divided into two groups depending on whether distant metastasis were encountered during exploration. Univariate and multivariate logistic regression analyses were used to identify risk factors for occult metastasis. A nomogram to predict occult metastasis of PDAC on exploration was developed and evaluated.

Results: A total of 273 patients who underwent diagnostic laparoscopic exploration were included in this study. Nineteen (7.0%) patients were found with distant metastasis during exploration. Multivariate logistic regression analysis showed that ALT>40U/L, CA19-9, CA125 and regional nodes enlargement were independent predictors for occult metastasis. Incorporating these four factors, the nomogram achieved concordance index of 0.799, with a well-fitted calibration curve.

Conclusions: Occult metastasis is not unusual during surgical exploration in patients with resectable or borderline resectable PDAC. The nomogram could achieve a personal prediction of unexpected distant metastasis on exploration. It may help to sift through patients with PDAC who would benefit from laparoscopic exploration.

Keywords: Pancreatic ductal adenocarcinoma (PDAC); laparoscopic exploration; nomogram; occult metastasis; risk factors.

Figure 1

Suspicious metastatic disease (arrow) detected during laparoscopic exploration. (A) Peritoneal metastasis; (B) phrenic metastatic nodules; (C) liver benign nodule; (D) peritoneal benign nodules. These nodules were missed by preoperative imaging check.

Video 1

A 70-year-old man had abdominal pain and yellow skin mucous membrane with a 2-week history. Abdominal MDCT showed a 28-mm pancreatic head lesion in August 2012. The tumor was characterized as resectable by traditional radiographic examination. Therefore, he was prepared to undergo pancreaticoduodenctomy. Phrenic nodules were detected during laparoscopic exploration and sent to frozen section analysis. Results showed that the nodules were metastatic disease. Therefore, radical surgery was abandoned and cholecystoenterostomy was undergone instead to treat jaundice.

Video 2

A 69-year-old woman presented with a 1-month history of a space-occupying pancreatic lesion. She had experienced no discomfort. MDCT showed a 36-mm pancreatic tail lesion in August 2019, and no abdominal metastases were detected. A liver nodule was found during laparoscopic exploration. Results of the biopsy showed that the nodule was a benign lesion. She underwent radical antegrade modular pancreatosplenectomy after then.

Figure 2

Flow chart of the study. N, number, PDAC, pancreatic ductal adenocarcinoma, IPMN, intraductal papillary mucinous neoplasm.

Figure 3

Nomogram for predicting occult metastasis during laparoscopic exploration. To estimate the risk of occult metastasis, mark patients value at each axis, then draw a straight line perpendicular to the point axis. After sum the points of all variables, mark the sum on the total points axis and make a vertical line to the risk axis. ALT, alanine aminotransferase.

Figure 4

Receiver operating characteristic (ROC) curves of the nomogram, CA125 and CA19-9.

Figure 5

Calibration curve for nomogram. The curve depicts the calibration of the nomogram in terms of agreement between predicted and actual probability of occult metastasis.

Figure 6

The decision curve analysis (DCA) for the nomogram. The net benefit was plotted versus the threshold probability. The blue, red and green line represents the nomogram, CA19-9 and CA125 respectively. The gray line represents the treat-all-patients scheme and the black line represents the treat-none scheme.