Reduction of intrapancreatic neural density in cancer tissue predicts poorer outcome in pancreatic ductal carcinoma

Abstract

Neural invasion is one of the malignant features contributing to locally advanced and/or metastatic disease progression in patients with pancreatic ductal adenocarcinoma (PDAC). Few studies exist on the distribution and state of nerve fibers in PDAC tissue and their clinicopathological impacts. The aim of the present study was to investigate the clinicopathological characteristics and prognostic value of intrapancreatic neural alterations in patients with PDAC. We retrospectively analyzed 256 patients with PDAC who underwent macroscopic curative surgery. Nerve fibers, immunolabeled with a specific neural marker GAP-43, were digitally counted and compared among PDAC, chronic pancreatitis (CP) and normal pancreatic tissues. Interlobular nerve fibers were apparently hypertrophic in both CP and PDAC, although intrapancreatic neural density and nerve number decreased characteristically in PDAC. They tended to decrease toward the center of the tumor. Kaplan-Meier survival analyses revealed a statistically significant correlation between low neural density and shorter overall survival (OS) (P = 0.014), and between high neural invasion and shorter OS (P = 0.017). Neural density (P = 0.04; HR = 1.496; 95% CI 1.018-2.199) and neural invasion ratio (P = 0.064; HR = 1.439; 95% CI .980-2.114) were prognostic factors of shorter OS in the multivariate analysis. These findings suggest low intrapancreatic neural density in patients with PDAC as an independent prognosticator, which may represent aggressive tumor behavior. Furthermore, we propose a simple, practical and reproducible method (to measure neural density and the neural invasion ratio during conventional histopathological diagnosis of PDAC), which has been validated using another cohort (n = 81).

Keywords: ductal carcinoma; intrapancreatic neural density; intrapancreatic neural invasion; pancreas; prognosis.

Figure 1

Comparisons of the neural alteration among normal pancreas (NP), chronic pancreatitis (CP), and pancreatic ductal adenocarcinoma (PDAC) tissue. Representative immunohistochemical images with GAP‐43 in NP, CP and PDAC, and the corresponding HE‐stained images. In NP (A to D), there are large numbers of intraparenchymal minute nerve fibers. Several hypertrophic nerves are seen but intraparenchymal nerve fibers are decreased in the atrophy of the acinar cell within CP (E to H), and in the desmoplastic stroma within PDAC (I to L). Adenocarcinoma cells proliferate with desmoplastic stromal reaction in the right half of the panel, and invade into the surrounding pancreatic parenchyma with chronic inflammation and fibrosis in I and J. Low‐power field in A, B, E, F, I and J, and middle‐power field in C, D, G, H, K and L

Figure 2

Quantitative analysis of the neural alterations in normal pancreas (NP), chronic pancreatitis (CP) and pancreatic ductal adenocarcinoma (PDAC). The intrapancreatic neural density (A), the nerve number (B) and the mean nerve size (C) are measured and compared among NP, CP and PDAC. Defining the intra‐tumoral measurement field by drawing a dotted line (yellow) along the cancer invasive margin (D), which is divided into 3 areas from “Area 1” to “Area 3” (refer to Materials and Method). In addition, the “surrounding area” is defined (E). The neural density (F), the nerve number (G) and the mean nerve size (H) are measured and compared among these areas. N.S., not significant

Figure 3

Kaplan‐Meier survival curves for overall survival (OS) in patients with pancreatic ductal adenocarcinoma (PDAC) according to the intrapancreatic neural density (A), nerve number (B) and neural invasion ratio (C)

Figure 4

Receiver operating characteristic (ROC) curve analysis and Kaplan‐Meier survival curves. Cut‐off point of the intrapancreatic nerve number (A) or the intrapancreatic neural invasion ratio (B) in ROC curve analysis. The cut‐off value corresponding to the point with the lowest distance to the upper‐left corner of the ROC curve (and also the point with the highest distance to the reference line) is indicated by arrow in A. The candidate cut‐off value corresponding to the point with the lowest distance to the upper‐left corner of the ROC curve (dotted arrow) is 46.2% and another candidate corresponding to the point with the highest distance to the reference line of the ROC curve (black arrow) is 55.6%. To carry out a simple and practical method, we set the cut‐off value as 50%, with the corresponding point indicated with a red arrow in B. Kaplan‐Meier survival curves showing the comparison of OS (C, E) and DFS (D) between the high (red) and low (blue) nerve number groups (C, D) and those of the neural invasion ratio groups (E) in the validation cohort. P‐values are obtained from the log‐rank test