Impact of radiation and surgery for intraductal papillary neoplasm of the bile duct: A population-based analysis.

Abstract

363 Background: Intraductal Papillary Neoplasm of the Bile Duct (IPNB), either in-situ or invasive, is a histological variant with better prognosis then the more common adenocarcinoma. This study’s purpose is to use the Surveillance Epidemiology and End Results (SEER) database to evaluate prognostic factors: histology, stage, location, extent of surgery and the use of radiation therapy (RT). Methods: Cases from 1973-2011 were acquired. Inclusion criteria included intrahepatic (IHD), extrahepatic bile duct (EHD) or ampulla of vater (AoV) locations, first primary, extent of surgery and RT history. Kaplan-Meier and Log-Rank methods measured overall survival (OS) and disease specific survival (DSS) in months (m) and their medians (MOS, MDSS). Cox multivariate regression computed hazard ratios (HR) controlling for stage, treatment, surgical extent and histology. Results: . For non-invasive cases, 14% were IPNB (n = 31). Survival was similar for EHD & AoV cases. Surgery was associated with prolonged MOS of 120m compared to 8m without surgery or RT. A trend suggested better survival with lesser extent of surgery for EHD & AoV cases (p < 0.16, n = 8 at both sites). For invasive cases, 5% were IPNB (n = 1309). For cases not receiving surgery, RT was associated with prolonged OS & DSS from 3 to 7m (p = 0.026) and 4 to 8m (p = 0.074). In T1N0M0 EHD cases, surgery with and without RT had similar OS & DSS. Cox analysis observed similar OS & DSS for surgery with and without RT for EHD and AoV cases. Mucin-producing IPNB was less likely local stage disease (10% vs 39%, p < 0.01), with shorter OS 5m vs 23 m (p < 0.01) and DSS 6m vs 28m (p < 0.01), and for EHD cases, with HR = 2.0 (p < 0.01) compared to papillary type IPNB. Conclusions: For non-invasive IPNB, surgery with less extensive resections was associated with better prognosis. For invasive IPNB cases not amenable to surgery, RT improved short term survival. If high-risk factors such as suboptimal surgical margins, which are not recorded in SEER, correlated with the use of RT, then the outcomes in EHD & AoV locations could be explained by an imparted benefit. As well, mucin-associated IPNB, was associated with worse survival than papillary type. Further work is necessary to validate these findings.