A Retrospective Analysis of The Long-Term Operative Outcomes for T2 Carcinoma of Gallbladder

Introduction

Gallbladder Carcinoma (GBC) remains a rare disease except in some geographical areas like central and south America, central and eastern Europe, and Japan [1]. Patients with GBC usually present late as symptoms usually occur at advanced stage. Earlier disease is usually detected incidentally on imaging or after cholecystectomy for other indications commonly gallstone disease. Incidentally detected GBC after routine cholecystectomy was reported to be 0.6-1.6% [2,3]. The disease is regarded as early if the cancer cells are limited to the lamina propria (T1a) or just invade to the muscle layer (T1b) [4]. T2 disease implies the tumour has invaded through the muscle layer to reach the perimuscular connective tissue. Recently T2 is further classified to T2a if tumour locates at peritoneal side while T2b for tumour locates at liver side [5,6]. If tumour perforates the serosa or directly invades the liver, it becomes T3 stage. In T4 disease, the tumour invades to main portal vein or hepatic artery or 2 or more extrahepatic organs. T3 and T4 disease carry a dismal prognosis as even en bloc adjacent organ resection is permissible, this is not associated with improved long-term survival [7]. It is the tumour biology and stage rather than the extent of resection more important predictors for survival [8]. Curative surgery is only possible for earlier stage disease. Currently, simple cholecystectomy is considered as an adequate treatment for T1a disease. For T1b disease that is discovered after cholecystectomy, it is still controversial whether reoperation for liver resection and porta lymph node dissection is beneficial for survival. For T2 disease, reoperation is recommended to achieve disease clearance and to prolong survival. Study has shown that T2a had better prognosis than T2b [9]. Whether T2a disease desires reoperation after previous cholecystectomy yet to be answered. This study aims to assess the long-term operative outcomes for T2 GBC, with special reference to any difference in survival between cholecystectomy alone and extended surgery which includes liver resection, Lymph Node (LN) dissection with or without bile duct resection. Subgroup analysis is also made between T2a and T2b disease to evaluate any prognostic difference.

Methods

This was a retrospective study based on the pathological diagnosis of GBC on resected specimens from two hospitals (Prince of Wales Hospital and North District Hospital within the New Territories East Cluster of Hong Kong). The study period was between April 2003 and February 2022. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Informed consent was waived due to retrospective nature of study. Patient confidentiality and privacy rights were all protected. During the study period, 83 consecutive patients were identified to have GBC on resected specimens, which were all only surgically treated histologically confirmed cases of GBC. The number of patients in different T and M staging was as follows: 4 T1a, 5 T1b, 21 T2a, 21 T2b, 20 T3, 3 T4 and 9 M1 disease. T2 stage was the commonest stage for operated GBC in our centre and formed the patient cohort for the present study.In case of uncertainty of histological diagnosis or T staging of disease, the histology slides would be reviewed by a designated pathologist (HHWL) who was specialised on hepatobiliary pathology. Only T2 stage GBC was included in this study to make interpretation of result and recommendation of treatment for this intermediate stage of disease more precise. Besides, this was also the commonest stage of disease in GBC. The demographic data, operation details and follow-up outcomes were retrieved from a prospectively collected electronic database. No patient was lost to follow-up. Missing data was drop out for analysis. The study was approved by the Joint Chinese University of Hong Kong - New Territories East Cluster Clinical Research Ethics Committee (CREC Ref No. 2022.320).

Surgical Approach

For patients with suspected GBC, pre-operative imaging including Computed Tomography (CT) and Positron Emission Tomography (PET) were performed for delineation of lesion and staging of disease. During surgery, diagnostic laparoscopy was performed first to rule out liver or peritoneal metastasis before contemplating radical resection. Usually gallbladder would be removed first and together with the cystic duct margin these were sent for frozen section. Extended resection was defined as including liver resection (gallbladder fossa wedge resection or segment 4b/5 resection) and porta LN dissection (resection of cystic LNs, common bile duct LNs, LNs around hepatoduodenal ligament and the posterior superior pancreaticoduodenal LNs). It was performed if GBC stage T1b or above was found. Bile duct resection with Roux-en-Y hepatojejunostomy reconstruction was performed if the cystic duct margin close to common duct was involved. Occasionally pancreaticoduodenectomy was required if gross peripancreatic head LN involvement was noted. For patients who were found to have incidental GBC stage T1b or above after cholecystectomy, staging of disease was performed with PET CT. If there was no distant metastasis, surgery which included diagnostic laparoscopy, liver resection, cystic duct stump or bile duct resection with porta LN dissection was offered. If previous cholecystectomy was done laparoscopically, port site excision was also performed.Adjuvant chemotherapy was not given for T2 disease. Patients were regularly follow-up in outpatient clinic. Ultrasound (USG) was performed every 6 months for first two years and then yearly. Tumour markers including CEA and CA19.9 were monitored regularly. CT or PET would be performed when recurrent disease was suspected. Patients would be referred to oncologist for consideration of systemic chemotherapy if recurrent disease was diagnosed.

Statistical Analysis

Continuous variables were expressed as median (range), whilst categorical variables were presented in number (percentage). Mann-Whitney U test was used in continuous variable, Chi-square test or Fisher’s exact test was used in categorical variable. Overall Survival (OS) was calculated from time of last operation to date of death or last available follow-up. Disease-Free Survival (DFS) was calculated from time of last surgery to date of diagnosis of first disease recurrence or last available follow-up. Survival curves were calculated by Kaplan-Meier method and was compared with the log-rank test. Univariate and multivariable analysis were performed by Cox regression to identify the risk factors of OS and DFS, with proportional hazards assumptions assessed. Those variables with P-values less than 0.1 in univariate analysis were used in multivariable analysis. IBM SPSS version 27 was used for statistical analysis, P< 0.05 was considered as statistically significant.

Results

Out of the 42 patients with T2 GBC, 6 patients had just cholecystectomy done while 36 patients had extended resection. Two patients had bile duct resection and hepaticojejunostomy due to involvement of cystic duct margin by tumour. Two other patients had primary diagnosis of choledochal cyst requiring bile duct resection and hepaticojejunostomy. Two further patients had pancreaticoduodenectomy done due to massive peripancreatic head lymphadenopathy. The follow-up period was 36.8 months (1.1-121.8) for cholecystectomy alone group versus 44.8months (0.2-252.7) for extended resection group (P=0.408). Out of the 36 patients with extended surgery, 12 patients had the extended resection done at second operation after initial cholecystectomy showing T2 disease. For the 6 patients with only cholecystectomy done, 5 patients were considered too old or too high risk for extended surgery while one patient refused re-operation. The patient demographics, pathology finding and survival outcomes of these 6 patients were compared with the 36 patients who had extended surgery (Table 1). Patients with cholecystectomy only were older and as expected had less LNs removed. They had significantly more recurrent disease (83.3% vs 30.6%, P=0.023) and worse 5-year DFS (16.7% vs 58.9%, P=0.007) as compared with patients with extended surgery. There was no difference on site of recurrent disease. When separate analysis was done for T2a and T2b subgroups, cholecystectomy alone still had worse 5-year DFS in T2a and more recurrent disease in T2b subgroups (P<0.001 and =0.035 respectively) (Table 2). On the contrary, there was no difference in disease recurrence and survival outcomes between the 24 patients with single stage extended surgery and the 12 patients with second stage extended surgery after previous cholecystectomy (Table 3).When the 21 patients with T2a disease were compared with the other 21 patients with T2b disease, there was no difference in patient demographics and disease characteristics (Table 4). One patient (2.8%) was found to have residual disease in liver after extended surgery following cholecystectomy showing T2b GBC. The recurrence rate for T2a disease was 28.6%, which was lower than T2b disease (47.6%), but it was not statistically significant (P=0.204). The 5-year OS and DFS for T2a were 63.5% and 58.7% respectively, while the corresponding values were 52.9% and 47.1% for T2b, there was no significant difference (P= 0.429 and 0.224 respectively) (Table 4 and Figure 1). Nine patients (25.0%) had LN metastasis for those with LN dissection performed. All of them had either one or two positive LNs (all N1 stage). Three patients had positive nodes along hepatoduodenal ligament only, five patients had further nodal involvement in posterosuperior pancreatic head and one last patient had further LN spread along coeliac trunk and superior mesenteric artery. When the nine patients with LN metastasis compared with 33 patients without LN metastasis, there was no difference in patient demographics, tumour size and biology. There was no difference in tumour recurrence and OS but node positive group had significantly worse 5-year DFS (14.8% vs 62.1%, P=0.009) (Figure 2). On univariate and multivariable analysis for OS and DFS, no predictor for OS was found. For DFS, LN metastasis was a negative predictor while extended resection was a positive predictor for survival (Table 5). T2a and T2b stages were not predictors for both OS and DFS.

Table 1: Patient demographics, histology finding and survival outcome between patients with cholecystectomy alone and patients with extended resection.

ASA: American Society of Anaesthesiologists; N/A: not available; LVI: lympho vascular invasion; PNI: perineural invasion; LN: lymph node; * Statistically significant

Table 2: Survival outcome between patients with cholecystectomy alone and patients with extended resection stratified to T2a and T2b diseases; * Statistically significant

Table 3: Survival outcome between patients with single stage extended resection and second stage extended resection after previous cholecystectomy alone.

Table 4: Patient demographics, histology finding and survival outcome between patients with T2a and T2b tumors.

Table 5: Univariate and multivariable analysis for disease-free survival. LVI: lymphovascular invasion; PNI: perineural invasion; LN: lymph node; *Statistically significant

Figure 1: Overall and disease-free survival between T2a and T2b disease after surgery.

Figure 2: Overall and disease-free survival between patients with or without lymph node metastasis.