World J Surg (2014) 38:484–489 DOI 10.1007/s00268-013-2266-4
Patterns of Failure and Determinants of Outcomes Following Radical Re-resection for Incidental Gallbladder Cancer Savio George Barreto • Satish Pawar • Sanket Shah • Sanjay Talole • Mahesh Goel Shailesh V. Shrikhande
•
Published online: 2 October 2013 Ó Socie´te´ Internationale de Chirurgie 2013
Abstract Background Better appreciation of the course and factors that influence incidental gallbladder cancer is needed to develop treatment strategies aimed at improved outcomes. The purpose of this study was to determine pattern of disease recurrence and influencing factors in patients undergoing radical re-resection for incidental gallbladder cancer. Methods Patients undergoing radical re-resection from February 2003 to May 2010 were analyzed. Influence of variables (lymph node ratio, ASA grade, gender, adjuvant treatment, time interval between cholecystectomy and radical re-resection (in months), and TNM stage) on disease-free survival was assessed. Results Of 163 patients, 127 (92 female and 35 male patients; median age 50 years) underwent successful radical re-resection. Median duration between two surgeries was 2 months (range 1–10). Twenty-five percent of patients with pT1b disease had lymph node metastases. Two-year disease-free survival rate was 79.6 % (median follow-up, 16 months). On follow-up, 18 of 24 patients
S. G. Barreto S. Pawar S. Shah M. Goel S. V. Shrikhande Department of Gastrointestinal and Hepato-Pancreato-Biliary Surgical Oncology, Tata Memorial Hospital, Parel, Mumbai, India S. Talole Department of Biostatistics and Epidemiology, Tata Memorial Hospital, Parel, Mumbai, India S. V. Shrikhande (&) GI Management Group, Tata Memorial Centre, Parel, Mumbai 400 012, India e-mail:
[email protected];
[email protected]
123
developed recurrences at distant sites. Lymph node metastasis was the single variable significantly influencing disease-free survival. Adjusting for disease stage when analyzing time interval between cholecystectomy and radical re-resection on a continuous scale as a prognostic factor for recurrence revealed no significant impact of increasing interval between surgeries (hazard ratio 1.12; 95 % confidence interval 0.95–1.34; p = 0.17). Conclusions The most important predictor of disease recurrence is lymph node metastases. In patients who undergo curative radical re-resection for incidental gallbladder cancer, recurrent disease is more likely to occur at distant sites. Patients with pT1b disease should be offered radical re-resection with a radical lymphadenectomy. It is not the delay in revision surgery but TNM stage that influences outcomes in incidental gallbladder cancer.
Introduction Gallbladder cancer is a relatively uncommon cancer worldwide with age-standardized incidence rates of 2/100,000 [1]. Recent reports have further suggested a decline in the incidence of the disease in different parts of the world [2, 3]. However in India, especially in the North of the country, gallbladder cancer remains a major problem [4, 5] (age standardized rates from 3.9 to 8.6/100,000) [6] with the vast majority of patients presenting with advanced disease [7, 8]. The uncommonness of gallbladder cancer has contributed to the generally poor understanding of the disease [9, 10]. However, despite the suggested declining trend of this cancer in the world, it remains imperative that an improved appreciation of the course of the disease and factors influencing this course is needed to develop treatment strategies designed to improve its outcomes not only
World J Surg (2014) 38:484–489
in countries that continue to have a high incidence of the cancer but in the rest of the world, as well. ‘‘Incidental’’ gallbladder cancer was a term coined to denote a histological surprise of a cancer when analyzing a pathological specimen of simple cholecystectomy performed for presumed gallstone disease. The ideal treatment for patients with incidental gallbladder cancer consists of a radical re-resection (including a radical portal lymphadenectomy, resection of a 2–3 cm wedge of liver, and revision of the cystic duct margin, with or without an excision of the extrahepatic bile duct) so long as the disease is CT1b with no evidence of distant metastasis. The purpose of the current study was to evaluate a cohort of patients who underwent radical re-resection surgery for incidentally detected gallbladder cancer to determine the prognosis of initial disease recurrence and the factors influencing these outcomes.
Methods A retrospective analysis was performed of patients who were referred to the Department of Gastrointestinal and Hepatopancreatobiliary Surgical Oncology of the Tata Memorial Centre with incidentally detected gallbladder cancer and who successfully underwent radical re-resection from February 2003 to May 2010. All surgeries were performed by, or under the supervision of, the consultant surgeons in the unit. Preoperatively, all patients were investigated with routine blood investigations, including blood counts, liver and renal functions, and tumor marker serum carbohydrate antigen 19-9 (CA 19-9). Preoperative evaluation also included contrast-enhanced computed tomography (CT scan) of the abdomen and pelvis and a review of the histopathology (paraffin-embedded tissue blocks and/or slides) of the gallbladder specimen excised during the primary surgery (laparoscopic or open cholecystectomy) to determine the histologic type and grade (based on the areas of tumour with highest grade and classified as well-, moderately-, or poorly differentiated), depth of infiltration, and pathological T (pT) stage of the cancer. Since 2008, based on our experience with the use of positron emission tomography (PET) in gallbladder cancer [11, 12], patients with or without evidence of residual local disease but suspected to have systemic disease underwent a PET–CT. All patients (pT1b and above) with non-metastatic, locally resectable disease on PET–CT (after October 2007) were offered surgical exploration with an intent to perform a radical re-resection. The procedure has been previously described in a publication from the unit 13. Histopathological examination was performed on the resected wedge of liver, lymph nodes, and cystic duct
485
margin tissues that were fixed in 10 % buffered formaldehyde solution. The liver wedge was subjected to serial sectioning to determine presence of residual disease and involvement or freedom of the resection margin. Immunohistochemistry for the lymph nodes was not performed routinely. All patients had their pathology re-staged by the 7th Edition of the American Joint Cancer Committee TNM staging system for purpose of uniformity [14]. Data, such as age, sex, American Society of Anesthesiologists (ASA) grade, duration between the primary cholecystectomy and the radical re-resection (in months), and the use of adjuvant treatment—chemotherapy (regimens included gemcitabine monotherapy or in combination with cisplatin or 5-fluorouracil with cisplatin) and/or radiotherapy administered as an external beam therapy, were recorded. The patients in this study span a period of 10 years. Given the lack of consensus on the benefit of adjuvant therapy not only throughout the study period but even to this day in gallbladder cancer [15], as well as the fact that our patients come from different social and financial backgrounds, the decision for adjuvant chemotherapy was not based on a fixed protocol but decided on a case to case basis taking into account the patient’s performance status after surgery as well as financial status. In general, patients with lymph node positive disease were offered the option of chemotherapy and/or radiotherapy if they were performance status 1 or 2 and if they could afford the therapy after being explained the potential of benefit without any assurance of it. Follow-up Postoperatively, patients were followed up at 3-monthly intervals for the first 2 years, then every 6 months for another year and then yearly. The follow-up regimen consisted of a clinical history and examination, liver function tests and serum CA 19-9, and an ultrasound of the abdomen and pelvis. Recurrence of disease was defined by the presence of documented disease on CT scan/ultrasonography confirmed by histopathology and/or PET–CT imaging with or without a corresponding elevation in serum CA 19-9. Ethics The data of the present study were collected in the course of common clinical practice and, accordingly, the signed informed consent was obtained from each patient for any surgical and clinical procedure. The study protocol conforms to the ethical guidelines of the ‘‘World Medical Association Declaration of Helsinki—Ethical Principles for Medical Research Involving Human Subjects’’ adopted by
123
486
the 18th WMA General Assembly, Helsinki, Finland, June 1964, as revised in Tokyo 2004. No approval of the institutional review committee was needed. Statistical analysis Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) Inc., Version 18.0 for Windows and Stata 11.0 version. Nominal data was provided as number (%) and continuous data as median (range). Univariate analysis of prognostic factors affecting survival was performed. Factors that seemed determinants were evaluated by Kaplan–Meier survival curves. Finally, the significant factors found in univariate analysis were tested in multivariate analysis using the Cox regression method. Kaplan–Meier survival technique was used to plot and study the distribution of disease-free period. Log-rank (LR) test was used to measure the equality of survival distributions for variables including disease stage, lymph node ratio, use of adjuvant therapy in the postoperative period, gender (male vs. female), ASA grade (1 vs. 2), and the time interval between the cholecystectomy and radical re-resection (\2 vs. C2 months). Lymph node ratio was derived as ratio of positive to total number of nodes dissected. Lymph node ratio was coded as 0 for none of the nodes being positive and one if the there was even a single positive metastatic node. Multivariate analysis was performed using Cox regression forward stepwise LR model. Multivariate model included variables, such as lymph node ratio, ASA grade, gender, adjuvant treatment (if any), time interval between the cholecystectomy and the radical reresection (in months) and TNM stage of the disease.
Results A total of 163 patients underwent a surgical exploration with intent to cure within the study period. A successful radical re-resection could be undertaken in 127 patients due to evidence of metastatic disease on exploration in the remaining 36 patients.
World J Surg (2014) 38:484–489
(24 %). Of the 14 patients diagnosed on histopathological examination to have liver infiltration, 2 patients had been classified as having pT1, 3 patients as having pT2 disease, and 9 patients as having pT3 disease on histopathological review of their simple cholecystectomy pathology slides/ blocks.
Table 1 Patient demographic and pertinent histopathological data Patient data (N = 127) Median age (years) (range)
50 (24–77)
Sex (male:female)
35:92
Median interval between the two surgeries (mo) (range) Median total lymph nodes dissected (range)
2 (1–10)
Median positive lymph nodes dissected (range)
0 (0–9)
TNM stagea
123
I
24 (19 %)
II
55 (43 %)
IIIA
15 (12 %)
IIIB
30 (24 %)
a
As per the 7th edition of the TNM staging for gallbladder cancer, in three patients the pT could not be assessed accurately (details in text)
Table 2 Patient demographic and pertinent histopathological data from the cohort of 36 patients found to have unresectable disease at exploration Patient data (N = 36) Median age (years) (range)
45 (35–73)
Sex (male:female)
9:27
Median interval between the two surgeries (mo) (range)
2 (1–11)
pT stage (at primary surgery)a T1
6 (17 %)
T2
15 (42 %)
T3 Reasons for unresectabilityb
Patient demography and histopathology The demographic data of the 127 patients is presented in Table 1. The data of the 36 patients who had metastatic disease at the time of exploration and who thus did not undergo a radical re-resection have been shown in Table 2. There were significantly more (p \ 0.02) patients with pT2/ T3 disease in the unresectable group than in the patients who could undergo a successful radical re-resection. Residual disease in the lymph nodes (n = 30) as well as the liver wedge (n = 14) was noted in a total of 31 patients
5 (1–23)
13 (36 %)
Extensive local liver infiltration
8
Distant liver metastases
5
Peritoneal/omental deposits ± ascites
21
Unresectable lymph node disease (extensive disease/para-aortic/paracaval)
13
Port-site diseasec
2
a
As per the 7th edition of the TNM staging for gallbladder cancer, in two patients the pT could not be assessed accurately
b
A single patient may have had more than one site of metastatic disease
c
Associated with liver metastases
World J Surg (2014) 38:484–489
487
0.00
0.25
0.50
0.75
1.00
Kaplan-Meier survival estimate
0 Number at risk 108
6
12
18
24
30
36
41
28
Disease Free Months 89
75
62
54
Fig. 1 Kaplan–Meier survival curves depicting up to 3-year, diseasefree survival of patients undergoing radical re-resection for gallbladder cancer (19 patients did not follow-up after their discharge following the surgery)
A total of 750 lymph nodes were dissected with a median of 5 lymph nodes (range 1–23) per patient. The median lymph node ratio was 0 (mean 0.1; range 0–1). In five patients (4 %), based on the slide/paraffin block provided from their primary surgery, only a positive confirmation of diagnosis of adenocarcinoma could be made. In these patients, the pT stage could not be determined with complete accuracy. All five patients were, however, offered surgery as a precautionary measure and in two of them lymph node metastases was detected. Twenty-five percent (8/32) patients with pT1b disease had lymph node metastases compared with 17 % (12/70) in patients with pT2 disease and 44 % (8/20) with pT3 disease.
Fig. 2 Kaplan–Meier survival curves comparing up to 3-year, disease-free patient survival based on the presence (green line) and absence (blue line) of lymph node metastases. Patients with even a single lymph node metastasis had a significantly poorer outcome (Color figure online)
Adjuvant treatment Fifty-two patients were advised adjuvant treatment, which included 46 patients who received adjuvant chemotherapy, 5 who received chemoradiotherapy, and only 1 patient who received adjuvant radiotherapy. Survival data The median follow-up was 16 months with a 2-year disease-free survival rate of 79.6 % (Fig. 1). Twenty-four patients developed recurrences during the entire follow-up period. Of these three patients, 12.5 % had local recurrences, 18 patients (75 %) had distant recurrent/metastatic disease, and 3 patients (12.5 %) recurred both at local and distant sites. On univariate analysis, 2-year disease-free survival rate was 90 % in patients without lymph node metastases (lymph node ratio, 0) compared with 50 % in patients with
Fig. 3 Kaplan–Meier survival curves comparing up to 3-year, disease-free survival based on the TNM stage
even a single positive lymph node (lymph node ratio, 1; p \ 0.001; Fig. 2). Two-year disease-free survival rates for stage I, II, IIIA, and IIIB were 100, 89, 76.2, and 47.4 % respectively (LR; p \ 0.001; Fig. 3). The use of adjuvant therapy in the postoperative period, age, gender (male vs. female), ASA grade (1 vs. 2), and the time interval between
123
488
the cholecystectomy and radical re-resection (\2 vs. C2 months) did not significantly influence the disease-free survival rates. On multivariate analysis, possible prognostic factors, such as age, gender, ASA grade, time interval between two surgeries, adjuvant treatment, lymph node ratio, and stage of the disease, were included into the model. Lymph node ratio was an independent factor significantly affecting disease-free survival. The hazard ratio for lymph node ratio 1 was 4.2 compared with lymph node ratio 0 [p = 0.002, 95 % confidence interval (CI) 1.7–10.5]. Impact of time interval between two surgeries Analysis of the time interval between the primary cholecystectomy and radical re-resection on a continuous scale as a prognostic factor for recurrence revealed a significant impact of delay in second surgery for every additional month [hazard ratio (HR) 1.22; 95 % CI 1.01–1.47; p = 0.04]. However, after adjusting for stage of disease, this significance disappeared (HR 1.12; 95 % CI 0.95–1.34; p = 0.17).
Discussion These data suggest that lymph node metastasis is the most important predictor of disease recurrence even after complete radical re-resection for incidental gallbladder cancer. Patients with gallbladder cancer tend to have a higher risk of recurrence at distant sites rather than locally. Moreover, patients with pT1b disease do have lymph node metastases and hence should be offered radical re-resection with an aim to perform an adequate lymphadenectomy. The need for radical surgery pT1b gallbladder cancer has been debated with some surgeons suggesting that it is a localized disease [16]. However, we [13] and others [17] previously have suggested the need for radical surgery in this subgroup of patients. The results of the current study thus further lend credibility to the need for radical surgery in patients with CpT1b disease. de Aretxabala et al. [18] recently showed that invasion of the Rokitansky–Aschoff sinuses was associated with a significantly poorer survival even in patients with early gallbladder cancer. The validation of this pathological aspect could lead to it being an important prognostic factor with a potential to guide therapeutic approach. An inadequately appreciated issue in incidental gallbladder cancer is the timing of the radical re-resection following the primary surgery. We have previously shown that successful radical re-resection could be undertaken in selected patients even at 11 months following the primary surgery [13]. The findings of this study clearly indicate that
123
World J Surg (2014) 38:484–489
the risk of recurrence does not depend on the delay of the radical re-resection following the simple cholecystectomy, but rather on the disease stage per se. Thus, radical reresection (especially lymphadenectomy, which has shown to improve survival [19]) should be performed in patients with disease CpT1b even if they present after 2 months of the primary surgery so long as the disease is not metastatic based on an adequate staging performed before revision surgery. Based on the multivariate analysis of factors impacting disease-free survival, the only significant factor was lymph node metastases. Recently, Kishi et al. [20] suggested that the Japanese Society of Biliary Surgery (JSBS) staging systems better stratified the prognosis of patients with gallbladder cancer due to their more accurate definition of T3/4 and regional lymph nodes. Our study, however, clearly indicates that even the presence of a single metastatic lymph node (defined by the TNM staging system as stage IIIB) is associated with a poor outcome irrespective of the T stage compared with node-negative disease (47 vs. [76 %; 2-year disease-free survival). This is unlike the study by Negi et al. [21] who found that in addition to lymph node involvement, T-stage and tumor differentiation also influenced disease-free survival. One potential reason for the 2-year disease-free survival of 79.6 % noted in our study could be a bias towards good outcomes in patients with node-negative disease who formed a large subset in the current cohort (95 patients or 77 %). Another potential hypothesis is whether patients with incidental gallbladder cancer harbor better tumor biology than their counterparts with the same tumor stage who present as a primary gallbladder cancer up front. We have shown previously that not all patients labeled as having incidental gallbladder cancer are truly, incidental [22] based on the fact the pre-cholecystectomy imaging often suggests a possibility of an underlying malignancy and yet these patients tend to be treated with a simple cholecystectomy—with the confirmation of the cancer being made at pathological examination. However, despite this there remain patients who are truly, incidental. It would be interesting to analyse the tumor characteristics in the above two groups of patients. The study is unable to demonstrate a significant difference in disease-free survival for patients depending on adjuvant treatment. This finding is similar to a recent observation by Glazer et al. [23] who observed no survival benefit with adjuvant or neoadjuvant therapy in patients with biliary tract cancers who underwent a complete surgical resection. This cohort included 63 (out of 157) patients with gallbladder cancer. However, our data (Fig. 3) indicates that in patients with node-positive disease (stage IIIB), the majority of recurrences are noted only after the first year of follow-up. Moreover, similar to the
World J Surg (2014) 38:484–489
findings of Jarnagin et al. [24] (85 % distant sites of recurrence), our data suggest that patients with gallbladder cancer are more likely to have their disease fail at distant sites rather than locally (75 %). Thus, rather than the current practice of adjuvant treatment strategies with limited cycles of chemotherapy, the use of prolonged chemotherapy to ‘‘cover’’ the patients with nodal disease in the period around the first year following surgery may help reduce the incidence of recurrences. While routinely available chemotherapy may prove to be toxic for prolonged periods, the chronic administration of chemotherapeutic agents at relatively low, minimally toxic doses, and with no prolonged drug-free breaks a method of dosing more commonly known as ‘‘metronomic chemotherapy’’ may offer some promise [25].
Conclusions The most important predictor of disease recurrence is the presence of lymph node metastases. In patients who undergo curative radical re-resection for incidental gallbladder cancer, recurrent disease is more likely to occur at a distant site. Thus, the role of adjuvant chemotherapy strategies, including metronomic dosing, needs evaluation. Patients with pT1b disease should be offered radical reresection with a radical lymphadenectomy. It is not the duration between the primary surgery and the radical reresection but the stage of the disease that influences outcomes in incidental gallbladder cancer. Acknowledgments The authors would like to thank Dr. Sudeep Gupta, Professor and Medical Oncologist, Department of Medical Oncology, Tata Memorial Centre for constructive academic inputs. Conflict of interest
None.
References 1. Ferlay J, Shin HR, Bray F et al (2010) Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 127:2893–2917 2. Alexander S, Lemmens VE, Houterman S et al (2012) Gallbladder cancer, a vanishing disease? Cancer Causes Control 23:1705–1709 3. Le MD, Henson D, Young H et al (2011) Is gallbladder cancer decreasing in view of increasing laparoscopic cholecystectomy? Ann Hepatol 10:306–314 4. Dhir V, Mohandas KM (1999) Epidemiology of digestive tract cancers in India IV. Gall bladder and pancreas. Indian J Gastroenterol 18:24–28
489 5. Shrikhande S, Barreto S, Singh S et al (2010) Cholelithiasis in gallbladder cancer: coincidence, cofactor, or cause! Eur J Surg Oncol 36:514–519 6. Curado M, Edwards B, Shin H et al (2007) Cancer incidence in five continents. IARC Scientific Publications IX, Lyon 7. Shukla PJ, Neve R, Barreto SG et al (2008) A new scoring system for gallbladder cancer (aiding treatment algorithm): an analysis of 335 patients. Ann Surg Oncol 15:3132–3137 8. Doval D, Sekhon J, Gupta S et al (2004) A phase II study of gemcitabine and cisplatin in chemotherapy-naive, unresectable gallbladder cancer. Br J Cancer 90:1516–1520 9. Boutros C, Gary M, Baldwin K et al (2012) Gallbladder cancer: past, present and an uncertain future. Surg Oncol 21:e183–e191 10. Shrikhande SV, Barreto SG (2009) Surgery for gallbladder cancer: the need to generate greater evidence. World J Gastrointest Surg 1:26–29 11. Shukla PJ, Barreto SG, Arya S et al (2008) Does PET–CT scan have a role prior to radical re-resection for incidental gallbladder cancer? HPB (Oxford) 10:439–445 12. Shukla PJ, Barreto SG, Shrikhande SV et al (2007) Detection of gall bladder cancer metastases in rare sites by PET scan. Indian J Gastroenterol 26:303–304 13. Shukla PJ, Barreto G, Kakade A et al (2008) Revision surgery for incidental gallbladder cancer: factors influencing operability and further evidence for T1b tumours. HPB (Oxford) 10:43–47 14. Sobin L, Gospodarowicz M, Wittekind C (2009) UICC: TNM classification of malignant tumours. Wiley, Oxford 15. Horgan AM, Amir E, Walter T et al (2012) Adjuvant therapy in the treatment of biliary tract cancer: a systematic review and meta-analysis. J Clin Oncol 30:1934–1940 16. Lee SE, Jang JY, Lim CS et al (2011) Systematic review on the surgical treatment for T1 gallbladder cancer. World J Gastroenterol 17:174–180 17. Ouchi K, Suzuki M, Tominaga T et al (1994) Survival after surgery for cancer of the gallbladder. Br J Surg 81:1655–1657 18. de Aretxabala X, Roa I, Hepp J et al (2009) Early gallbladder cancer: is further treatment necessary? J Surg Oncol 100:589–593 19. Shirai Y, Wakai T, Sakata J et al (2012) Regional lymphadenectomy for gallbladder cancer: rational extent, technical details and patient outcomes. World J Gastroenterol 18:2775–2783 20. Kishi Y, Shimada K, Hata S et al (2012) Definition of T3/4 and regional lymph nodes in gallbladder cancer: which is more valid, the UICC or the Japanese staging system? Ann Surg Oncol 19:3567–3573 21. Negi S, Singh A, Choudhary A (2011) Lymph nodal involvement as prognostic factor in gallbladder cancer: location, count or ratio? J Gastrointest Surg 15:1017–1025 22. Shukla PJ, Barreto G, Neve R et al (2007) Can we do better than ‘incidental’ gallbladder cancer? Hepatogastroenterology 54:2184–2185 23. Glazer E, Liu P, Abdalla E et al (2012) Neither neoadjuvant nor adjuvant therapy increases survival after biliary tract cancer resection with wide negative margins. J Gastrointest Surg 16:1666–1671 24. Jarnagin WR, Ruo L, Little SA et al (2003) Patterns of initial disease recurrence after resection of gallbladder carcinoma and hilar cholangiocarcinoma: implications for adjuvant therapeutic strategies. Cancer 98:1689–1700 25. Pasquier E, Kavallaris M, Andre N (2010) Metronomic chemotherapy: new rationale for new directions. Nat Rev Clin Oncol 7:455–465
123
