ORIGINAL ARTICLE ANZJSurg.com

Prognostic factors in 151 patients with surgically resected non-functioning pancreatic neuroendocrine tumours Ki Byung Song,* Song Cheol Kim,* Ji Hoon Kim,† Seung-Mo Hong,† Kwang-Min Park,* Dae Wook Hwang,* Jae Hoon Lee* and Young-Joo Lee* *Division of Hepatobiliary and Pancreatic Surgery, Department of Surgery, Ulsan University College of Medicine and Asan Medical Center, Seoul, South Korea and †Department of Pathology, Ulsan University College of Medicine and Asan Medical Center, Seoul, South Korea

Key words classification, disease-specific survivor, non-functioning pancreatic neuroendocrine tumour, prognostic factor, stage. Correspondence Professor Song Cheol Kim, Division of Hepatobiliary and Pancreatic Surgery, Department of Surgery, Ulsan University College of Medicine and Asan Medical Center, 86 Asan Byeongwon-gil, Songpa-gu, Seoul 138-736, Korea. Email: [email protected] K. B. Song MD, PhD; S. C. Kim MD, PhD; J. H. Kim MD, PhD; S.-M. Hong MD, PhD; K.-M. Park MD, PhD; D. W. Hwang MD, PhD; J. H. Lee MD, PhD; Y.-J. Lee MD, PhD. Accepted for publication 25 May 2014. doi: 10.1111/ans.12738

Abstract Background: Recently, non-functioning pancreatic neuroendocrine tumors (NF-PNETs) are increasing. It is important to know about the prognostic factors and long-term survival rates in patients with NF-PNET for the management of these diseases. Methods: We retrospectively analysed the records of 151 patients with NF-PNET who had pancreatic resection between January 1995 and December 2010. Result: The 10-year overall survival, disease-specific survival and disease-free survival rate of the patients who underwent surgery for NF-PNET were 72.6%, 85.1% and 57.2% respectively. The three staging systems (2004 World Health Organization classification, 2006 European Neuroendocrine Tumor Society stages and 2010 grading system) showed comparable prognostic relevance in disease-specific survival of patients with resected NF-PNETs. Lymph nodes metastasis, high Ki-67 index and mitotic rate were the independent poor prognostic factors for disease-specific survival in the patients with surgically resected NF-PNET on multivariate analysis. Conclusion: We suggested that the three staging systems accurately reflect the prognosis in disease-specific survival of patients with resected NF-PNETs. Presence of lymph nodes metastases and high Ki-67 index and mitotic rate were the independent poor prognostic factors after resection of NF-PNET.

Introduction Neuroendocrine neoplasms, defined as epithelial neoplasms with neuroendocrine differentiation, can occur in many organs. Pancreatic neuroendocrine tumours (PNETs) are uncommon and slowgrowing neoplasms of the pancreas. Their incidence is estimated to range from four to five cases per 1 million persons in the United States,1 which accounts for approximately 1–2% of all pancreatic neoplasms. However, the detection of PNET has increased recently, most likely because of the easy availability and widespread use of advanced imaging modality. During the past few years, the percentage of patients diagnosed with non-functioning pancreatic neuroendocrine tumours (NF-PNET) has increased significantly.2,3 NF-PNETs represent one of the most aggressive type of gastroenteropancreatic neuroendocrine tumour.3 Although, some studies have been reported to evaluate the survival rate for patients with PNET based on the 2004 World Health Organization (WHO) classifications, the 2006 European Neuroendocrine Tumor Society © 2014 Royal Australasian College of Surgeons

(ENETS) stages and the 2010 WHO grades, there are few studies with adequate number of surgically resected NF-PNET and longterm follow-up. Moreover, the predictive factors for disease-specific survival in patients with NF-PNET are not fully understood. The objective of this study was to assess the clinical relevance of all three classification/grading systems and to identify factors predicting long-term survival for surgically resected NF-PNET in a large number of patients.

Methods Two hundred and twenty-five patients underwent PNET surgery at our institution between January 1995 and December 2010. We retrospectively analysed the records of 151 patients with NF-PNET who had pancreatic resection. All patients were followed-up by a return visit, correspondence with primary physicians or telephone contact during the final month of this study (September 2013). The follow-up examination included a clinical examination, blood ANZ J Surg •• (2014) ••–••

2

tests, abdominal ultrasonography and computed tomography and chest radiography. All patient follow-ups were completed by 30 April 2013, and the median follow-up period was 70.4 (range: 2.6– 221.8) months. Complete follow-up data were obtained for all patients.

Pathological assessment Diagnosis was based on histopathology and immunohistochemistry results, and a tumour was considered to be a NF-PNET in the absence of hormonal excess and hormone-related symptoms. Patient records were reviewed with respect to patient demographics, clinical features, pathological findings, surgical details and outcome, medical treatment and long-term survival. Patients with multiple endocrine neoplasia type I syndrome, von Hippel-Lindau disease, tuberous sclerosis and neurofibromatosis were excluded. NF-PNET staging was based on the 2004 WHO classification4 and 2010 WHO grading system5 and 2006 ENETS stages.6 In 2004, the WHO introduced a system for naming and classifying endocrinological tumours of the pancreas into four categories: (1) WHO1, welldifferentiated (WD)-PNET (i) WHO1a, a WD-PNET (benign); (ii) WHO1b, a WD-PNET (uncertain); (2) WHO2, a WD-pancreatic neuroendocrine carcinoma (PNECa); and (3) WHO3, a poorly differentiated-PNECa. In 2006, the ENETS introduced staging for gastroenteropancreatic neuroendocrine tumours as follows: (i) stage I, a primary tumour confined to the pancreas and 4 cm or one that invades the duodenum or bile duct; (iv) stage IIIa, a tumour invading adjacent organs (i.e. stomach, spleen, colon or adrenal gland) or the wall of large vessels (i.e. celiac axis or superior mesenteric artery); (v) stage IIIb, a tumour with lymph node metastasis; and (vi) stage IV, a tumour with distant metastasis. In 2010, the WHO updated its classifications by including mitotic and Ki-67 proliferation indexes.

Treatment strategy Patients with a tumour in the head of the pancreas underwent standard pancreaticoduodenectomy or pylorus-preserving pancreaticoduodenectomy, whereas those with a tumour in the body or tail underwent distal pancreatectomy with or without splenectomy. When a tumour extended past the neck of the pancreas, margined by the left border of the superior mesenteric vessels, and into the gland body, total pancreatectomy was performed. When the lesion was small, single, capsulated and did not involve the main pancreatic duct, enucleation was performed. Aggressive treatment of liver metastases included surgery, transarterial chemoembolization or radiofrequency ablation. When complete pancreatic resection (R0) was not possible or when hepatic metastasis or advanced PNET was present, adjuvant therapy such as antihormonal therapy with longacting octreotide (Sandostatin LAR, Novartis, Basel, Switzerland) systemic chemotherapy and external beam radiation were performed.

Statistical analyses Mean, median, standard deviation, frequencies and range were obtained as descriptive statistics. Groups were compared with Student’s t-test or Mann–Whitney U-test for continuous variables.

Song et al.

Qualitative data were compared with chi-square test and Fisher’s exact test as needed. All time points were reported in months, with the date of surgery taken as day 0 and the date of last follow-up or death taken as the end point. Disease-specific mortality was defined as a death caused by a tumour. Patients who died of other causes, regardless of disease status, were censored at the time of death in the analysis of disease-specific mortality. The end point for univariate analyses was disease-specific survival. Survival curves were generated by the Kaplan–Meier method, and the log-rank test was used to compare survival curves. Statistical analyses were performed with SPSS 21.0 for Windows (SPSS Inc., Chicago, IL, USA). A P value of 20%), (iv) distant metastasis (yes or no), and (v) resection margin status (R0 versus R1 or R2).

Results Clinical characteristics Two hundred and twenty-five patients underwent PNET surgery at the Asan Medical Center between January 1990 and December 2010. The percentage of NF-PNET patients was 67.1% (151 of 225). Of 151 patients that underwent surgery for NF-PNET, there were 65 men (43%) and 86 women (57%). The median age was 53 (range: 17–77) years. The most common symptom for NF-PNET was abdominal pain (29.1%). Seventy-one patients (47%) showed no disease symptoms. The median size of tumours was 2.6 cm (range: 0.5–15) in diameter. The most common tumour occurrence site was the head of pancreas.

Operation and adjuvant therapy The most frequent operation was pancreaticoduodenectomy (64 cases, 42.4%). Resection of adjacent organs, such as the adrenal gland, transverse colon, kidney, stomach and portal vein, was performed in a few cases (12 cases, 7.9%), and three patients (2%) underwent hepatic resection for metastatic synchronous disease. The rates of laparoscopic pancreatic resection and pancreas preserving pancreatectomy (central pancreatectomy and enucleation) in patients with NF-PNET were 21.2% (32 of 151) and 13.2% (20 of 151) respectively. Curative primary surgery (R0) was performed in 139 cases (92%), whereas R1 and R2 resections were performed in eight and four cases respectively. The recurrent rate was 14.3% (21 of 147), and the median recurrence time was 16.1 months (range: 3–87.2). Recurrences were not calculated for patients who underwent R2 resection. The liver was the most common recurrent site (18 of 21, 85.7%). Twelve patients underwent adjuvant therapy. Specifically, four patients received adjuvant chemotherapy, five patients received long-acting octreotide and three patients received postoperative external beam radiation. © 2014 Royal Australasian College of Surgeons

Non-functioning pancreatic neuroendocrine tumour

3

Table 1 Univariate and multivariate analysis of disease-specific survival of the patients with non-functioning pancreatic neuroendocrine tumour Disease-specific survival

Univariate analysis

(15/151) Ki-67 index

Lymph node Mitosis

Resection Margin status Distant metastasis

Hazard ratio ≤2 3–20 >20 Negative Positive 20 R0 R1 or R2 Yes No

120 22 9 123 23 113 33 5 139 12 146 5

1 9.984 30.198 1 12.149 1 9.423 44.909 1 8.905 1 15.253

Multivariate analysis

95% confidence interval –– 2.380 7.197 –– 3.802 –– 2.820 9.464 –– 3.150 –– 4.049

–– 41.888 126.708 –– 38.828 –– 31.489 213.111 –– 25.180 –– 57.457

P value

Hazard ratio

Prognostic factors in 151 patients with surgically resected non-functioning pancreatic neuroendocrine tumours.

Recently, non-functioning pancreatic neuroendocrine tumors (NF-PNETs) are increasing. It is important to know about the prognostic factors and long-te...
170KB Sizes 0 Downloads 3 Views