Scandinavian Journal of Gastroenterology. 2014; 49: 734–741

ORIGINAL ARTICLE

Survival and prognostic factors in well-differentiated pancreatic neuroendocrine tumors

RAZIYE BOYAR CETINKAYA1, MORTEN VATN1,3, LARS AABAKKEN1,2, DEIDI S. BERGESTUEN1 & ESPEN THIIS-EVENSEN1 1

Department of Gastroenterology, Rikshospitalet, Oslo University Hospital, Oslo, Norway, 2University of Oslo, Institute of Clinical Medicine, OUS, Rikshospitalet, Norway, and 3University of Oslo, Institute of Clinical Medicine, Campus Ahus, Nordbyhagen, Norway

Abstract Objectives. Pancreatic neuroendocrine tumors (P-NETs) account for 2–3% of all pancreatic neoplasms. The aim of our study was to investigate survival and prognostic factors in patients with P-NETs. Material and methods. We retrospectively reviewed the medical records of 114 patients diagnosed with well-differentiated P-NETs from 1982 through 2010. We studied demographical, clinical, radiological, and histopathological characteristics. Results. Median age at diagnosis was 57 years (range 32–83); 53% were men and 78% had nonfunctional tumors. The most common presenting symptoms were abdominal pain (41%), weight loss (36%), and diarrhea (25%); 19% of the tumors were incidental findings. Median duration of symptoms before diagnosis was 4 months, 3 months for nonfunctional, and 12 months for functional tumors (p < 0.001). At diagnosis, 32.5% patients had local, 22.8% had regional, and 44.7% had distant disease. Men had more often distant disease at diagnosis (p = 0.02). Median survival was 6.6 years (95% confidence interval [CI]: 4.0–9.2). Overall 5-year survival was 53.9% (95% CI: 43.4–63.3). For those with local disease and those with distant disease at diagnosis, the 5-year survivals were 70.2% (95% CI: 49.9–83.6) and 33.0% (95% CI: 19.7–46.7), respectively. Surgery with curative intent was performed on 46 patients. Newly detected liver metastases were diagnosed up to 10 years after surgery in this group. In patients with metastases, palliative surgery (debulking) did not have a significant effect on survival compared to those who had no surgery. Conclusion. Distant metastases, Ki-67 > 2%, nonfunctional tumors, elevated level of chromogranin A, and palliative treatment were associated with poor survival.

Key Words: neuroendocrine tumors, pancreatic neoplasms, prognosis, survival analysis

Introduction Pancreatic neuroendocrine tumors (P-NETs) are rare with an incidence of 0.3 per 100,000 [1], comprising 2 to 3% of all pancreatic tumors [2,3]. While the incidence of most other pancreatic neoplasm is stable, recent studies show a rising incidence of P-NETs [4– 6]. The incidence is probably even higher as an autopsy study by Kimura et al. found neuroendocrine tumors (NET) in 10% of the cases [7]. An increasing number of P-NETs are detected incidentally, probably due to the widespread use of abdominal imaging.

Although P-NETs usually have metastasized at diagnosis, they have a better prognosis than adenocarcinoma [3]. Compared to other localizations of NET, P-NETs seem to have a poor prognosis [1,8,9]. The aim of our study was to investigate survival and prognostic factors in patients with well-differentiated P-NETs treated at a tertiary referral center. Methods Our hospital is a tertiary referral center, serving approximately 2.7 million people, 54% of the total

Correspondence: Raziye Boyar Cetinkaya, MD, Department of Gastroenterology, Rikshospitalet, Oslo University Hospital, 4950 Nydalen, 0424 Oslo, Norway. Tel: +47 23071961. Fax: +47 23070670. E-mail: [email protected].

(Received 19 December 2013; revised 10 February 2014; accepted 8 March 2014) ISSN 0036-5521 print/ISSN 1502-7708 online  2014 Informa Healthcare DOI: 10.3109/00365521.2014.903432

Prognosis in pancreatic neuroendocrine tumors Norwegian population. Altogether 714 patients with NETs, of whom 114 (16%) had well-differentiated P-NETs, were registered from 1982 through 2010. Demographical, clinical, radiological, and histopathological characteristics were prospectively recorded and included in a database. During follow up, most of the patients underwent CT scan with arterial and portal venous phases. MRI and ultrasound were used to a lesser extent. Variables registered included age at diagnosis, gender, symptoms, biochemistry, tumor size, proliferation index, stage at diagnosis and therapy given. A pancreatic tumor was defined as functional when a clinical syndrome of hormonal overexpression was evident. The proliferation index, Ki-67, was retrieved from pathology reports of surgical specimens or liver biopsies. In cases where more than one tissue sample were analyzed for Ki-67, the highest value was used. The Ki-67 was not obtained in 35 patients, mainly due to lack of remaining tumor specimen for the analyses. The Ki-67 value was used to grade the tumor according to WHO (2010) and the European Neuroendocrine Tumors Society (ENETS) grading systems: grade 1 Ki-67 < 2%, grade 2 Ki-67 3–20%, and grade 3 with Ki-67 > 20% [10]. Only grade 1 and grade 2 tumors were included in our study. Stage was divided into local, regional, and distant disease. Local disease was defined as the tumor confined to the pancreas. Regional disease was defined as extension of tumor into adjacent tissue or spread of tumor to regional lymph nodes and distant disease was defined as the presence of metastases in other parts of the body [9]. Debulking was performed if the patient was in good clinical condition and the primary tumor was regarded as resectable and/or more than 90% of the liver metastases were resectable. Chromogranin A (CgA) was analyzed using different laboratory methods over the study period. To compare results, CgA was expressed as ratio (measured value:upper limit of normal). In case of several measurements, we used the first measured values of CgA. For survival analyses, we obtained information about time of death by linkage of our data to Statistics Norway. Patients are identified in the registry by their individually national registration number, which includes date of birth. Statistics Mann–Whitney, Kruskal–Wallis and chi-square were used to compare groups. Kaplan–Meier and log-rank were used for survival analyses. Multivariate analysis was considered but could not be done due to small numbers. The p-Values were two-sided and considered statistically significant when 40 mm Stage at diagnosis Local Regional Distant Liver Liver + other distant Distant without liver Ki-67% (n = 79) 2) in 87.8% of the cases. Only two patients had distant disease without liver involvement. Of the patients with local disease at diagnosis, 1 (3%) patient later developed regional disease and 16 (43%) patients developed distant disease. Of those with regional disease at diagnosis, 14 (54%) patients developed distant disease, mainly liver metastases. CgA was elevated in 85.5% of the patients with median ratio of 3.3 (range 0–640). CgA ratio increased with increasing stage (p < 0.001), number of liver metastases (p < 0.001) and tumor size > 40 mm (p = 0.03). There was no difference in CgA, whether the tumor was functional or not. Somatostatin receptor scintigraphy results were available for 88 patients with 80 (91%) patients having positive results. Treatment Tumor-directed therapy was given to 110 (96%) patients. The treatment modalities used are presented in Table III. Surgery was performed in 63 patients, 46 (42%) patients with curative intent and 17(15%) patients with palliative intent (debulking). Among the 46 patients with intended curative surgery, 27 patients had local, 14 patients had regional, and 5 patients had distant disease at diagnosis. In 10 (37%) of the 27 patients with local disease, distant metastases were later detected with a median time of 14.5 months (range 3–126) after surgery. Among the 14 patients with regional disease, who received surgery with curative intent, 8 (57%) patients developed distant disease after a median time of 22 months (range 3–57).

Table IV. Survival. Variables Total group Stage Local Regional Distant Functionality Functional Nonfunctional Ki-67% 2% predicted poorer prognosis (p = 0.04) with reduced 5- and 10-year survival rates (Figure 2). Patients with functional tumors had better 5- and 10-year survival rates than those with nonfunctional tumors, p = 0.01 and p = 0.03, respectively (Figure 3). High CgA level at diagnosis was associated with reduced survival (p = 0.008). Survival was reduced in those who received debulking surgery compared to those who had surgery with a curative intent (p < 0.001). There was, however, no statistically significant difference in survival between

1

Overall survival

.75

.5

.25

0 0

50

100

150

Time (months) 95% CI Ki7% < = 2

95% CI Ki7% < = 3–20

Figure 2. Survival rate according to Ki-67 is shown. Log-rank p = 0.04.

738

R. Boyar Cetinkaya et al. 1

Overall survival

.75

.5

.25

0 0

50

100

150

Time (months) 95% CI F

95% CI NF

Figure 3. Survival rate according to functionality is shown. Log rank p = 0.01.

patients who received debulking surgery compared to patients who did not receive surgery (p = 0.32). There was, however, a trend toward better short-time survival (Figure 4). Discussion This study presents all patients with P-NETs followed at our institution from 1982 through 2010. Patients were prospectively registered and included in a database; hence very few patients admitted to our hospital were missed. As a tertiary referral center, our patients are selected and do not give a complete overview over

this patient category. Some patients with tumors that were completely resected at local hospitals, and some patients in their terminal stage of the disease, are probably not referred. As a consequence, we might have missed some of the patients with the best, and some with the worst, prognosis. Median survival for our study population was 6.6 years. The 5-year overall survival rate was 54%. Bergestuen et al. reported 258 small intestinal NETs (SI-NETs) in our center and found that median survival was 9.3 years [11]. The 5-year overall survival for SI-NETs was 78% which is better than P-NETs. This is in accordance with other studies that report a

1 .9

Overall survival

.8 .7 .6 .5 .4 .3 .2 .1 0 0

6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108114120

Analysis time (months) 95% CI Primary

95% CI Debulking

95% CI No surgery

Figure 4. Survival rate and treatment modality are shown. Log rank p < 0.001. Primary surgery with curative intent shows better survival compared to debulking and no surgery, but no significant difference between the latter two (p = 0.33).

Prognosis in pancreatic neuroendocrine tumors better 5-year overall survival of 68–90% in gastrointestinal NET versus 38–74% in P-NET [8,12,13]. In our study, men had a more advanced stage at diagnosis and reduced survival compared to women, although the latter did not reach statistical significance. The difference in stage could not be explained by patient delay as men had a trend toward shorter duration of symptoms before diagnosis. There was, however no difference in Ki-67 in regard to gender. There are studies that reports that men have reduced overall survival [8,9,14,15]. In a population-based study [9], men were more likely to present with metastases, but to our knowledge, there are no other single center reports with the finding of more advanced disease in men. These findings might be an indication of men having a more aggressive course. Almost 80% of our patients had a nonfunctional tumor which is in accordance with the findings in other studies [2,16–18]. Duration of symptoms before diagnosis was longer for functional tumors, as found by Wang et al. [19]. Despite longer duration of symptoms, functional tumors had better prognosis in our population. Functionality and its prognostic role has been a matter of debate. Whereas some studies report better prognosis with functional tumors [14,19–21], others show no difference [18,22–25]. To our knowledge, only one study reports better prognosis in patients with nonfunctioning tumors [26]. In our study, functional tumors were more frequent among women, patients with lower age, and lower grade at diagnosis. There was no difference in tumor size and stage between functional and nonfunctional tumors. This might indicate that the malignant potential of functional tumors differs from the nonfunctional tumors. Distant metastases were present at diagnosis in 45% of the patients. During the follow up, another 25% of the patients developed distant metastases, most often in the liver. Kim et al. compared the clinical course of NETs with different sites of origin and showed that whereas small intestinal NET most often recur locally, P-NETs recur in the liver [8]. Some of the patients were diagnosed with newly discovered liver metastases up to 10 years after surgery with curative intent. This supports the recommendation for long-time surveillance, even in those that are considered cured by surgery. Distant disease was a predictor of poor survival, but we found no significant difference in survival between patients with local and regional disease. Stage is found to be a predictor of survival either if TNM with grade I–IV or local/regional/distant is used [9,27]. Halfdanarson analyzed P-NETs in the SEER registries and found that distant disease significantly reduced survival [14]. They found, however, no difference

739

in survival between local and regional disease. Ekeblad et al. analyzed 324 P-NETs and they did not show any survival difference between TNM stages I versus II and III [22]. In their multivariate analysis, there was only a significant difference between metastatic disease (stage IV) and any other stage. As reported by others [13,22,28], we found reduced survival in patients with Ki-67 > 2%. Panzuto et al. analyzed the factors associated with disease progression and found a 2% increased risk of progression with every unit increase in Ki-67 [29]. As found in our patients, high Ki-67 is also associated with more advanced stage [16]. Whether debulking surgery should be done for metastatic P-NETs is an ongoing discussion as the advantage in survival has not been clearly shown. According to ENETS guidelines [10], debulking of an unresectable primary tumor is advised in selected cases to avoid tumor-related complications. Surgery of liver metastases has been recommended if 90% of the tumor mass can be reduced. In our population, patients who received surgery with curative intent had, as expected, a better survival than non-curative treatment. However, there was no statistical difference in survival among patients who had debulking surgery compared to patients who did not receive surgery at all. There was, however, a non-statistical difference in favor of debulking in the first 5 years after surgery. P-NET is a rare disease and the number of patients in our study, and in most published reports, is small, which makes it difficult to demonstrate the differences as statistical significance. We do lack some information; especially Ki-67 values have been difficult to retrieve. Ki-67 was not a routine test until the past decade. Further, some of the biopsies were too small to be analyzed for Ki-67. Due to missing Ki-67 values for many of our patients, we might have included some grade 3 tumors in our analyses. This could have influenced the survival data. There has been little change in the treatment options for P-NETs in our cohort, until recently. We report results from patients treated through 2010. For the past few years, an increasing number of patients at our institution have been treated with peptide receptor radionuclide therapy, temozolomide/capecitabine combination, everolimus and sunitinib. We, therefore, hope that the survival in advanced P-NET might be better in future reports. In conclusion, 45% of the patients in our cohort had distant metastases at diagnosis. The median survival was 6.6 years with men having more advanced disease at diagnosis and there was trend toward reduced survival. In some of our patients, new liver metastases were diagnosed up to 10 years after

740

R. Boyar Cetinkaya et al.

surgery with curative intent, emphasizing the importance of long-time follow up in this patient group. Patients with metastases who received debulking surgery did not have a favorable survival compared to patients who were not operated on; however, a trend toward improved short-term prognosis may support debulking as a treatment option. Distant metastases, Ki-67 > 2%, nonfunctional tumors, elevated CgA, and non-curative treatment were associated with poor survival.

Acknowledgments The authors thank Kristian Holm for technical assistance with the database and Milada Småstuen for assistance with the survival analyses. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. This work was supported by South-Eastern Norway Regional Health Authority.

References [1] Hauso O, Gustafsson BI, Kidd M, Waldum HL, Drozdov I, Chan AK, et al. Neuroendocrine tumor epidemiology: contrasting Norway and North America. Cancer 2008;113: 2655–64. [2] Figueiredo FA, Giovannini M, Monges G, Charfi S, Bories E, Pesenti C, et al. Pancreatic endocrine tumors: a large single-center experience. Pancreas 2009;38:936–40. [3] Fesinmeyer MD, Austin MA, Li CI, De Roos AJ, Bowen DJ. Differences in survival by histologic type of pancreatic cancer. Cancer Epidemiol Biomarkers Prev 2005;14:1766–73. [4] Fitzgerald TL, Hickner ZJ, Schmitz M, Kort EJ. Changing incidence of pancreatic neoplasms: a 16-year review of statewide tumor registry. Pancreas 2008;37:134–8. [5] Franko J, Feng W, Yip L, Genovese E, Moser AJ. Nonfunctional neuroendocrine carcinoma of the pancreas: incidence, tumor biology, and outcomes in 2,158 patients. J Gastrointest Surg 2010;14:541–8. [6] Martin RC, Kooby DA, Weber SM, Merchant NB, Parikh AA, Cho CS, et al. Analysis of 6,747 pancreatic neuroendocrine tumors for a proposed staging system. J Gastrointest Surg 2011;15:175–83. [7] Kimura W, Kuroda A, Morioka Y. Clinical pathology of endocrine tumors of the pancreas. Analysis of autopsy cases. Dig Dis Sci 1991;36:933–42. [8] Kim SJ, Kim JW, Oh DY, Han SW, Lee SH, Kim DW, et al. Clinical Course of Neuroendocrine Tumors With Different Origins (the Pancreas, Gastrointestinal Tract, and Lung). Am J Clin Oncol 2011;35:549–56. [9] Yao JC, Hassan M, Phan A, Dagohoy C, Leary C, Mares JE, et al. One hundred years after "carcinoid": epidemiology of and prognostic factors for neuroendocrine tumors in 35,825 cases in the United States. J Clin Oncol 2008;26: 3063–72.

[10] Falconi M, Bartsch DK, Eriksson B, Kloppel G, Lopes JM, O’Connor JM, et al. ENETS Consensus Guidelines for the management of patients with digestive neuroendocrine neoplasms of the digestive system: well-differentiated pancreatic non-functioning tumors. Neuroendocrinology 2012; 95:120–34. [11] Bergestuen DS, Aabakken L, Holm K, Vatn M, ThiisEvensen E. Small intestinal neuroendocrine tumors: prognostic factors and survival. Scand J Gastroenterol 2009;44: 1084–91. [12] Lawrence B, Gustafsson BI, Chan A, Svejda B, Kidd M, Modlin IM. The epidemiology of gastroenteropancreatic neuroendocrine tumors. Endocrinol Metab Clin North Am 2011;40:1–18; vii. [13] Panzuto F, Nasoni S, Falconi M, Corleto VD, Capurso G, Cassetta S, et al. Prognostic factors and survival in endocrine tumor patients: comparison between gastrointestinal and pancreatic localization. Endocr Relat Cancer 2005;12: 1083–92. [14] Halfdanarson TR, Rabe KG, Rubin J, Petersen GM. Pancreatic neuroendocrine tumors (PNETs): incidence, prognosis and recent trend toward improved survival. Ann Oncol 2008;19:1727–33. [15] Caldarella A, Crocetti E, Paci E. Distribution, incidence, and prognosis in neuroendocrine tumors: a population based study from a cancer registry. Pathol Oncol Res 2011;17: 759–63. [16] Zerbi A, Falconi M, Rindi G, Delle Fave G, Tomassetti P, Pasquali C, et al. Clinicopathological features of pancreatic endocrine tumors: a prospective multicenter study in Italy of 297 sporadic cases. Am J Gastroenterol 2010;105:1421–9. [17] Bilimoria KY, Tomlinson JS, Merkow RP, Stewart AK, Ko CY, Talamonti MS, et al. Clinicopathologic features and treatment trends of pancreatic neuroendocrine tumors: analysis of 9,821 patients. J Gastrointest Surg 2007;11:1460– 7; discussion 1467-1469. [18] Oh TG, Chung MJ, Park JY, Bang SM, Park SW, Chung JB, et al. Prognostic factors and characteristics of pancreatic neuroendocrine tumors: single center experience. Yonsei Med J 2012;53:944–51. [19] Wang SE, Su CH, Kuo YJ, Shyr YM, Li AF, Chen TH, et al. Comparison of functional and nonfunctional neuroendocrine tumors in the pancreas and peripancreatic region. Pancreas 2011;40:253–9. [20] Bilimoria KY, Talamonti MS, Tomlinson JS, Stewart AK, Winchester DP, Ko CY, et al. Prognostic score predicting survival after resection of pancreatic neuroendocrine tumors: analysis of 3851 patients. Ann Surg 2008;247:490–500. [21] Eriksson B, Oberg K, Skogseid B. Neuroendocrine pancreatic tumors. Clinical findings in a prospective study of 84 patients. Acta Oncol 1989;28:373–7. [22] Ekeblad S, Skogseid B, Dunder K, Oberg K, Eriksson B. Prognostic factors and survival in 324 patients with pancreatic endocrine tumor treated at a single institution. Clin Cancer Res 2008;14:7798–803. [23] Gao C, Fu X, Pan Y, Li Q. Surgical treatment of pancreatic neuroendocrine tumors: report of 112 cases. Dig Surg 2010; 27:197–204. [24] Hill JS, McPhee JT, McDade TP, Zhou Z, Sullivan ME, Whalen GF, et al. Pancreatic neuroendocrine tumors: the impact of surgical resection on survival. Cancer 2009;115: 741–51. [25] Martin-Perez E, Capdevila J, Castellano D, JimenezFonseca P, Salazar R, Beguiristain-Gomez A, et al. Prognostic Factors and Long-Term Outcome of Pancreatic

Prognosis in pancreatic neuroendocrine tumors Neuroendocrine Neoplasms: Ki-67 Index Shows a Greater Impact on Survival than Disease Stage. The Large Experience of the Spanish National Tumor Registry (RGETNE). The Large Experience of the Spanish National Tumor Registry (RGETNE). Neuroendocrinology 2013; 98:156–68. [26] Fischer L, Kleeff J, Esposito I, Hinz U, Zimmermann A, Friess H, et al. Clinical outcome and long-term survival in 118 consecutive patients with neuroendocrine tumours of the pancreas. Br J Surg 2008;95:627–35. [27] Strosberg JR, Cheema A, Weber J, Han G, Coppola D, Kvols LK. Prognostic validity of a novel American Joint

741

Committee on Cancer Staging Classification for pancreatic neuroendocrine tumors. J Clin Oncol 2011;29:3044–9. [28] Pomianowska E, Gladhaug IP, Grzyb K, Rosok BI, Edwin B, Bergestuen DS, et al. Survival following resection of pancreatic endocrine tumors: importance of R-status and the WHO and TNM classification systems. Scand J Gastroenterol 2010;45:971–9. [29] Panzuto F, Boninsegna L, Fazio N, Campana D, Pia Brizzi M, Capurso G, et al. Metastatic and locally advanced pancreatic endocrine carcinomas: analysis of factors associated with disease progression. J Clin Oncol 2011; 29:2372–7.

Copyright of Scandinavian Journal of Gastroenterology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.