- Email: [email protected]
Repeat hepatectomy for patients with recurrent neuroendocrine liver metastasis: Comparison with first hepatectomy
Surgery xxx (2019) 1e6
Contents lists available at ScienceDirect
Surgery journal homepage: www.elsevier.com/locate/surg
Repeat hepatectomy for patients with recurrent neuroendocrine liver metastasis: Comparison with first hepatectomy Sho Kiritani, MD, Junichi Arita, PhD, Masaru Matsumura, MD, Yujiro Nishioka, PhD, Hiroki Kudo, MD, Akihiko Ichida, PhD, Takeaki Ishizawa, PhD, Nobuhisa Akamatsu, PhD, Junichi Kaneko, PhD, Kiyoshi Hasegawa, PhD* Hepato-Biliary-Pancreatic Surgery Division, Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 20 August 2019 Available online xxx
Background: The value of repeat hepatectomy for intrahepatic recurrence after curative resection of neuroendocrine liver metastasis (NELM) remains unclear. The aim of this retrospective cohort study was to determine the significance of repeat hepatectomy for recurrent NELM. Methods: Patients who underwent hepatectomy for NELM between 1994 and 2016 were identified. The indications for a first hepatectomy were adequate liver remnant volume and no extrahepatic metastasis. The diagnosis of recurrent NELM was based on radiographic examinations. The indications for a repeat hepatectomy were the same as those for the first hepatectomy. Clinicopathologic factors, short-term survival, and long-term survival were investigated using clinical records. Results: Forty-four patients enrolled in this study. Thirty-three patients among them underwent a curative hepatectomy, and 28 of them developed recurrence. Of them, 16 patients underwent a repeat hepatectomy. The overall survival of the repeat hepatectomy cohort (n ¼ 16) was significantly better than that of the no repeat hepatectomy cohort (n ¼ 12) (P < .001). The progression free survival after the first hepatectomy (n ¼ 44) and that after repeat hepatectomy (n ¼ 16) were similar (P ¼ .546). No significant difference was seen between the frequency of major complications (Clavien-Dindo score 3a) after the first and repeat hepatectomy (P ¼ .279). No repeat hepatectomy (hazard ratio [HR] 5.0, P ¼ .036) was identified as an independent predictive factor of a poor outcome among the recurrent cohort, along with the presence of multiple nodules (HR 26.2, P ¼ .008) and a CA19-9 level 40 U/mL (HR 11.2, P ¼ .012). Conclusions: A repeat hepatectomy is feasible in selected patients with recurrent NELM. © 2019 Elsevier Inc. All rights reserved.
Introduction Although neuroendocrine tumors (NETs) are rare with an incidence of about 5 per 100,000 according to a large exploratory report of The Surveillance, Epidemiology, and End Results data by Yao et al,1 the recognized incidence of this disease increased by about 6.4-fold from 1973 to 2012.1e3 Approximately 60% to 90% of patients with NETs have synchronous or metachronous neuroendocrine liver metastasis (NELM), and many of these patients have constitutional symptoms and a poor prognosis.1,4e7 Populationbased studies have reported that the median overall survival (OS) for patients with metastatic pancreatic and small bowel NETs are 24 * Reprint requests: Kiyoshi Hasegawa, PhD, Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail address: [email protected] (K. Hasegawa). https://doi.org/10.1016/j.surg.2019.08.020 0039-6060/© 2019 Elsevier Inc. All rights reserved.
and 56 months, respectively.1,8 Operative resection is recommended if curative resection is possible; however, surgery with curative intent was reportedly adopted for only half or less of all cases at the time of the diagnosis of NELM.4,9,10 The Cochrane systematic review suggested that the better prognosis after hepatic resection for NELM was attributable to a selection bias.11 Thus, the operative indications for NELM have been controversial. Recurrence after resection is common and decreases both the quality of life and the long-term survival of patients. Intrahepatic recurrence after a curative hepatectomy for NELM can be as high as 70% to 94% at 5 years11e13; however, the optimal treatment strategy for patients with recurrent tumors has remained unclear. Several medications, including somatostatin analogues, everolimus, sunitinib, and streptozocin, have been used for the treatment of NETs, which paradoxically complicate the selection of the optimal treatment option.14e18 Because few reports assessing recurrent NELM have been published, the purpose of
2
S. Kiritani et al. / Surgery xxx (2019) 1e6
the present study was to determine the oncologic impact of repeat hepatectomy for recurrent NELM after a first hepatectomy.
Patients and Methods This retrospective cohort study was approved by the Institutional Ethics Committee of Tokyo University. An opt-out manuscript for this study was published on our website (http://www.u-tokyohbp-transplant-surgery.jp/). None of the patients were opposed to being included in the present study. All operative candidates with NETs who were treated between January 1994 and December 2016 at our department were identified from the maintained database. Among them, all the patients undergoing resection for NELM were included in this study. Contrast-enhanced computed tomography or magnetic resonance imaging of the chest, abdomen, and pelvis using a triphasic liver protocol was performed for all the patients to assess resectability and extrahepatic disease. Neoadjuvant treatment was not considered, in principle. The indications for hepatectomy were as follows: (1) an acceptable general condition, (2) a favorable Child-Pugh Classification (A plus selected class B) and indocyanine green retention rate at 15 min, (3) adequate liver remnant volume,19 and (4) no extrahepatic lymph node or organ metastasis. In addition, hepatic resection was indicated for patients with hormonal symptoms even if an R2 resection was expected. A R1 resection was allowed when the resection of all the tumors was otherwise impossible. In principle, parenchymasparing, nonanatomic hepatic resection was selected for NELM whenever possible; however, an anatomic resection including a segmentectomy, sectionectomy, or hemihepatectomy was selected for large tumors or tumors invading major hepatic vessels.20 Patients undergoing curative hepatic resection were followed up every 3 months using serologic examinations, including the evaluation of transaminases and bilirubin levels and imaging studies, such as abdominal ultrasonography, contrast-enhanced computed tomography, and magnetic resonance imaging. Recurrent NELM was diagnosed when new hypervascular lesions were identified in the imaging studies or when small lesions with any vascularity exhibited gradual growth in size. Most patients who did not undergo hepatic resection or who underwent an R2 resection were subjected to medical treatments, including cytotoxic agents, molecular target drugs, and somatostatin analogs. The indications for a repeat hepatectomy for recurrent NELM were decided using the same criteria as those used for the first hepatectomy. Patients with early recurrence, usually within 3 months, were allocated to a waitand-see policy, and operative indications were reconsidered after several months of chemotherapy or observation. Transcatheter arterial chemoembolization, radiofrequency ablation, or best-supportive care was selected in selected patients in whom neither curative surgery nor systemic drug therapy was effective or amenable. In patients with recurrent NELM, the clinicopathologic characteristics and long-term results were compared between patients with repeat hepatectomy and those without. A multivariable analysis was performed to explore independent prognostic factors. Clinicopathologic characteristics, short-term outcomes, and progression-free survival (PFS) after the first hepatectomy were compared with those after repeat hepatectomy.
Fig 1. Flow chart of patients.
Statistical analysis Categoric data were expressed as the number and proportion, and continuous data were expressed as the median value with the range. A comparison of the 2 groups was performed using a c2 test or the Fisher exact test for categoric data and using the Mann-Whitney U test for continuous data. OS and PFS were calculated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. Univariable and multivariable analyses for OS were performed using the Cox proportional hazard model. Variables with a P value of < .05 in the univariable analyses were entered into the multivariable analysis. The statistical analyses were performed using SPSS Statistics, version 25.0 (SPSS Inc., Chicago, IL). Results Patient selection A flow chart showing the treatment selection for the patients with NELM is shown in Fig 1. Among the 44 patients who were reviewed, a curative hepatectomy defined as an R0 or R1 resection was performed in 33 patients; 28 patients developed recurrent NELM. A repeat hepatectomy was performed in 16 of these 28 patients. Comparison of repeat hepatectomy and no repeat hepatectomy The clinicopathologic characteristics of the 28 patients in the recurrent NELM cohort and a comparison between the repeat hepatectomy (n ¼ 16) and no hepatectomy (n ¼ 12) cohorts are summarized in Table I. Significant differences between the 2 groups were found in terms of the presence of microscopic lymphatic invasion of the primary tumor, the recurrent interval from the first hepatectomy, and the incidence of extrahepatic metastasis (P .029 each). Of the 12 patients with no repeat hepatectomy, streptozocin was administered in 6 patients, everolimus in 3, radiofrequency ablation in 2, and the other patient received best supportive care. The reasons for no repeat hepatectomy were as follows: inadequate liver remnant volume or technically unresectable disease in 6 patients, extrahepatic metastases in 3 patients, and quite early recurrence in the remaining 3 patients. The median OS after the diagnosis of recurrent NELM was 92 months (95% confidence interval [CI], 47e172), and the 1-, 5-, and 10-year OS rates were 96.2%, 61.5%, and 37.2%, respectively (Supplementary Fig 1). Multivariable analyses for OS after the time of recurrence identified multiple nodules (hazard ratio [HR] 26.2; 95% CI, 2.08e329.52; P ¼ .008), a carbohydrate antigen (CA19-9) level 40
S. Kiritani et al. / Surgery xxx (2019) 1e6
3
Table I Clinicopathologic characteristics about the patients of recurrent NELM (n ¼ 28)
Age (y)y Male Primary tumor factors Functionality Location, pancreas/duodenum/others CEA (ng/mL)y CA19-9 (U/mL)y Tumor size (mm)y Grade,x G1/G2/G3 Ki-67 labeling indexy Microscopic vascular invasion Microscopic lymphatic invasion Lymph node metastasis Curability, R0/R1/R2 Initial NELM factors Onset, synchronous CEA (ng/mL)y CA19-9 (U/mL)y No. of nodulesy Maximum tumor size (mm)y Extra hepatic metastasis Recurrent NELM factors CEA (ng/mL)y CA19-9 (U/mL)y Interval between the first hepatectomy and recurrence (mo)y No. of nodulesy Maximum tumor size (mm)y Extra hepatic metastasis Repeat hepatectomy
Whole (n ¼ 28)
Repeat hepatectomy (n ¼ 16)
No hepatectomy (n ¼ 12)
P value*
57 (36e85) 16 (57.1)
53 (37e75) 9 (56.3)
59 (36e85) 7 (58.3)
.391z .908
3 (10.7) 21/4/3 2.2 (0.5e16.2) 17 (1e96) 40 (3.0e150.0) 4/15/1 6.0 (1.0e25.0) 21 (75.0) 15 (53.6) 14 (50.0) 18/10/0
3 (18.8) 13/2/1 1.9 (0.5e5.7) 16 (1e88) 50 (3.0e150.0) 2/7/1 6.0 (1.0e25.0) 13 (81.3) 11 (68.8) 6 (37.5) 9/7/0
0 (0.0) 8/2/2 3.7 (0.9e16.2) 19 (1e96) 30 (11.0e110.0) 2/8/0 6.0 (1.0e20.0) 8 (66.7) 4 (48.0) 8 (66.7) 9/3/0
.112 .647 .132z .247z .574z .589 .728z .373 .027 .156 .332
18 (64.3) 2.3 (0.9e22.0) 18 (1e575) 4 (1e67) 35.0 (9.0e130.0) 0 (0.0)
10 (62.5) 2.5 (1.0e22.0) 18 (1e575) 4 (1e26) 34.0 (10.0e130.0) 0 (0.0)
8 (66.7) 2.3 (0.9e16.2) 21 (1e96) 7 (1e67) 50.0 (9.0e62.0) 0 (0.0)
.820 .727z .291z .178z .577z e
3.5 (0.7e17.4) 18 (1e130) 16 (3e97)
3.5 (0.7e8.4) 16 (1e65) 19 (9e100)
3.1 (0.7e17.4) 22 (1e130) 8 (3e35)
1.000z .431z .007z
5 (1e60) 14.0 (5.0e110.0) 3 (10.7) 16 (57.1)
3.5 (1e24) 15.0 (6.0e110.0) 0 (0.0) d
5 (3e60) 12.0 (5.0e16.0) 3 (25.0) d
.372z .214z .029 d
Values in parentheses are percentages unless indicated otherwise. CEA, carcinoembryonic antigen. * c2 test was performed except Mann-Whitney U test. z Mann-Whitney U test. y Values are median (range). x The grade is indicated according to World Health Organization 2010 classification.
Table II Multivariable analysis of predictors of overall survival in recurrent NELM (n ¼ 28) Univariable analysis
Age 60 y Male Primary tumor factors Functionality no Location pancreas Maximum tumor size 30.0 mm CEA 5.0 ng/mL CA19-9 40 U/mL Ki-67 labeling index 15.0% Microscopic vascular invasion Microscopic lymphatic invasion Lymph node metastasis Curability R1/R2 Initial NELM factors Onset, synchronous CEA 5.0 ng/mL CA19-9 40 U/mL Recurrent NELM factors CEA 5.0 ng/mL CA19-9 40 U/mL Multiple nodules Maximum tumor size 15.0 mm Time to recurrence after first hepatectomy 12 mo Repeat hepatectomy no CEA, carcinoembryonic antigen. * Cox proportional hazards model.
Multivariable analysis
HR (95% CI)
P value*
2.27 (0.78e7.39) 1.51 (0.53e4.39)
.132 .441
2.68 1.04 0.69 2.23 3.11 3.76 0.58 0.34 1.59 0.48
.287 .945 .489 .202 .086 .164 .391 .074 .426 .256
(0.50e50.12) (0.32e4.64) (0.23e2.04) (0.65e7.60) (0.85e11.31) (0.53e17.62) (0.19e2.17) (0.09e1.07) (0.50e5.50) (0.16e1.65)
HR (95% CI)
P value*
1.67 (0.54e5.42) 2.42 (0.81e7.23) 3.00 (0.97e9.26)
.374 .113 .056
6.47 (1.46e44.90) 8.14 (1.96e31.66) 10.04 (1.24e81.50) 0.40 (0.11e1.34) 2.64 (0.80e8.79)
.008 <.001 .032 .138 .113
2.33 (0.70e7.72) 11.48 (1.97e67.17) 26.17 (2.08e329.52)
.174 .012 .008
4.61 (1.41e16.20)
.009
5.49 (1.13e26.62)
.036
4
S. Kiritani et al. / Surgery xxx (2019) 1e6
Fig 2. Kaplan-Meier estimates of overall survival between repeat hepatectomy and no repeat hepatectomy cohorts with recurrent NELM. The day of diagnosis was regarded as the first day of the survival period. P < .001 (log-rank test).
Fig 3. Kaplan-Meier estimates of PFS after presumed curative surgery for the initial NELM and recurrent NELM. The day of each operation was regarded as the starting point for these curves. P ¼ .546 (log-rank test).
U/mL (HR 11.19; 95% CI, 2.23e59.60; P ¼ .012), and no resection of recurrent NELM (HR 4.95; 95% CI, 1.35e19.58; P ¼ .036) as independent predictors of a poor outcome (Table II). The OS curves for the 2 groups are shown in Fig 2. The median OS after the diagnosis of recurrent NELM and a repeat hepatectomy was 126 months (95% CI, 57e172), whereas the OS in patients without a repeat hepatectomy was 41 months (95%, CI 23e92); this difference was statistically significant (P < .001).
with NELM,21 although some controversy persists mainly because of the high frequency of intrahepatic and extrahepatic recurrence after hepatic resection.11e13,22e24 Furthermore, although recurrence after initial hepatic resection for NELM is common (84% in the present series), the optimal treatment for such recurrent tumors has continued to be debated. In the present study, the role of the reresection of recurrent NELM tumors was examined. Two important results of this study support the usefulness of reresection of recurrent NELMs. First, treatment other than a hepatectomy was shown to be an independent prognostic factor among patients with recurrent NELM tumors. Second, the PFS curves after a first hepatectomy for initial NELM and those after a repeat hepatectomy for recurrent NELM were similar. Thus, we demonstrated that repeat hepatectomy would be applicable for patients with recurrent NELM. However, it was not appropriate to compare OS after first and repeat hepatectomy cohorts because some patients overlapped the other cohort. To our knowledge, only one report by Spolverato et al has focused on the clinical course after various treatments for recurrent NELM.25 They reported that 35.8% of the patients with recurrent NELM underwent a repeat hepatectomy and that 19.7% of the patients underwent intra-arterial therapy; these 2 treatment methods resulted in better patient survival than the use of somatostatin analogs and systemic cytotoxic chemotherapy. In the present study, 57% of the patients with recurrent NELM who
Comparison of first and repeat hepatectomy A comparison of the characteristics between the first (n ¼ 44) and repeat (n ¼ 16) hepatectomy cohorts is shown in Table III. The maximum tumor size and the rate of preoperative treatment were different between the 2 groups. No significant differences in shortterm outcomes were detected between the first and repeat hepatectomy cohorts (Supplementary Table I). The PFS curves are shown in Fig 3. The respective median PFS times after a first and a repeat hepatectomy were 13 and 6 months (95% CI, 8e18 and 2e10, respectively) and were not different (P ¼ .546). Discussion Curative liver resection and debulking procedures are recommended as the treatments of choice for selective patients
Table III Comparison of characteristics between first and repeat hepatectomy
Liver metastasis factors CEA (ng/mL)y CA19-9 (ng/mL)y Synchronous onset Interval from primary resection (mo)y Interval from first hepatectomy (mo)y Number of nodulesy Maximum tumor size (mm)y Preoperative treatment Extra hepatic metastasis Operative procedure limited/anatomic Curability R0/R1/R2
First hepatectomy (n ¼ 44)
Repeat hepatectomy (n ¼ 16)
P value*
2.7 (0.9e22.0) 15.0 (1.0e575.0) 31 (70.5) 37 (4e151)
3.5 (0.7e8.4) 16.0 (1.0e65.0) 10 (62.5)
.637z .720z .564 d d .145z .008z .014 .543 .083 .862
7 (1e139) 28.0 (9.0e130.0) 14 (23.3) 1 (2.3) 25 (56.8) 19/14/11
Values in parentheses are percentages unless indicated otherwise. * c2 test was performed except Mann-Whitney U test. z Mann-Whitney U test. y Values are median (range).
16 (3e97) 4 (1e24) 15.0 (6.0e110.0) 11 (68.8) 0 (0.0) 13 (81.3) 7/6/3
S. Kiritani et al. / Surgery xxx (2019) 1e6
underwent a repeat hepatectomy, which was more than the percentage in the previous report, had acceptable oncologic effects. Spolverato et al also reported that the 5-year OS of a group of patients with recurrent hepatic NELM who had no operative reintervention was 45.9%, which was less than the result for repeat hepatectomy in our study (69%). Nonoperative treatment, including streptozocin (7 patients), everolimus (2 patients), radiofrequency ablation (2 patients), and best supportive care (1 patient), were administered to the remaining 12 patients. The median PFS of streptozocin and other treatments were each 3.5 and 3.0 months (95% CI, 3e5 and 1e8), which were less than that of repeat hepatectomy (6.0 mo; 95% CI, 2e10). The PFS curves after the first hepatectomy and those after the repeat hepatectomy were similar. Similar findings have been reported for patients with colorectal liver metastasis26e28; however, no previous reports, to our knowledge, have reported such findings in patients with NELM. The characteristics of the first hepatectomy cohort and those of the repeat hepatectomy cohort were similar except for the maximum tumor size. The tumor size was smaller in the repeat hepatectomy cohort than in the first hepatectomy cohort, primarily because of the close follow-up after the first hepatectomy. Although the smaller tumor size might have had a favorable effect on the long-term outcomes of the repeat hepatectomy cohort, this factor was excluded from the independent prognostic factors after both the first and repeat hepatectomy, indicating a limited prognostic impact. Additionally, the short-term outcomes of a repeat hepatectomy, despite this procedure being more technically demanding, were comparable to those of a first hepatectomy. The aforementioned results supported the usefulness of repeated hepatectomy for recurrent NELM. Although certain oncologic benefits of hepatectomy, including both a first and a repeat hepatectomy, were recognized from our results, confirming that patients who undergo a hepatectomy for NELM can obtain a true cure is difficult for the following reasons. In the present study, the OS curve after the initial hepatectomy decreased even after 10 years, and only 1 patient had reached a 10year PRS after a first hepatectomy; none of the patients undergoing a repeat hepatectomy achieved such a milestone (Fig 3). A greater observation period is thus needed to evaluate the true curability of repeat hepatectomy. In the multivariable analysis of recurrent NELM patients in the present study, a high CA19-9 value, multiple nodules, and treatment other than a hepatectomy were shown to be independently correlated with a poor outcome. For the initial NELM, CA19-9 was usually not associated with differentiated NET, although several factors, such as the amount of tumor, operative curability, presence of extrahepatic disease, functionality, or poor differentiation, have been reported as prognostic factors.9,24,29,30 A retrospective cohort study by Luo et al partially supported our results that a high serum CA19-9 level was correlated with lymph node metastasis and a poor prognosis in patients with pancreatic NET. The present study, however, is slightly different from the report by Luo et al31; Luo et al suggested a relationship between the CA19-9 level and the biology of primary pancreatic NET, the biology of recurrent NELM was associated with CA19-9 level at the resection of recurrent NELM, not at the resection of the primary site or at the first hepatectomy in our study. Because any relation between an increase in CA19-9 and poor outcome for recurrent NELM remains unknown, further research would be warranted. The present study has several limitations. First, retrospective nature of this study may have led to a favorable patient selection especially for recurrent NELM cohorts. A possible selection bias may influence the result of comparison between OS after the
5
first hepatectomy and that after the repeat hepatectomy, because the patients undergoing the second hepatectomy may have more favorable characteristics considering the technical and anatomic difficulty. The type of hepatic resection would often be limited, because the hepatic parenchymal volume would be decreased from the first hepatectomy. Second, there were several missing values in this study. Levels of neuronspecific enolase, which was reportedly a potential poor prognostic marker of NETs, was not measured in all patients in the present study because the measurement of it was unavailable and not covered by health insurance system in Japan. The utility of this marker should be clarified in further studies. Although a greater Ki-67 labeling index in NELM compared to primary tumor has been reported as a poor prognostic factor,32 we did not measure the Ki-67 labeling index in the initial and recurrent NELM in most patients because the period of this study was long and its importance was only recognized recently. Finally, the relatively small numbers of patients undergoing a first and a repeat hepatectomy might not reveal all the possible findings.25 In conclusion, although NELM frequently reoccurs after curative surgery, a repeat hepatectomy for recurrent NELM may be beneficial in selected patients. Conflict of interest/Disclosure The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. Funding/Support This work was not supported by any grants. Supplementary materials Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.1016/j.surg.2019. 08.020. References 1. Yao JC, Hassan M, Phan A, 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;26:3063e3072. 2. Lawrence B, Gustafsson BI, Chan A, et al. The epidemiology of gastroenteropancreatic neuroendocrine tumors. Endocrinol Metab Clin North Am. 2011;40:1e18, vii. 3. Dasari A, Shen C, Halperin D, et al. Trends in the incidence, prevalence, and survival outcomes in patients with neuroendocrine tumors in the United States. JAMA Oncol. 2017;3:1335e1342. 4. Pape UF, Jann H, Muller-Nordhorn J, et al. Prognostic relevance of a novel TNM classification system for upper gastroenteropancreatic neuroendocrine tumors. Cancer. 2008;113:256e265. 5. Majeed F, Porter TR, Tarantolo S, et al. Carcinoid crisis and reversible right ventricular dysfunction after embolization in untreated carcinoid syndrome. Eur J Echocardiogr. 2007;8:386e389. 6. Fujie S, Zhou W, Fann P, et al. Carcinoid crisis 24 hours after bland embolization: A case report. Biosci Trends. 2010;4:143e144. 7. Cives M, Strosberg J. Treatment strategies for metastatic neuroendocrine tumors of the gastrointestinal tract. Curr Treat Options Oncol. 2017;18:14. 8. Kunz PL. Carcinoid and neuroendocrine tumors: Building on success. J Clin Oncol. 2015;33:1855e1863. 9. Frilling A, Modlin IM, Kidd M, et al. Recommendations for management of patients with neuroendocrine liver metastases. Lancet Oncol. 2014;15:e8ee21. 10. Zhang XF, Beal EW, Chakedis J, et al. Early recurrence of neuroendocrine liver metastasis after curative hepatectomy: Risk factors, prognosis, and treatment. J Gastrointest Surg. 2017;21:1821e1830. 11. Gurusamy KS, Ramamoorthy R, Sharma D, et al. Liver resection versus other treatments for neuroendocrine tumours in patients with resectable liver metastases. Cochrane Database Syst Rev. 2009:Cd007060. 12. Pavel M, Baudin E, Couvelard A, et al. ENETS Consensus Guidelines for the management of patients with liver and other distant metastases from
6
13. 14.
15.
16. 17. 18.
19.
20.
21.
S. Kiritani et al. / Surgery xxx (2019) 1e6 neuroendocrine neoplasms of foregut, midgut, hindgut, and unknown primary. Neuroendocrinology. 2012;95:157e176. Farley HA, Pommier RF. Treatment of neuroendocrine liver metastases. Surg Oncol Clin N Am. 2016;25:217e225. Moertel CG, Lefkopoulo M, Lipsitz S, et al. Streptozocin-doxorubicin, streptozocin-fluorouracil or chlorozotocin in the treatment of advanced isletcell carcinoma. N Engl J Med. 1992;326:519e523. Rinke A, Muller HH, Schade-Brittinger C, et al. Placebo-controlled, doubleblind, prospective, randomized study on the effect of octreotide LAR in the control of tumor growth in patients with metastatic neuroendocrine midgut tumors: A report from the PROMID Study Group. J Clin Oncol. 2009;27: 4656e4663. Raymond E, Dahan L, Raoul JL, et al. Sunitinib malate for the treatment of pancreatic neuroendocrine tumors. N Engl J Med. 2011;364:501e513. Caplin ME, Pavel M, Cwikla JB, et al. Lanreotide in metastatic enteropancreatic neuroendocrine tumors. N Engl J Med. 2014;371:224e233. Yao JC, Fazio N, Singh S, et al. Everolimus for the treatment of advanced, nonfunctional neuroendocrine tumours of the lung or gastrointestinal tract (RADIANT-4): A randomised, placebo-controlled, phase 3 study. Lancet. 2016;387:968e977. Kubota K, Makuuchi M, Kusaka K, et al. Measurement of liver volume and hepatic functional reserve as a guide to decision-making in resectional surgery for hepatic tumors. Hepatol. 1997;26:1176e1181. Strasberg SM. Nomenclature of hepatic anatomy and resections: A review of the Brisbane 2000 system. J Hepatobiliary Pancreat Surg. 2005;12: 351e355. Sarmiento JM, Heywood G, Rubin J, et al. Surgical treatment of neuroendocrine metastases to the liver: A plea for resection to increase survival. J Am Coll Surg. 2003;197:29e37.
22. Fraenkel M, Kim M, Faggiano A, et al. Incidence of gastroenteropancreatic neuroendocrine tumours: A systematic review of the literature. Endocr Relat Cancer. 2014;21:R153eR163. 23. Kunz PL, Reidy-Lagunes D, Anthony LB, et al. Consensus guidelines for the management and treatment of neuroendocrine tumors. Pancreas. 2013;42: 557e577. 24. Maxwell JE, Sherman SK, O'Dorisio TM, et al. Liver-directed surgery of neuroendocrine metastases: What is the optimal strategy? Surgery. 2016;159:320e333. 25. Spolverato G, Bagante F, Aldrighetti L, et al. Management and outcomes of patients with recurrent neuroendocrine liver metastasis after curative surgery: An international multi-institutional analysis. J Surg Oncol. 2017;116:298e306. 26. Oba M, Hasegawa K, Shindoh J, et al. Survival benefit of repeat resection of successive recurrences after the initial hepatic resection for colorectal liver metastases. Surgery. 2016;159:632e640. 27. Bozzetti F, Bignami P, Montalto F, et al. Repeated hepatic resection for recurrent metastases from colorectal cancer. Br J Surg. 1992;79:146e148. 28. Adair RA, Young AL, Cockbain AJ, et al. Repeat hepatic resection for colorectal liver metastases. Br J Surg. 2012;99:1278e1283. 29. Mayo SC, de Jong MC, Pawlik TM. Surgical management and emerging therapies to prolong survival in metastatic neuroendocrine cancer. Ann Surg Oncol. 2011;18:S220eS221; author reply S222e233. 30. Bagante F, Spolverato G, Merath K, et al. Neuroendocrine liver metastasis: The chance to be cured after liver surgery. J Surg Oncol. 2017;115:687e695. 31. Luo G, Jin K, Cheng H, et al. Carbohydrate antigen 19-9 as a prognostic biomarker in pancreatic neuroendocrine tumors. Oncol Lett. 2017;14:6795e6800. 32. Lv Y, Han X, Huang C, et al. Risk factors affecting prognosis in metachronous liver metastases from WHO classification G1 and G2 gastroenteropancreatic neuroendocrine tumors after initial R0 surgical resection. BMC Cancer. 2019;19:335.
Contents lists available at ScienceDirect
Surgery journal homepage: www.elsevier.com/locate/surg
Repeat hepatectomy for patients with recurrent neuroendocrine liver metastasis: Comparison with first hepatectomy Sho Kiritani, MD, Junichi Arita, PhD, Masaru Matsumura, MD, Yujiro Nishioka, PhD, Hiroki Kudo, MD, Akihiko Ichida, PhD, Takeaki Ishizawa, PhD, Nobuhisa Akamatsu, PhD, Junichi Kaneko, PhD, Kiyoshi Hasegawa, PhD* Hepato-Biliary-Pancreatic Surgery Division, Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 20 August 2019 Available online xxx
Background: The value of repeat hepatectomy for intrahepatic recurrence after curative resection of neuroendocrine liver metastasis (NELM) remains unclear. The aim of this retrospective cohort study was to determine the significance of repeat hepatectomy for recurrent NELM. Methods: Patients who underwent hepatectomy for NELM between 1994 and 2016 were identified. The indications for a first hepatectomy were adequate liver remnant volume and no extrahepatic metastasis. The diagnosis of recurrent NELM was based on radiographic examinations. The indications for a repeat hepatectomy were the same as those for the first hepatectomy. Clinicopathologic factors, short-term survival, and long-term survival were investigated using clinical records. Results: Forty-four patients enrolled in this study. Thirty-three patients among them underwent a curative hepatectomy, and 28 of them developed recurrence. Of them, 16 patients underwent a repeat hepatectomy. The overall survival of the repeat hepatectomy cohort (n ¼ 16) was significantly better than that of the no repeat hepatectomy cohort (n ¼ 12) (P < .001). The progression free survival after the first hepatectomy (n ¼ 44) and that after repeat hepatectomy (n ¼ 16) were similar (P ¼ .546). No significant difference was seen between the frequency of major complications (Clavien-Dindo score 3a) after the first and repeat hepatectomy (P ¼ .279). No repeat hepatectomy (hazard ratio [HR] 5.0, P ¼ .036) was identified as an independent predictive factor of a poor outcome among the recurrent cohort, along with the presence of multiple nodules (HR 26.2, P ¼ .008) and a CA19-9 level 40 U/mL (HR 11.2, P ¼ .012). Conclusions: A repeat hepatectomy is feasible in selected patients with recurrent NELM. © 2019 Elsevier Inc. All rights reserved.
Introduction Although neuroendocrine tumors (NETs) are rare with an incidence of about 5 per 100,000 according to a large exploratory report of The Surveillance, Epidemiology, and End Results data by Yao et al,1 the recognized incidence of this disease increased by about 6.4-fold from 1973 to 2012.1e3 Approximately 60% to 90% of patients with NETs have synchronous or metachronous neuroendocrine liver metastasis (NELM), and many of these patients have constitutional symptoms and a poor prognosis.1,4e7 Populationbased studies have reported that the median overall survival (OS) for patients with metastatic pancreatic and small bowel NETs are 24 * Reprint requests: Kiyoshi Hasegawa, PhD, Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail address: [email protected] (K. Hasegawa). https://doi.org/10.1016/j.surg.2019.08.020 0039-6060/© 2019 Elsevier Inc. All rights reserved.
and 56 months, respectively.1,8 Operative resection is recommended if curative resection is possible; however, surgery with curative intent was reportedly adopted for only half or less of all cases at the time of the diagnosis of NELM.4,9,10 The Cochrane systematic review suggested that the better prognosis after hepatic resection for NELM was attributable to a selection bias.11 Thus, the operative indications for NELM have been controversial. Recurrence after resection is common and decreases both the quality of life and the long-term survival of patients. Intrahepatic recurrence after a curative hepatectomy for NELM can be as high as 70% to 94% at 5 years11e13; however, the optimal treatment strategy for patients with recurrent tumors has remained unclear. Several medications, including somatostatin analogues, everolimus, sunitinib, and streptozocin, have been used for the treatment of NETs, which paradoxically complicate the selection of the optimal treatment option.14e18 Because few reports assessing recurrent NELM have been published, the purpose of
2
S. Kiritani et al. / Surgery xxx (2019) 1e6
the present study was to determine the oncologic impact of repeat hepatectomy for recurrent NELM after a first hepatectomy.
Patients and Methods This retrospective cohort study was approved by the Institutional Ethics Committee of Tokyo University. An opt-out manuscript for this study was published on our website (http://www.u-tokyohbp-transplant-surgery.jp/). None of the patients were opposed to being included in the present study. All operative candidates with NETs who were treated between January 1994 and December 2016 at our department were identified from the maintained database. Among them, all the patients undergoing resection for NELM were included in this study. Contrast-enhanced computed tomography or magnetic resonance imaging of the chest, abdomen, and pelvis using a triphasic liver protocol was performed for all the patients to assess resectability and extrahepatic disease. Neoadjuvant treatment was not considered, in principle. The indications for hepatectomy were as follows: (1) an acceptable general condition, (2) a favorable Child-Pugh Classification (A plus selected class B) and indocyanine green retention rate at 15 min, (3) adequate liver remnant volume,19 and (4) no extrahepatic lymph node or organ metastasis. In addition, hepatic resection was indicated for patients with hormonal symptoms even if an R2 resection was expected. A R1 resection was allowed when the resection of all the tumors was otherwise impossible. In principle, parenchymasparing, nonanatomic hepatic resection was selected for NELM whenever possible; however, an anatomic resection including a segmentectomy, sectionectomy, or hemihepatectomy was selected for large tumors or tumors invading major hepatic vessels.20 Patients undergoing curative hepatic resection were followed up every 3 months using serologic examinations, including the evaluation of transaminases and bilirubin levels and imaging studies, such as abdominal ultrasonography, contrast-enhanced computed tomography, and magnetic resonance imaging. Recurrent NELM was diagnosed when new hypervascular lesions were identified in the imaging studies or when small lesions with any vascularity exhibited gradual growth in size. Most patients who did not undergo hepatic resection or who underwent an R2 resection were subjected to medical treatments, including cytotoxic agents, molecular target drugs, and somatostatin analogs. The indications for a repeat hepatectomy for recurrent NELM were decided using the same criteria as those used for the first hepatectomy. Patients with early recurrence, usually within 3 months, were allocated to a waitand-see policy, and operative indications were reconsidered after several months of chemotherapy or observation. Transcatheter arterial chemoembolization, radiofrequency ablation, or best-supportive care was selected in selected patients in whom neither curative surgery nor systemic drug therapy was effective or amenable. In patients with recurrent NELM, the clinicopathologic characteristics and long-term results were compared between patients with repeat hepatectomy and those without. A multivariable analysis was performed to explore independent prognostic factors. Clinicopathologic characteristics, short-term outcomes, and progression-free survival (PFS) after the first hepatectomy were compared with those after repeat hepatectomy.
Fig 1. Flow chart of patients.
Statistical analysis Categoric data were expressed as the number and proportion, and continuous data were expressed as the median value with the range. A comparison of the 2 groups was performed using a c2 test or the Fisher exact test for categoric data and using the Mann-Whitney U test for continuous data. OS and PFS were calculated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. Univariable and multivariable analyses for OS were performed using the Cox proportional hazard model. Variables with a P value of < .05 in the univariable analyses were entered into the multivariable analysis. The statistical analyses were performed using SPSS Statistics, version 25.0 (SPSS Inc., Chicago, IL). Results Patient selection A flow chart showing the treatment selection for the patients with NELM is shown in Fig 1. Among the 44 patients who were reviewed, a curative hepatectomy defined as an R0 or R1 resection was performed in 33 patients; 28 patients developed recurrent NELM. A repeat hepatectomy was performed in 16 of these 28 patients. Comparison of repeat hepatectomy and no repeat hepatectomy The clinicopathologic characteristics of the 28 patients in the recurrent NELM cohort and a comparison between the repeat hepatectomy (n ¼ 16) and no hepatectomy (n ¼ 12) cohorts are summarized in Table I. Significant differences between the 2 groups were found in terms of the presence of microscopic lymphatic invasion of the primary tumor, the recurrent interval from the first hepatectomy, and the incidence of extrahepatic metastasis (P .029 each). Of the 12 patients with no repeat hepatectomy, streptozocin was administered in 6 patients, everolimus in 3, radiofrequency ablation in 2, and the other patient received best supportive care. The reasons for no repeat hepatectomy were as follows: inadequate liver remnant volume or technically unresectable disease in 6 patients, extrahepatic metastases in 3 patients, and quite early recurrence in the remaining 3 patients. The median OS after the diagnosis of recurrent NELM was 92 months (95% confidence interval [CI], 47e172), and the 1-, 5-, and 10-year OS rates were 96.2%, 61.5%, and 37.2%, respectively (Supplementary Fig 1). Multivariable analyses for OS after the time of recurrence identified multiple nodules (hazard ratio [HR] 26.2; 95% CI, 2.08e329.52; P ¼ .008), a carbohydrate antigen (CA19-9) level 40
S. Kiritani et al. / Surgery xxx (2019) 1e6
3
Table I Clinicopathologic characteristics about the patients of recurrent NELM (n ¼ 28)
Age (y)y Male Primary tumor factors Functionality Location, pancreas/duodenum/others CEA (ng/mL)y CA19-9 (U/mL)y Tumor size (mm)y Grade,x G1/G2/G3 Ki-67 labeling indexy Microscopic vascular invasion Microscopic lymphatic invasion Lymph node metastasis Curability, R0/R1/R2 Initial NELM factors Onset, synchronous CEA (ng/mL)y CA19-9 (U/mL)y No. of nodulesy Maximum tumor size (mm)y Extra hepatic metastasis Recurrent NELM factors CEA (ng/mL)y CA19-9 (U/mL)y Interval between the first hepatectomy and recurrence (mo)y No. of nodulesy Maximum tumor size (mm)y Extra hepatic metastasis Repeat hepatectomy
Whole (n ¼ 28)
Repeat hepatectomy (n ¼ 16)
No hepatectomy (n ¼ 12)
P value*
57 (36e85) 16 (57.1)
53 (37e75) 9 (56.3)
59 (36e85) 7 (58.3)
.391z .908
3 (10.7) 21/4/3 2.2 (0.5e16.2) 17 (1e96) 40 (3.0e150.0) 4/15/1 6.0 (1.0e25.0) 21 (75.0) 15 (53.6) 14 (50.0) 18/10/0
3 (18.8) 13/2/1 1.9 (0.5e5.7) 16 (1e88) 50 (3.0e150.0) 2/7/1 6.0 (1.0e25.0) 13 (81.3) 11 (68.8) 6 (37.5) 9/7/0
0 (0.0) 8/2/2 3.7 (0.9e16.2) 19 (1e96) 30 (11.0e110.0) 2/8/0 6.0 (1.0e20.0) 8 (66.7) 4 (48.0) 8 (66.7) 9/3/0
.112 .647 .132z .247z .574z .589 .728z .373 .027 .156 .332
18 (64.3) 2.3 (0.9e22.0) 18 (1e575) 4 (1e67) 35.0 (9.0e130.0) 0 (0.0)
10 (62.5) 2.5 (1.0e22.0) 18 (1e575) 4 (1e26) 34.0 (10.0e130.0) 0 (0.0)
8 (66.7) 2.3 (0.9e16.2) 21 (1e96) 7 (1e67) 50.0 (9.0e62.0) 0 (0.0)
.820 .727z .291z .178z .577z e
3.5 (0.7e17.4) 18 (1e130) 16 (3e97)
3.5 (0.7e8.4) 16 (1e65) 19 (9e100)
3.1 (0.7e17.4) 22 (1e130) 8 (3e35)
1.000z .431z .007z
5 (1e60) 14.0 (5.0e110.0) 3 (10.7) 16 (57.1)
3.5 (1e24) 15.0 (6.0e110.0) 0 (0.0) d
5 (3e60) 12.0 (5.0e16.0) 3 (25.0) d
.372z .214z .029 d
Values in parentheses are percentages unless indicated otherwise. CEA, carcinoembryonic antigen. * c2 test was performed except Mann-Whitney U test. z Mann-Whitney U test. y Values are median (range). x The grade is indicated according to World Health Organization 2010 classification.
Table II Multivariable analysis of predictors of overall survival in recurrent NELM (n ¼ 28) Univariable analysis
Age 60 y Male Primary tumor factors Functionality no Location pancreas Maximum tumor size 30.0 mm CEA 5.0 ng/mL CA19-9 40 U/mL Ki-67 labeling index 15.0% Microscopic vascular invasion Microscopic lymphatic invasion Lymph node metastasis Curability R1/R2 Initial NELM factors Onset, synchronous CEA 5.0 ng/mL CA19-9 40 U/mL Recurrent NELM factors CEA 5.0 ng/mL CA19-9 40 U/mL Multiple nodules Maximum tumor size 15.0 mm Time to recurrence after first hepatectomy 12 mo Repeat hepatectomy no CEA, carcinoembryonic antigen. * Cox proportional hazards model.
Multivariable analysis
HR (95% CI)
P value*
2.27 (0.78e7.39) 1.51 (0.53e4.39)
.132 .441
2.68 1.04 0.69 2.23 3.11 3.76 0.58 0.34 1.59 0.48
.287 .945 .489 .202 .086 .164 .391 .074 .426 .256
(0.50e50.12) (0.32e4.64) (0.23e2.04) (0.65e7.60) (0.85e11.31) (0.53e17.62) (0.19e2.17) (0.09e1.07) (0.50e5.50) (0.16e1.65)
HR (95% CI)
P value*
1.67 (0.54e5.42) 2.42 (0.81e7.23) 3.00 (0.97e9.26)
.374 .113 .056
6.47 (1.46e44.90) 8.14 (1.96e31.66) 10.04 (1.24e81.50) 0.40 (0.11e1.34) 2.64 (0.80e8.79)
.008 <.001 .032 .138 .113
2.33 (0.70e7.72) 11.48 (1.97e67.17) 26.17 (2.08e329.52)
.174 .012 .008
4.61 (1.41e16.20)
.009
5.49 (1.13e26.62)
.036
4
S. Kiritani et al. / Surgery xxx (2019) 1e6
Fig 2. Kaplan-Meier estimates of overall survival between repeat hepatectomy and no repeat hepatectomy cohorts with recurrent NELM. The day of diagnosis was regarded as the first day of the survival period. P < .001 (log-rank test).
Fig 3. Kaplan-Meier estimates of PFS after presumed curative surgery for the initial NELM and recurrent NELM. The day of each operation was regarded as the starting point for these curves. P ¼ .546 (log-rank test).
U/mL (HR 11.19; 95% CI, 2.23e59.60; P ¼ .012), and no resection of recurrent NELM (HR 4.95; 95% CI, 1.35e19.58; P ¼ .036) as independent predictors of a poor outcome (Table II). The OS curves for the 2 groups are shown in Fig 2. The median OS after the diagnosis of recurrent NELM and a repeat hepatectomy was 126 months (95% CI, 57e172), whereas the OS in patients without a repeat hepatectomy was 41 months (95%, CI 23e92); this difference was statistically significant (P < .001).
with NELM,21 although some controversy persists mainly because of the high frequency of intrahepatic and extrahepatic recurrence after hepatic resection.11e13,22e24 Furthermore, although recurrence after initial hepatic resection for NELM is common (84% in the present series), the optimal treatment for such recurrent tumors has continued to be debated. In the present study, the role of the reresection of recurrent NELM tumors was examined. Two important results of this study support the usefulness of reresection of recurrent NELMs. First, treatment other than a hepatectomy was shown to be an independent prognostic factor among patients with recurrent NELM tumors. Second, the PFS curves after a first hepatectomy for initial NELM and those after a repeat hepatectomy for recurrent NELM were similar. Thus, we demonstrated that repeat hepatectomy would be applicable for patients with recurrent NELM. However, it was not appropriate to compare OS after first and repeat hepatectomy cohorts because some patients overlapped the other cohort. To our knowledge, only one report by Spolverato et al has focused on the clinical course after various treatments for recurrent NELM.25 They reported that 35.8% of the patients with recurrent NELM underwent a repeat hepatectomy and that 19.7% of the patients underwent intra-arterial therapy; these 2 treatment methods resulted in better patient survival than the use of somatostatin analogs and systemic cytotoxic chemotherapy. In the present study, 57% of the patients with recurrent NELM who
Comparison of first and repeat hepatectomy A comparison of the characteristics between the first (n ¼ 44) and repeat (n ¼ 16) hepatectomy cohorts is shown in Table III. The maximum tumor size and the rate of preoperative treatment were different between the 2 groups. No significant differences in shortterm outcomes were detected between the first and repeat hepatectomy cohorts (Supplementary Table I). The PFS curves are shown in Fig 3. The respective median PFS times after a first and a repeat hepatectomy were 13 and 6 months (95% CI, 8e18 and 2e10, respectively) and were not different (P ¼ .546). Discussion Curative liver resection and debulking procedures are recommended as the treatments of choice for selective patients
Table III Comparison of characteristics between first and repeat hepatectomy
Liver metastasis factors CEA (ng/mL)y CA19-9 (ng/mL)y Synchronous onset Interval from primary resection (mo)y Interval from first hepatectomy (mo)y Number of nodulesy Maximum tumor size (mm)y Preoperative treatment Extra hepatic metastasis Operative procedure limited/anatomic Curability R0/R1/R2
First hepatectomy (n ¼ 44)
Repeat hepatectomy (n ¼ 16)
P value*
2.7 (0.9e22.0) 15.0 (1.0e575.0) 31 (70.5) 37 (4e151)
3.5 (0.7e8.4) 16.0 (1.0e65.0) 10 (62.5)
.637z .720z .564 d d .145z .008z .014 .543 .083 .862
7 (1e139) 28.0 (9.0e130.0) 14 (23.3) 1 (2.3) 25 (56.8) 19/14/11
Values in parentheses are percentages unless indicated otherwise. * c2 test was performed except Mann-Whitney U test. z Mann-Whitney U test. y Values are median (range).
16 (3e97) 4 (1e24) 15.0 (6.0e110.0) 11 (68.8) 0 (0.0) 13 (81.3) 7/6/3
S. Kiritani et al. / Surgery xxx (2019) 1e6
underwent a repeat hepatectomy, which was more than the percentage in the previous report, had acceptable oncologic effects. Spolverato et al also reported that the 5-year OS of a group of patients with recurrent hepatic NELM who had no operative reintervention was 45.9%, which was less than the result for repeat hepatectomy in our study (69%). Nonoperative treatment, including streptozocin (7 patients), everolimus (2 patients), radiofrequency ablation (2 patients), and best supportive care (1 patient), were administered to the remaining 12 patients. The median PFS of streptozocin and other treatments were each 3.5 and 3.0 months (95% CI, 3e5 and 1e8), which were less than that of repeat hepatectomy (6.0 mo; 95% CI, 2e10). The PFS curves after the first hepatectomy and those after the repeat hepatectomy were similar. Similar findings have been reported for patients with colorectal liver metastasis26e28; however, no previous reports, to our knowledge, have reported such findings in patients with NELM. The characteristics of the first hepatectomy cohort and those of the repeat hepatectomy cohort were similar except for the maximum tumor size. The tumor size was smaller in the repeat hepatectomy cohort than in the first hepatectomy cohort, primarily because of the close follow-up after the first hepatectomy. Although the smaller tumor size might have had a favorable effect on the long-term outcomes of the repeat hepatectomy cohort, this factor was excluded from the independent prognostic factors after both the first and repeat hepatectomy, indicating a limited prognostic impact. Additionally, the short-term outcomes of a repeat hepatectomy, despite this procedure being more technically demanding, were comparable to those of a first hepatectomy. The aforementioned results supported the usefulness of repeated hepatectomy for recurrent NELM. Although certain oncologic benefits of hepatectomy, including both a first and a repeat hepatectomy, were recognized from our results, confirming that patients who undergo a hepatectomy for NELM can obtain a true cure is difficult for the following reasons. In the present study, the OS curve after the initial hepatectomy decreased even after 10 years, and only 1 patient had reached a 10year PRS after a first hepatectomy; none of the patients undergoing a repeat hepatectomy achieved such a milestone (Fig 3). A greater observation period is thus needed to evaluate the true curability of repeat hepatectomy. In the multivariable analysis of recurrent NELM patients in the present study, a high CA19-9 value, multiple nodules, and treatment other than a hepatectomy were shown to be independently correlated with a poor outcome. For the initial NELM, CA19-9 was usually not associated with differentiated NET, although several factors, such as the amount of tumor, operative curability, presence of extrahepatic disease, functionality, or poor differentiation, have been reported as prognostic factors.9,24,29,30 A retrospective cohort study by Luo et al partially supported our results that a high serum CA19-9 level was correlated with lymph node metastasis and a poor prognosis in patients with pancreatic NET. The present study, however, is slightly different from the report by Luo et al31; Luo et al suggested a relationship between the CA19-9 level and the biology of primary pancreatic NET, the biology of recurrent NELM was associated with CA19-9 level at the resection of recurrent NELM, not at the resection of the primary site or at the first hepatectomy in our study. Because any relation between an increase in CA19-9 and poor outcome for recurrent NELM remains unknown, further research would be warranted. The present study has several limitations. First, retrospective nature of this study may have led to a favorable patient selection especially for recurrent NELM cohorts. A possible selection bias may influence the result of comparison between OS after the
5
first hepatectomy and that after the repeat hepatectomy, because the patients undergoing the second hepatectomy may have more favorable characteristics considering the technical and anatomic difficulty. The type of hepatic resection would often be limited, because the hepatic parenchymal volume would be decreased from the first hepatectomy. Second, there were several missing values in this study. Levels of neuronspecific enolase, which was reportedly a potential poor prognostic marker of NETs, was not measured in all patients in the present study because the measurement of it was unavailable and not covered by health insurance system in Japan. The utility of this marker should be clarified in further studies. Although a greater Ki-67 labeling index in NELM compared to primary tumor has been reported as a poor prognostic factor,32 we did not measure the Ki-67 labeling index in the initial and recurrent NELM in most patients because the period of this study was long and its importance was only recognized recently. Finally, the relatively small numbers of patients undergoing a first and a repeat hepatectomy might not reveal all the possible findings.25 In conclusion, although NELM frequently reoccurs after curative surgery, a repeat hepatectomy for recurrent NELM may be beneficial in selected patients. Conflict of interest/Disclosure The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. Funding/Support This work was not supported by any grants. Supplementary materials Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.1016/j.surg.2019. 08.020. References 1. Yao JC, Hassan M, Phan A, 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;26:3063e3072. 2. Lawrence B, Gustafsson BI, Chan A, et al. The epidemiology of gastroenteropancreatic neuroendocrine tumors. Endocrinol Metab Clin North Am. 2011;40:1e18, vii. 3. Dasari A, Shen C, Halperin D, et al. Trends in the incidence, prevalence, and survival outcomes in patients with neuroendocrine tumors in the United States. JAMA Oncol. 2017;3:1335e1342. 4. Pape UF, Jann H, Muller-Nordhorn J, et al. Prognostic relevance of a novel TNM classification system for upper gastroenteropancreatic neuroendocrine tumors. Cancer. 2008;113:256e265. 5. Majeed F, Porter TR, Tarantolo S, et al. Carcinoid crisis and reversible right ventricular dysfunction after embolization in untreated carcinoid syndrome. Eur J Echocardiogr. 2007;8:386e389. 6. Fujie S, Zhou W, Fann P, et al. Carcinoid crisis 24 hours after bland embolization: A case report. Biosci Trends. 2010;4:143e144. 7. Cives M, Strosberg J. Treatment strategies for metastatic neuroendocrine tumors of the gastrointestinal tract. Curr Treat Options Oncol. 2017;18:14. 8. Kunz PL. Carcinoid and neuroendocrine tumors: Building on success. J Clin Oncol. 2015;33:1855e1863. 9. Frilling A, Modlin IM, Kidd M, et al. Recommendations for management of patients with neuroendocrine liver metastases. Lancet Oncol. 2014;15:e8ee21. 10. Zhang XF, Beal EW, Chakedis J, et al. Early recurrence of neuroendocrine liver metastasis after curative hepatectomy: Risk factors, prognosis, and treatment. J Gastrointest Surg. 2017;21:1821e1830. 11. Gurusamy KS, Ramamoorthy R, Sharma D, et al. Liver resection versus other treatments for neuroendocrine tumours in patients with resectable liver metastases. Cochrane Database Syst Rev. 2009:Cd007060. 12. Pavel M, Baudin E, Couvelard A, et al. ENETS Consensus Guidelines for the management of patients with liver and other distant metastases from
6
13. 14.
15.
16. 17. 18.
19.
20.
21.
S. Kiritani et al. / Surgery xxx (2019) 1e6 neuroendocrine neoplasms of foregut, midgut, hindgut, and unknown primary. Neuroendocrinology. 2012;95:157e176. Farley HA, Pommier RF. Treatment of neuroendocrine liver metastases. Surg Oncol Clin N Am. 2016;25:217e225. Moertel CG, Lefkopoulo M, Lipsitz S, et al. Streptozocin-doxorubicin, streptozocin-fluorouracil or chlorozotocin in the treatment of advanced isletcell carcinoma. N Engl J Med. 1992;326:519e523. Rinke A, Muller HH, Schade-Brittinger C, et al. Placebo-controlled, doubleblind, prospective, randomized study on the effect of octreotide LAR in the control of tumor growth in patients with metastatic neuroendocrine midgut tumors: A report from the PROMID Study Group. J Clin Oncol. 2009;27: 4656e4663. Raymond E, Dahan L, Raoul JL, et al. Sunitinib malate for the treatment of pancreatic neuroendocrine tumors. N Engl J Med. 2011;364:501e513. Caplin ME, Pavel M, Cwikla JB, et al. Lanreotide in metastatic enteropancreatic neuroendocrine tumors. N Engl J Med. 2014;371:224e233. Yao JC, Fazio N, Singh S, et al. Everolimus for the treatment of advanced, nonfunctional neuroendocrine tumours of the lung or gastrointestinal tract (RADIANT-4): A randomised, placebo-controlled, phase 3 study. Lancet. 2016;387:968e977. Kubota K, Makuuchi M, Kusaka K, et al. Measurement of liver volume and hepatic functional reserve as a guide to decision-making in resectional surgery for hepatic tumors. Hepatol. 1997;26:1176e1181. Strasberg SM. Nomenclature of hepatic anatomy and resections: A review of the Brisbane 2000 system. J Hepatobiliary Pancreat Surg. 2005;12: 351e355. Sarmiento JM, Heywood G, Rubin J, et al. Surgical treatment of neuroendocrine metastases to the liver: A plea for resection to increase survival. J Am Coll Surg. 2003;197:29e37.
22. Fraenkel M, Kim M, Faggiano A, et al. Incidence of gastroenteropancreatic neuroendocrine tumours: A systematic review of the literature. Endocr Relat Cancer. 2014;21:R153eR163. 23. Kunz PL, Reidy-Lagunes D, Anthony LB, et al. Consensus guidelines for the management and treatment of neuroendocrine tumors. Pancreas. 2013;42: 557e577. 24. Maxwell JE, Sherman SK, O'Dorisio TM, et al. Liver-directed surgery of neuroendocrine metastases: What is the optimal strategy? Surgery. 2016;159:320e333. 25. Spolverato G, Bagante F, Aldrighetti L, et al. Management and outcomes of patients with recurrent neuroendocrine liver metastasis after curative surgery: An international multi-institutional analysis. J Surg Oncol. 2017;116:298e306. 26. Oba M, Hasegawa K, Shindoh J, et al. Survival benefit of repeat resection of successive recurrences after the initial hepatic resection for colorectal liver metastases. Surgery. 2016;159:632e640. 27. Bozzetti F, Bignami P, Montalto F, et al. Repeated hepatic resection for recurrent metastases from colorectal cancer. Br J Surg. 1992;79:146e148. 28. Adair RA, Young AL, Cockbain AJ, et al. Repeat hepatic resection for colorectal liver metastases. Br J Surg. 2012;99:1278e1283. 29. Mayo SC, de Jong MC, Pawlik TM. Surgical management and emerging therapies to prolong survival in metastatic neuroendocrine cancer. Ann Surg Oncol. 2011;18:S220eS221; author reply S222e233. 30. Bagante F, Spolverato G, Merath K, et al. Neuroendocrine liver metastasis: The chance to be cured after liver surgery. J Surg Oncol. 2017;115:687e695. 31. Luo G, Jin K, Cheng H, et al. Carbohydrate antigen 19-9 as a prognostic biomarker in pancreatic neuroendocrine tumors. Oncol Lett. 2017;14:6795e6800. 32. Lv Y, Han X, Huang C, et al. Risk factors affecting prognosis in metachronous liver metastases from WHO classification G1 and G2 gastroenteropancreatic neuroendocrine tumors after initial R0 surgical resection. BMC Cancer. 2019;19:335.