Endocrine Journal
Online ISSN : 1348-4540
Print ISSN : 0918-8959
ISSN-L : 0918-8959
ORIGINAL
A case of vasoactive intestinal peptide-secreting tumor (VIPoma) arising from MEN1 inactivation which recurred 15 years after the initial resection
Momoe SakuraiTetsuji WakabayashiYasuyuki KondoEriko IkedaKoichi WatanabeAkihito TakeiHiroaki OkazakiKenta OkadaKen EbiharaNobuyuki KakiuchiSeishi OgawaNoriyoshi FukushimaShun Ishibashi
Author information
JOURNAL FREE ACCESS FULL-TEXT HTML

2023 Volume 70 Issue 6 Pages 573-579

Details
Abstract

Vasoactive intestinal peptide-secreting tumors (VIPomas) are extremely rare functional pancreatic neuroendocrine neoplasms (p-NENs) characterized by watery diarrhea, hypokalemia, and achlorhydria. Here, we report the case of a 51-year-old female patient with VIPoma that recurred after a long-term disease-free interval. This patient had been asymptomatic for approximately 15 years after the initial curative surgery for pancreatic VIPoma, with no metastasis. The patient underwent a second curative surgery for the locally recurrent VIPoma. Whole-exome sequencing of the resected tumor revealed a somatic mutation in MEN1, which is reportedly responsible not only for multiple endocrine neoplasia type 1 (MEN1) syndrome but also sporadic p-NENs. Symptoms were controlled with lanreotide before and after surgery. The patient is alive with no relapse following 14 months after surgery. This case demonstrates the importance of long-term observation of patients with VIPoma.

PANCREATIC NEUROENDOCRINE NEOPLASMS (p-NENs) are rare heterogeneous tumors, comprising approximately 3–5% of all diagnosed pancreatic tumors [1]. They can be broadly classified into non-functional and functional (hormone-producing) tumors, including gastrinoma, insulinoma, glucagonoma, somatostatinoma, or vasoactive intestinal peptide-secreting tumor (VIPoma). VIPoma is extremely rare, with an estimated incidence of less than 1 in 10 million individuals per year in the general population [2]. Most VIPomas are sporadic, whereas 5% are associated with multiple endocrine neoplasia type 1 (MEN1) syndrome [3]. Excessive VIP secretion causes clinical symptoms characterized by refractory watery diarrhea, often associated with electrolyte imbalance, such as hypokalemia and achlorhydria, known as WDHA syndrome. Sixty to eighty percent VIPomas are metastatic at diagnosis, with the liver being the most frequent site of metastasis [3]. VIPomas often exhibit slow growth despite malignant behavior, and their prognosis depends on pathological characteristics, tumor development, metastatic spread, and surgical resectability [4]. The long-term outcome of VIPomas remains unclear due to their infrequency.

Here, we report a rare case of VIPoma that recurred approximately 15 years after initial curative resection.

Case Presentation

A 51-year-old female presented with watery diarrhea, hypokalemia, and hyperglycemia that had persisted for several years. At the age of 32, she was diagnosed with VIPoma that occurred in the uncinate process of the pancreas and underwent enucleation of the pancreatic tumor at our hospital. The pathology was consistent with R0 resection of p-NET grade 2 (synaptophysin (+), chromogranin A (+), neuron specific enolase (NSE) (+), VIP (+), and Ki-67 index:10.0%). After surgery, the symptoms were attenuated by normalizing the serum VIP concentration. She had been followed up at our hospital for several years; however, she discontinued the hospital visit at her discretion. At the age of 45, diabetes care was resumed at another hospital and glycemic control was maintained with oral antihyperglycemic drugs. At the age of 46 years, she began to present with watery diarrhea again; however, the recurrence of the pancreatic tumor was not observed on a computed tomography (CT) scan, which was performed at another hospital. At the age of 50, she developed symptoms of hypokalemia, which necessitated potassium replacement therapy. In addition, glycemic control deteriorated. She was suspected of having VIPoma recurrence and was referred to our hospital for further examination.

A contrast-enhanced CT and magnetic resonance imaging (MRI) scan showed a well-defined enhancing multinodular lesion, approximately 5 cm in size, with some cystic and calcified components localizing near the dorsal of the pancreatic body (Fig. 1A and 1B). No distant metastases were observed, including those in the liver. Endoscopic ultrasound-guided fine-needle aspiration of the lesion was performed, and histopathological evaluations were consistent with a neuroendocrine neoplasm. Somatostatin receptor scintigraphy was selectively positive for the pancreatic lesion (Fig. 1C). Since it was unclear whether the tumor had continuity with the pancreas, we could not clarify whether it was derived from the pancreas or lymph nodes. Retrospective analysis of previous CT images performed at other hospitals revealed that the lesion was detectable not at the age of 45 years but at the age of 47 years (Fig. 1A).

Fig. 1

Preoperative imaging

Representative images of computed tomography (CT) scans (A), abdominal magnetic resonance imaging (MRI) (B), and somatostatin receptor scintigraphy (C). The red arrows show the recurred VIPoma lesion.

Initial laboratory findings showed hypokalemia (2.3 mmol/L), hyperglycemia (preprandial glucose, 228 mg/dL; HbA1c, 10.8%), and metabolic alkalosis on blood gas test (pH: 7.51, bicarbonate level: 31.3 mmol/L) (Table 1). The stool osmotic gap was calculated to be less than 50 mOsm/kg, suggesting secretory diarrhea (Table 1). We measured the serum levels of VIP using a Human VIP ELISA Kit (Cat No: E-EL-H2155, Elabscience, Houston, TX, United States). The maximum VIP concentration (257 pg/mL) was detected to be above the normal range in several measurements. This value was similar to that at the initial presentation (318 pg/mL). Based on these results, the patient was diagnosed with VIPoma relapse.

Table 1 Laboratory findings
WBC 5,500/μL Endocrine data Blood gas
Hb 12.2 g/dL ACTH 3.9 pg/mL pH 7.51
Plt 24.4 × 104/μL Cortisol 11.9 μg/dL pCO2 40.0 mmHg
TP 5.8 g/dL TSH 1.14 μU/mL pO2 68.5 mmHg
Albumin 3.4 g/dL FT3 2.49 pg/mL HCO3 31.3 mmol/L
AST 27 U/L FT4 1.33 ng/dL BE 7.8 mmol/L
ALT 35 U/L Adrenalin 7 pg/mL
LDH 185 U/L Noradrenaline 98 pg/mL Stool
UN 6 mg/dL Dopamine 6 pg/mL OP 289 mOsm/kg
Creatinine 0.43 mg/dL Intact PTH 21 pg/mL Na 104 mmol/L
Na 147 mmol/L Aldosterone 5.2 pg/mL K 51.4 mmol/L
K 2.3 mmol/L PRA 2.1 ng/mL/hr
Cl 101 mmol/L Gastrin 120 pg/mL
Ca 9.1 mg/dL Insulin 3.1 μU/mL
P 2.6 mg/dL Glucagon 26.1 pg/mL
Amylase 41 U/L VIP 257 pg/mL
Glucose 228 mg/dL
HbA1c 10.8%

Abbreviations: WBC, white blood cell; Hb, hemoglobin; Plt, platelet; TP, total protein; AST, aspartate aminotransferase; ALT, alanine aminotransferase; LDH, lactate dehydrogenase; UN, urea nitrogen; ACTH, adrenocorticotropic hormone; TSH, thyroid-stimulating hormone; PTH, parathyroid hormone; PRA, plasma renin activity; VIP, vasoactive intestinal peptide; BE, base excess; OP, osmotic pressure.

Given the absence of detectable metastases, we hypothesized that curative surgery would be feasible. Surgical debulking was necessary to alleviate the symptoms, although not feasible. Administration of lanreotide (a long-acting somatostatin analog) twice prior to surgery alleviated the symptoms of diarrhea but did not improve hypokalemia. We performed an enucleation of the tumor as the lesion appeared to be localized with no apparent invasive growth.

The dissected tumor grossly measured 5.8 × 4.5 × 4.2 cm and contained yellow-white colored multinodular components with some calcifications (Fig. 2A). Microscopically, the main part of the tumor consisted of atypical cells arranged in organoid patterns, with abundant fibrovascular stroma (Fig. 2B). The tumor cells had a granular eosinophilic cytoplasm with small round nuclei. Most nuclei were uniform in size, and mitosis was less than two per ten high-power fields (HPF), suggesting well-differentiated cells. The specimen demonstrated blood vascular invasion without lymphovascular or perineural invasion. Surgical resected margins were tumor-free. No regional lymph node metastases were observed. There was no evidence of lymphatic tissues in the tumor; therefore, the tumor, which was suspected to be a recurrence in a lymph node preoperatively, might have originated either from the residual pancreatic tumor or de novo. Immunohistochemical staining was positive for synaptophysin (Fig. 2C), insulinoma-associated protein 1 (INSM1), and somatostatin receptor 2 (SSTR2) (Volante score = 3+), but negative for insulin, glucagon, somatostatin, gastrin, and serotonin (data not shown). Ki-67 index in a hot spot with manual counting was 19.4% (Fig. 2C). An immunohistochemical stain with anti-human VIP monoclonal antibody (Cat No: sc-25347, Santa Cruz Biotechnology, Dallas, TX, United States; 1:50) yielded a positive result (Fig. 2C). Collectively, our final diagnosis was grade 2 p-NET (VIPoma), according to the 2017 WHO classification, which was classified as stage II.

Fig. 2

Pathological findings of the resected tumor

Macroscopy of the resected tumor (A), hematoxylin and eosin staining of the resected tumor (magnification power, 40×) (B). Tumor cells were well-differentiated and arranged in organoid patterns. Immunohistochemical examination of the resected tumor (magnification power, 40×) (C) showed positivity for synaptophysin, Ki-67, and VIP. “Primary onset” shows the histology at the initial surgery, and “Recurrence” shows the histology at the secondary surgery.

Subsequently, we performed whole-exome sequencing of the resected tumor to confirm the genomic events that characterize the pathogenesis. To identify somatic mutations in the tumor, a matched pair of tumor and normal DNA obtained from the patient’s blood sample was used. The MEN1 gene, known as one of the causative genes of p-NEN, showed two-hit inactivation: a truncating mutation and a loss of heterozygosity (LOH) on chromosome 11, while there were no other significant somatic mutations in driver genes involved in tumor etiology (Table 2 and Fig. 3), suggesting that MEN1 inactivation was responsible for tumorigenesis.

Table 2 Somatic mutations revealed by whole-exome analysis
Gene Chr Ref Alternate Mutation Type Amino-acid change VAF
MEN1 11 GGCGGGTC frameshift deletion p.D70Wfs*44 0.659
TMEM30A 6 C T missense p.G21S 0.829
C1orf87 1 C T nonsense p.W5X 0.808
BCO2 11 C T missense p.P540S 0.805
BMPR1B 4 T C synonymous p.Y431Y 0.8
KDM4E 11 C T missense p.P473L 0.773
ZNF382 19 T C missense p.F388S 0.523
GTF2I 7 G C missense p.D118H 0.464
HK3 5 G A missense p.R644W 0.46
RASA1 5 G A missense p.M861I 0.455
SNRPB 20 C T synonymous p.R221R 0.449
SULT2B1 19 G C missense p.W174S 0.448
C12orf42 12 G A synonymous p.A349A 0.429
CETN1 18 G A missense p.E99K 0.426
PRKDC 8 T unknown 0.422
COL3A1 2 A C synonymous p.A667A 0.41
DNAH9 17 C T nonsense p.Q1225X 0.408
CPA4 7 G A synonymous p.S152S 0.377
TTN 2 C T synonymous p.V21102V 0.355
LGR6 1 C T missense p.A909V 0.286
TCHH 1 A T missense p.L1267Q 0.273
MKLN1 7 C T nonsense p.Q660X 0.265
CARMIL2 16 G A missense p.D580N 0.212
DSG3 18 G A missense p.D103N 0.137
CCP110 16 TGAGTTGGACATTAA non-frameshift p.E704_N708del 0.133
ANAPC10 4 C T deletion p.D12N 0.131
PROM2 2 G A missense p.C430Y 0.093
EYA4 6 T G missense p.S447R 0.066
KMT2C 7 T A missense p.Q873L 0.063
missense

Abbreviations: Chr, chromosome; Ref, reference; VAF, variant allele frequency

Fig. 3

Summary of copy-number aberrations

Copy-number analysis indicated the loss of chromosomes 1p, 4, 6, 10q, 11, 13q, and X, and the gain of chromosome 1q. The red arrow indicates the location of MEN1.

Post-surgery, the serum VIP concentration decreased to an undetectable range, and the symptoms, such as diarrhea and hypokalemia, had dramatically disappeared; however, the patient required insulin injection therapies, postoperatively, because of persistent inadequate glycemic control. Monthly lanreotide injections were administered as adjuvant therapy for more than 14 months. The patient is still alive without definite evidence of recurrence.

Discussion

Here, we report the case of a female patient with VIPoma that recurred after a long-term disease-free period following initial surgery. In general, most VIPomas are metastatic at diagnosis [3]; however, this case, fortunately, did not demonstrate metastasis either at the primary onset or at the time of recurrence, allowing curative surgeries. Nearly all recurrences in patients with p-NEN were detected within 10 years after surgery [5, 6]. To the best of our knowledge, no VIPoma cases that recurred more than 10 years after radical surgery have been reported in the literature. The adequate follow-up period after surgery for patients with VIPoma is controversial; however, it was suggested that long-term follow-up was crucial due to a high recurrence rate of 40.4% [7] and because some cases might demonstrate excessively slow progression, as observed in this patient. In particular, careful attention should be paid to the reappearance of symptoms that might precede the detection of abnormal findings in imaging tests, as observed in this case.

Lanreotide was used as neoadjuvant therapy in this case, resulting in the improvement of diarrhea. Somatostatin analogs (SSAs), including lanreotide, are considered to inhibit VIP secretion and are often used to control symptoms preoperatively [8]. In addition, lanreotide treatment was maintained as postoperative adjuvant therapy. Although SSAs have not yet been proven to prevent the relapse of VIPoma, we expected that lanreotide would minimize the chance of further relapse because of SSAs’ potential antiproliferation and tumor stabilization effects [2, 9].

Our patient had diabetes mellitus as a comorbidity since the initial onset of VIPoma. Hyperglycemia or impaired glucose tolerance reportedly affects 20–50% of patients with VIPoma [10]. Multiple mechanisms have been proposed to explain this association such as the direct glycogenolytic and gluconeogenic activity of VIP in the liver and the inhibitory effect of hypokalemia on insulin release from pancreatic beta cells [2]. Glycemic control appeared to be correlated with the tumor burden of the VIPoma at the time of recurrence, suggesting that VIPoma was involved in the worsening of glycemic control in this case.

The majority of p-NENs are sporadic, while 5–10% of them may arise from hereditary tumor syndromes such as MEN1, von Hippel–Lindau syndrome (VHL), neurofibromatosis type 1 (NF1), and tuberous sclerosis complex (TSC) [11]. MEN1 is the most common genetic syndrome associated with p-NEN, comprising 5% of pancreatic VIPomas [3]. Recent whole-exome analysis of p-NENs has shown that a few somatic alterations are commonly identified in genes implicated in four main pathways in sporadic p-NEN cases: chromatin remodeling, DNA damage repair, telomere maintenance, and activation of mTOR signaling [12-14]. Somatic mutations of MEN1 were found in 41–44% of patients with p-NEN, not only in MEN1 syndrome but also in sporadic cases [1]. MEN1, located on chromosome 11, encodes the protein menin, which plays an important role in chromatin remodeling, genomic stability, and histone methylation-driven gene transcription [15]. Wild-type menin negatively regulates cell proliferation and acts as a tumor suppressor for p-NEN development; hence, its loss drives tumor initiation and progression. Somatic MEN1 mutations have been found in approximately 45% of sporadic VIPoma cases [15-18]. In addition, DAXX/ATRX gene mutations, which participate in the alternative lengthening of telomeres and chromatin remodeling, and mTOR pathway gene mutations are known to cause p-NEN. In our case, whole-exome sequencing of a matched pair of tumor and normal DNA revealed two-hit MEN1 inactivation: a truncating mutation and LOH (Table 2 and Fig. 3). There were no germline gene alterations detected which indicated the possibility of hereditary tumor syndromes including MEN1. In addition, the tumor DNA showed no typical mutated driver genes, including DAXX/ATRX and mTOR pathway genes. We hypothesized that MEN1 inactivation was responsible for the VIPoma development in this case; however, it is possible that some chromosomal aberrations, such as loss of 1q, 6q, and 10q could have caused the tumor progression as they commonly identified in p-NENs [19].

The elevated serum VIP concentration in this patient was consistent with the diagnosis of VIPoma (in general, >200 pg/mL is diagnostic); however, the magnitude of elevation was relatively modest compared with the tumor size and symptoms. As previously reported, serum VIP levels vary considerably (100–7,200 pg/mL) [7]. VIPoma might secrete VIP intermittently, and the serum VIP level might not increase between episodes of diarrhea; thus, it is important to measure serum VIP concentrations repeatedly when the patient is symptomatic [20]. The reason for the discrepancy between VIP levels and the symptoms cannot be explained completely at present; however, it cannot be ruled out that the tumor may secrete unknown peptides that have physiological functions similar to those of VIP. Further research is needed to address this issue.

In conclusion, we report a case of VIPoma that recurred after a long-term disease-free interval. This case underscores the significance of long-term follow-up for patients with VIPoma. Postoperative lanreotide treatment may prevent further recurrence. Moreover, the whole-exome analysis revealed that two-hit inactivation of the MEN1 gene was involved in the development of VIPoma. Further studies are necessary to elucidate the molecular pathogenesis and clinical course of VIPomas, which are extremely rare.

Acknowledgments

This study was not funded by any grant. We are grateful to Yuichi Aoki for his support in collecting tumor samples. The patient provided informed consent for the publication of this report. We would like to thank Editage (www.editage.com) for the English language editing.

Ethics Approval and Consent to Participate

This study was approved by the ethics committee of Jichi Medical University. Written informed consent was obtained from the patient for genetic testing, publication of this case report, and any accompanying data and images. A copy of the written consent form is available for review.

Disclosure

The authors declare no conflicts of interest.

Authors’ Contributions

MS, TW, and YK: patient management and data acquisition; EI and NF: pathologic evaluation; KW, NK, and SO: genetic testing; MS, TW, AT, HO, KO, KE, and SI: conceptualization; MS, TW, and SI: manuscript editing. All authors have read and approved the final manuscript.

References
 
© The Japan Endocrine Society
feedback
Top