A patient with a pancreatic neuroendocrine tumor and multiple liver metastases achieved a long-term partial response to third-line streptozocin treatment

Abstract

The Japanese guideline for gastroenteropancreatic neuroendocrine tumor treatment recommends everolimus or sunitinib as first-line treatment for unresectable pancreatic neuroendocrine tumors (PNETs). Streptozocin (STZ) is recommended as an alternative. We encountered a patient with PNET who had multiple liver metastases and who showed a remarkable response to third-line STZ. The patient was a 50-year-old man with a pancreatic head tumor 32 mm in diameter. We planned to perform subtotal stomach-preserving pancreaticoduodenectomy, but abandoned resecting the pancreas during the surgery upon discovering a small liver tumor 3 mm in diameter, examination of frozen sections of which revealed a poorly differentiated adenocarcinoma. However, the final pathological examination revealed that the liver nodule was a PNET; hence, we completed the subtotal stomach-preserving pancreaticoduodenectomy 5 weeks after the first laparotomy. The patient received no adjuvant chemotherapy after surgery. Twenty-one months later, we discovered 20 scattered liver metastases via computed tomography; these were considered unresectable. Therefore, we administered everolimus for 7 months, but the patient developed interstitial pneumonia and experienced metastatic progression. Subsequent sunitinib administration for 6 months was ineffective. Finally, we chose STZ (1000 mg/m², weekly) as a third-line treatment, which produced a partial response for 10 months. The patient remains alive 38 months after the detection of recurrence. As the order in which anti-cancer drugs should be administered to treat PNET has not been determined, additional predictors of their therapeutic efficacy should be investigated.

Introduction

Pancreatic neuroendocrine tumors (PNETs) usually progress slowly, and aggressive surgical resection is recommended even for patients with locally extensive disease. However, it is unclear whether aggressive surgery or antitumor drug administration should be the first choice of treatment for patients with multiple liver metastases. It has been reported that 20% of patients with PNET have distant metastases at the time of initial diagnosis [1, 2], most frequently in the liver [2]. According to the guidelines used in Japan [3], systemic treatment with everolimus [4, 5] or sunitinib [6] is recommended for unresectable PNET with distant metastases. Streptozocin (STZ), which had long been recommended as a treatment for PNET in Western countries, became available in Japan in September 2014 [7, 8]. We recently encountered a patient with PNET who had multiple liver metastases and showed a significant response only after third-line STZ treatment following disease progression with everolimus and sunitinib.

Case report

A 50-year-old man with obstructive jaundice caused by a pancreatic head tumor was referred to our hospital by a local physician. His total and direct serum bilirubin levels were 19.5 mg/dL and 14.5 mg/dL, respectively. Dynamic computed tomography (CT) showed a pancreatic head tumor that was 32 mm in maximum diameter and was less dense than the pancreatic parenchyma in the arterial phase (Fig. 1). Serum carbohydrate 19–9 was slightly elevated at 45.3 U/mL. He had no symptoms related to excess releasing hormone. Dynamic magnetic resonance imaging showed no liver metastases. A somatostatin receptor scintigraphy was not carried out because there was no positive suspicion of PNETs before operation. We diagnosed the patient with a resectable pancreatic head adenocarcinoma and did not perform a tumor biopsy. Instead, we planned a subtotal stomach-preserving pancreaticoduodenectomy (SSPPD) following biliary drainage with an endoscopic biliary stent tube. On laparotomy with an upper midline incision, we found a tiny tumor 3 mm in diameter on the liver surface. Pathologically frozen sections of this liver tumor revealed poorly differentiated adenocarcinoma. Therefore, we performed only palliative hepaticojejunostomy and needle biopsy of the pancreatic tumor. The postoperative pathological examination of the formalin-fixed specimens from the pancreas and liver tumor revealed a PNET. According to the World Health Organization 2017 classification [9], the grade was G2. We performed SSPPD 5 weeks later and removed all of the tumor detected; the resected pancreatic tumor measured 43 × 24 × 43 mm, and the cut surface comprised fusing white solid nodules. A histopathological examination revealed that the tumor was composed of trabecular and nested growths of cuboidal tumor cells that had oval nuclei with increased chromatin and eosinophilic cytoplasm. On immunohistochemistry, the tumor cells were positive for synaptophysin and somatostatin receptor type 2a (Fig. 2), and negative for somatostatin, insulin, pancreatic polypeptide, glucagon and gastrin. The Ki67 proliferation index was 5.9%. The final pathological diagnosis was PNET-G2, and 11 of 69 resected lymph nodes were positive for metastases.

Fig. 1.

Computed tomography findings. Plain and contrast-enhanced computed tomography showed a pancreatic head tumor measuring 32 mm in diameter and exhibiting lower signal density than the normal pancreas

Fig. 2.

Pathologic findings of the resected pancreatic tumor. The tumor was 43 × 24 × 43 mm in size. Grossly, the cut sections showed multiple white fused solid nodules. Histopathological examination revealed that the tumor was composed of trabecular and nested growths of cuboidal tumor cells that exhibited oval nuclei with increased chromatin and eosinophilic cytoplasm. Immunohistochemically, the tumor cells were positive for synaptophysin and somatostatin receptor type 2a antibodies

The patient did not receive postoperative adjuvant chemotherapy. After 21 months, a contrast-enhanced CT scan revealed 20 bilaterally scattered liver metastases, the largest of which was 20 mm in diameter (Fig. 3a, b). We selected everolimus with a daily dose of 10 mg as a first-line treatment, but the patient experienced moderate dermatitis and aggravated hemorrhoids. Furthermore, CT performed 7 months later showed disease progression (Fig. 3c) and revealed interstitial pneumonia. We stopped treatment with everolimus and switched to 37.5 mg/day sunitinib as a second-line treatment. CT performed 6 months later again showed disease progression (Fig. 3d). Thereafter, we began administering STZ monotherapy as a third-line treatment. The initial dose was 1000 mg/m² per week, and the relative dose intensity was maintained at 88.8% for 12 months. After the administration of the first dose of STZ, the patient developed hyperglycemia considered as a side effect of STZ that required insulin treatment. We reduced the dose of STZ starting from the second course; no subjective symptoms were observed with this reduced dose and hyperglycemia improved without insulin treatment. CT performed 3 months later revealed a partial response with a 74% size reduction in the liver lesions; these lesions had almost disappeared after 7 months of STZ treatment (Fig. 3e). However, tumor regrowth was observed 12 months after the start of the STZ therapy. At that point, we started another regimen that included a weekly administration of STZ at 500 mg/m² and 5-fluorouracil at 400 mg/m², with withdrawal every 6 weeks. The patient experienced adverse events including anemia, general fatigue, and nausea; moreover, the liver metastases progressed (Fig. 3f). This combination regimen was abandoned after three cycles. Further treatment with STZ plus lanreotide also ceased after two cycles owing to disease progression. After 35 months of drug therapy, the patient has been receiving only best supportive care and has remained alive for 59 months after SSPPD and 38 months after the tumor’s recurrence.

Fig. 3.

Time course of liver metastases as detected by computed tomography. a Before the initial surgery. b Multiple liver metastases appeared 21 months after pancreatoduodenectomy. c Disease progression after 7 months of treatment with everolimus. d Disease progression after 6 months of treatment with sunitinib. e A marked response was observed after 7 months of treatment with streptozocin. f The latest computed tomography scan obtained 52 months after pancreatoduodenectomy

Discussion

The progression-free survival periods of patients treated with everolimus, sunitinib, or STZ for advanced PNET are reportedly 19.5, 11.5, and 11.8 months, respectively [5, 8, 10]. Table 1 summarizes the outcomes of patients who were treated with each drug. It has remained unclear which of these three drugs should be used as first-line treatment.

Table 1.

Treatment outcomes in Japanese patients with unresectable pancreatic neuroendocrine tumors who were administered molecular-targeting agents and streptozocin

Agent	N	PFS (months) (Non-functional)	CR + PR (%)	CR + PR + SD (%)
Everolimus [5]	23	19.5	4	87
Sunitinib [10]	12	11.5 (12.1)	42	92
Streptozocin [8]	39	11.8 (16.8)	35.9	48.7

N number of patients, PFS progression-free survival, CR complete response, PR partial response, SD stable disease

Although STZ-based chemotherapy has been used as a first-line treatment for PNET, especially in Western countries, everolimus and sunitinib are usually used for earlier-line treatment in Japan. Shibuya et al. [11] reported that STZ is non-inferior to everolimus and sunitinib as first- or second-line treatment. Furthermore, weekly STZ monotherapy was found to be as safe and effective as a daily STZ regimen administered alone or in combination with 5-fluorouracil. Although our patient developed hyperglycemia after the administration of the first dose of STZ, there were no additional side effects, and the patient was able to continue receiving weekly STZ monotherapy for as long as 1 year. Shibuya et al. [11] reported that a Ki67 index > 5% predicted a response to STZ.

The difference in responses to each of the agents administered to our patient could be ascribed to their distinct pharmacological mechanisms. Everolimus is a mammalian target of rapamycin (mTOR) inhibitor and prevents cell division and proliferation. Sunitinib is a multi-target kinase inhibitor and inhibits vascular endothelial growth factor receptor, platelet-derived growth factor receptor, CD117, and other kinase-containing molecules. On the other hand, STZ is an agent that alkylates DNA and inhibits its synthesis. Temozolomide is a drug with a similar mechanism to STZ that is used to treat glioblastoma and other tumors. The status of the O6-methylguanine DNA-methyltransferase-encoding gene MGMT can also be used to predict the effects of temozolomide treatment, as the MGMT gene product is involved in repairing the DNA alkylation induced by temozolomide [12]. Therefore, temozolomide may be effective in patients with MGMT that itself contains an alkylated region. Kulke et al. [13] also suggested that the absence of MGMT might explain the sensitivity of some PNETs to temozolomide-based regimens. STZ, which is also an alkylating agent, presumably has the same mechanism of action as temozolomide, and research aimed at identifying the factors that predict the efficacy of STZ is being conducted [14]. Our patient may have a mutation in his MGMT gene, but this has not been confirmed (Fig. 4). The ability to predict the effects of STZ based on the presence/absence of MGMT gene mutations would aid in the selection of treatment regimens for patients with unresectable PNET.

Fig. 4.

A schematic representation of the pharmacological mechanism of streptozocin and the relationship between the streptozocin efficacy and the status of the MGMT gene

In summary, the administration of STZ as a third-line treatment in our patient with PNET who had multiple liver metastases showed a partial response. Currently, the optimal order of the administration of anticancer drugs for PNET has not been determined. Identifying the predictors of the therapeutic effects of such drugs against PNET remains desirable.

Funding

The study received no financial support.

Compliance with ethical standards

Conflicts of interest

The authors declare that they have no conflict of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from the patient reported in this study.