Patient-derived organoids for the personalized treatment of pancreatic neuroendocrine tumor with liver metastases: A case report

Abstract

BACKGROUND

Liver metastases are very common in pancreatic neuroendocrine tumors (pNETs). When surgical resection is possible, it is typically associated with survival benefits in patients with pNET and liver metastases. Patient-derived organoids are a powerful preclinical platform that show great potential for predicting treatment response, and they have been increasingly applied in precision medicine and cancer research.

CASE SUMMARY

A 51-year-old man was admitted to the hospital with the chief complaint of intermittent dull pain in the upper abdomen for over 3 years. Computerized tomography showed multiple space-occupying lesions in the liver and a neoplasm in the pancreatic body. Pathological results suggested a grade 3 pancreas-derived hepatic neuroendocrine tumor. In combination with relevant examinations, the patient was diagnosed with pNET with liver metastases (grade 3). Transarterial chemoembolization was initially performed with oxaliplatin and 5-fluorouracil, after which the chemotherapy regimen was switched to liposomal irinotecan and cisplatin for a subsequent perfusion, guided by organoid-based drug sensitivity testing. Following interventional treatment, the tumor had decreased in size. However, due to poor treatment compliance and the patient’s preference for surgical management, multiple resections were performed. Postoperatively, liposomal irinotecan combined with cisplatin was continuously admnistered. To date, the patient has survived 18 months with tumor presence, and tumor markers have returned to normal.

CONCLUSION

This case suggests that patient-derived organoids can aid in the optimization of therapeutic decisions in pNET patients with liver metastases to guide personalized treatment and improve survival outcomes.

Key Words: Pancreatic neuroendocrine tumors; Liver metastases; Patient-derived organoids; Personalized treatment; Transarterial chemoembolization; Case report

Core Tip: Patient-derived organoids (PDOs) show great potential for predicting treatment response in various cancer types. Although gastroenteropancreatic neuroendocrine neoplasm organoids can retain the pathohistological and functional phenotypes of parental tumors, their roles in predicting clinical outcomes are not known. Here, we report a case of pancreatic neuroendocrine tumor with liver metastases in which personalized treatment guided by PDO-based drug sensitivity testing enabled successful surgery and resulted in a favorable prognosis. Therefore, for patients with pancreatic neuroendocrine tumor patients and liver metastases, PDOs can help identify effective chemotherapy regimens, enabling personalized treatment, facilitating surgical resection, and improving survival outcomes.

INTRODUCTION

Pancreatic neuroendocrine tumors (pNETs) are a rare and heterogeneous group of neoplasms arising from pancreatic islet cells, with variations in histology, clinical characteristics, and prognosis[1]. They may present as non-infiltrative, slow-growing tumors, locally invasive tumors, or even rapidly metastasizing tumors[2]. Most metastases localize to the liver, and approximately 28%-77% of patients with pNETs will experience liver metastases in their lifetime[3]. Patients with liver metastases may be subjected to local complications such as biliary obstruction, liver insufficiency, and carcinoid syndrome. Additionally, liver metastases are a major risk factor for the prognosis of patients with pNETs[4]. When feasible, surgical resection is significantly associated with the best long-term survival outcomes[5]. Therefore, for patients with pNET liver metastases, comprehensive assessment and multidisciplinary approaches are required to determine the feasibility of surgical resection and the optimal treatment to improve the prognosis.

Over the past decade, the advent of in vitro three-dimensional technologies including organoids has revolutionized the development of human cancer models. Patient-derived organoids (PDOs), an in vitro three-dimensional microstructure, can faithfully recapitulate the intricate spatial architecture and cell heterogeneity of the tissue, and simulate the biological behaviors and functions of parental tumors while preserving biological, genetic and molecular features[6,7]. As a powerful preclinical platform, PDOs have been increasingly applied in precision medicine and cancer research. Importantly, there is a significant association between the use of PDO-based drug sensitivity testing and clinical responses to chemotherapy, radiotherapy and targeted therapy in multiple cancer types[8-10]. Although gastroenteropancreatic neuroendocrine neoplasm organoids have been confirmed to retain the pathohistological and functional phenotypes of parental tumors[7], their roles in the prediction of clinical outcomes have not been presented. Here, we report a case of pNET with liver metastases who successfully received surgical resection after personalized treatment guided by PDO-based drug sensitivity testing, resulting in a favorable prognosis.

CASE PRESENTATION

Chief complaints

A 51-year-old man was admitted to the hospital with the primary complaint of intermittent dull pain in the upper abdomen for over 3 years.

History of present illness

Intermittent dull pain in the upper abdomen for over 3 years.

History of past illness

The patient had no history of malignancies.

Personal and family history

The patient denied any personal and family history.

Physical examination

No abnormalities were found during the patient’s general physical examination.

Laboratory examinations

Laboratory examinations showed alpha fetal protein of 2.88 ng/mL, carbohydrate antigen 19-9 of 75.96 U/mL, carbohydrate antigen 15-3 of 35.10 U/mL and carcinoembryonic antigen of 1.93 ng/mL (Table 1).

Table 1 Changes in tumor markers before and after treatment.

Indicators	AFP (ng/mL)	CA19-9 (U/mL)	CEA (ng/mL)	CA15-3 (U/mL)
Before treatment	2.88	75.96	1.93	35.10
After second TACE	2.99	5.07	1.56	9.10
2 months after surgery	2.10	39.00	1.51	20.0
18 months after surgery	2.15	9.19	1.51	6.61

TACE: Transarterial chemoembolization; AFP: Alpha fetal protein; CA: Carbohydrate antigen; CEA: Carcinoembryonic antigen.

Imaging examinations

Computerized tomography (CT) scans showed multiple space-occupying lesions in the liver and a neoplasm in the pancreatic body (Figure 1). Positron emission tomography (PET)/CT examinations further revealed a pancreatic body tumor and multiple liver metastases. Meanwhile, endoscopic ultrasound-guided liver biopsy was performed, among which a portion of the biopsy sample was collected for organoid culture according to the protocols from Kingbio Medical (Chongqing) Co., Ltd. (China) after obtaining informed consent. The reagents related to organoid culture and drug sensitivity testing were all provided by Kingbio Medical (Chongqing) Co., Ltd. (China). Briefly, the biopsy tissue was minced and digested after rinsing with precooled phosphate-buffered saline, and cell pellets were collected by centrifugation. Matrigel was added, and the cells and Matrigel suspension were seeded onto a 6-well plate, with 100 cells per well. The plate was placed in an incubator at 37 °C for 15 minutes. Once the droplets were solidified, Jiabili^® culture medium [Cabio Biotech (Wuhan) Co., Ltd., China] for pancreatic cancer was supplemented. At last, the plate was put in a 37 °C, 5% CO₂ incubator for 13 days. Drug sensitivity testing was conducted when the organoids became solid spheroids with approximately 70 μm diameters (Figure 2). Subsequently, organoid cell masses were digested into single cells, and a total of 2 × 10⁴ cells were obtained. Cells were resuspended in Jiabili^® tumor tissue basic medium II [Cabio Biotech (Wuhan) Co., Ltd., China]. As described previously, organoid drug sensitivity testing concentrations were established based on the blood drug concentration of each drug, including the upper and lower limits of the blood drug concentration[11]. The pre-prepared drug solution and CelltiterGlo 2.0 (Promega Corporation, WI, United States) were supplemented successively, and a microplate reader was used to read the value. Notably, a binary classification method was utilized to evaluate the potential of drugs according to the quantification of the drug’s inhibitory effects on tumor organoid growth. The drug was considered resistant when the inhibitory rate < 50% or sensitive for an inhibitory rate ≥ 50%.

Figure 1 Computed tomography images of the patient before treatment, after transarterial chemoembolization, 2 months and 18 months after surgery. The orange arrows point to the liver lesions, and the yellow arrows point to the pancreatic lesions. TACE: Transarterial chemoembolization.

Figure 2  The growth status of organoids from liver metastatic lesions in a pancreatic neuroendocrine tumor patient.

FINAL DIAGNOSIS

Pathological results of the patient suggested a hepatic neuroendocrine tumor derived from the pancreas, with a propensity for grade 3. Immunohistochemistry from liver puncture tissues showed hepatocyte (-), cytokeratin (CK)-7 (-), cluster of differentiation (CD)-34 (blood vessels +), CD10 (+), CK18 (+), glypican-3 (-), CK (+), chromogranin A (+), synapsin (+), Ki-67 (30%), S100P (-), p53 (wild-type), somatostatin receptor 2 (+), and X-linked alpha-thalassemia/mental retardation syndrome (-) (Figure 3). In combination with relevant examinations, the patient was finally diagnosed as pNET with liver metastases (grade 3) and only palliative debulking could be performed.

Figure 3 Immunohistochemical and hematoxylin and eosin staining results of the liver puncture tissue from a patient with pancreatic neuroendocrine tumors. CK: Cytokeratin; CD: Cluster of differentiation; CgA: Chromogranin A; Syn: Synapsin; SSTR2: Somatostatin receptor 2; ATRX: X-linked alpha-thalassemia/mental retardation syndrome; HE: Hematoxylin and eosin.

TREATMENT

After discussion, the patient was advised to first receive anti-tumor treatment, followed by evaluation for later treatment. After communication with the patient and his family, transarterial chemoembolization (TACE) plus oxaliplatin (85 mg/m²), 5-fluorouracil (400 mg/m²) and 5-fluorouracil (2400 mg) was initially perfused for 23 hours. Subsequently, the chemotherapy drugs were switched to liposomal irinotecan (43 mg/m²) and cisplatin (60 mg) for the second 23 hours according to the PDO-based drug sensitivity testing results (Figure 4). CT images revealed that the largest tumor diameters in the pancreas and liver were decreased from 4.6 cm to 3.3 cm and from 3.2 cm to 2.5 cm after interventional treatment, respectively (Figure 1). Based on the Response Evaluation Criteria in Solid Tumors (version 1.1), partial response was achieved in the pancreatic lesion or in the hepatic lesion. However, there was a cancer thrombus at the confluence of the splenic vein and the superior mesenteric vein, and the patient had poor treatment compliance. Considering that the patient was willing to undergo surgical resection, pancreatic body-tail resection, partial resection and anastomosis of the portal vein, total splenectomy, and liver tumor resection were all performed after discussion. Postoperative pathological results showed pNET (grade 3) and hepatic neuroendocrine tumor (grade 3), in which tumor cell metastasis occurred in 1 of 9 lymph nodes. There were no tumor cells at the margins of the portal vein and the pancreas.

Figure 4 The growth status of organoids from liver metastatic lesions drug sensitivity testing result. FOLFIRI: Folinic acid, fluorouracil, and irinotecan; mFOLFOX6: Modified folinic acid, fluorouracil, and oxaliplatin 6.

OUTCOME AND FOLLOW-UP

Two months after surgery, CT scans showed multiple liver metastatic lesions (Figure 1). TACE with liposomal irinotecan and cisplatin was again continuously perfused for 23 hours. After surgical resection, the patient received five total treatment cycles. He is currently alive with tumors for 18 months (Figure 1), and the tumor markers have returned to normal (Table 1).

DISCUSSION

Liver metastases are very common in pNETs. When possible, surgical resection is typically associated with better survival in pNET patients with liver metastases[5,12]. According to data from a large, prospective database, hepatic resection was a significant independent factor for overall survival in neuroendocrine liver metastases[5]. In a previous meta-analysis, a median 5-year overall survival rate of 71% was pooled in patients receiving hepatic resection, and more than 95% of them achieved symptomatic relief[13].

Multimodal therapies including systemic chemotherapy are usually used for patients with liver metastases of non-neuroendocrine origin. Neoadjuvant therapy aims to promote tumor downstaging and maintain stable tumor biology, thus ensuring the feasibility of surgical resection. To identify patients with disease stability and improve survival rates, neoadjuvant chemotherapy is commonly utilized in multiple cancer types before hepatic resection[14,15]. In our study, the patient was diagnosed with pNET with liver metastases (grade 3). Considering that the patient was willing to receive surgical resection, antitumor treatment was recommended after discussion. Therefore, TACE with oxaliplatin and 5-fluorouracil was initially used, then replaced with liposomal irinotecan and cisplatin for the second perfusion based on organoid drug sensitivity testing results. The pancreatic and hepatic lesions both significantly decreased in size after organoid-guided treatment.

Over the past decade, PDOs have been established for various cancer types, with high success rates of generation. In four patients with metastatic neuroendocrine prostate cancer, organoids were successfully generated and showed a high concordance with the original tumors at the genomic, transcriptomic, and epigenomic levels. Meanwhile, organoid-based drug screening generated hypotheses for both single-agent and combination therapies, thereby supporting the clinical use of certain drugs and facilitating the development of novel treatment strategies[16]. Consequently, organoid-based drug screening can serve as a valuable tool for identifying rational therapeutic approaches in cancer. In another study, organoids from neuroendocrine carcinoma of the ampulla of Vater were established. Based on gene expression profiling of the organoids, sex-determining region Y-box transcription factor 9 was identified as a potential prognostic biomarker for biliary tract carcinoma[17]. Additionally, an organoid biobank of neuroendocrine neoplasms has been constructed[7]. Encouraged by these studies, we established pNET organoids from liver metastases and found that tumor lesion size decreased after treatment with drugs guided by PDO-based testing. Due to poor treatment compliance and the patient’s preference for surgery, multiple resections including pancreatic body-tail resection, partial resection and anastomosis of the portal vein, total splenectomy, and liver tumor resection were performed. Postoperatively, liposomal irinotecan plus cisplatin was continuously perfused. The patient has survived to date, and tumor markers have returned to normal.

Notably, the results of the PDO-based drug sensitivity testing in our study only reflected the direct-action ability of drugs on “isolated tumor cells” and could not fully represent their comprehensive therapeutic efficacy in complex microenvironment in vivo. For drugs like sorafenib and lenvatinib, whose mechanisms rely on the tumor microenvironment including the vascular system, the indirect cytotoxicity observed in in vitro assays needs to be interpreted independently in combination with drug action mechanisms. In subsequent studies, we plan to further verify the efficacy of such anti-angiogenic drugs by constructing the a model system containing vascular endothelial cells[18].

CONCLUSION

For pNET patients with liver metastases, PDOs are conducive for screening effective chemotherapy drugs for TACE to facilitate personalized treatment plans, providing opportunities for surgical resection by decreasing tumor size and a novel platform for improving survival outcomes.