Molecular Characterization and Treatment of Pancreatic Neuroendocrine Tumors

Ana Maria Cristina De Jesus-Acosta, MD, is an assistant professor of oncology at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins Medicine.

ONCOLOGY recently spoke with Ana Maria Cristina De Jesus-Acosta, MD, an assistant professor of oncology at Johns Hopkins University, about the molecular characterization and treatment of pancreatic neuroendocrine tumors.

Q: What are pancreatic neuroendocrine tumors, and how frequently are patients typically diagnosed at an earlier or later stage?

DR. DE JESUS-ACOSTA: Neuroendocrine tumors are tumors that arise within the neuroendocrine tissues of the gastrointestinal tract or the pulmonary tract. However, when they originate from the islets of Langerhans within the pancreas, we call them pancreatic neuroendocrine tumors. This is not a very common tumor type compared with other cancers. Pancreatic neuroendocrine tumors occur in approximately 1 to 3 cases for every 100,000 individuals per year in the United States. These tumors also account for around 1% to 2% of all newly diagnosed pancreatic tumors. Certainly, with improvements in medical imaging techniques and an increased number of diagnostic procedures, we are identifying these tumors more often. Most patients do not have symptoms early on, so it’s not until later, when they have advanced disease or metastatic disease, that they tend to develop symptoms and seek medical guidance. Therefore, most of the patients we see have stage IV disease at the time of presentation.

Q: What have we learned about the molecular biology of this tumor type?

DR. DE JESUS-ACOSTA: Most of the knowledge that we have about the mutations and pathways involved in the biology of pancreatic neuroendocrine tumors comes from genomic exome sequencing results that were originally reported in 2011. The most frequent somatic mutations that have been identified occur in genes that encode for proteins implicated in chromatin remodeling. Approximately 44% of these tumors have a somatic inactivating mutation in a gene called MEN1, which is a tumor suppressor gene, and is linked to the menin pathway. There are around 43% of pancreatic neuroendocrine tumor cases that have mutations in genes that encode for either a transcription or chromatin remodeling complex composed of death-domain associated protein, also know as DAXX, and the thalassemia/mental retardation syndrome X-linked, which is called ATRX. This is also called the DAXX/ATRX pathway implicated in this disease. More importantly, around 14% of patients will have mutations in the genes that are key in the mammalian target of rapamycin, also known as the mTOR pathway, and mutations in the phosphatidylinositol 3-kinase pathway. Pancreatic neuroendocrine tumors are highly vascular tumors, so angiogenesis pathways are also very important for this disease. VEGF is expressed in approximately 78% to 80% of patients with pancreatic neuroendocrine tumors.

Knowledge of the molecular biology in pancreatic neuroendocrine tumors is very important and has clinical implications. For example, everolimus (Afinitor), which is an mTOR inhibitor, is now approved by the FDA for all patients with neuroendocrine tumors, including pancreatic primary tumors. Another agent, sunitinib (Sutent), that targets the VEGF receptor is also approved by the FDA specifically for patients with pancreatic neuroendocrine tumors, after a large phase III study noted improvement in progression-free survival in patients given sunitinib compared with placebo. Similarly, there are other agents that target the VEGF receptor. These are being tested at different stages in clinical trials. We have results for some and are still waiting for the results of other clinical trials.

Q: You mentioned some of the therapies used to treat patients with pancreatic neuroendocrine tumors. What are the initial, frontline therapy options and are there variations in therapies based on the biology and genetics of the tumor?

DR. DE JESUS-ACOSTA: This depends on the initial stage at presentation. Ideally, we want to see these patients in a multidisciplinary clinical setting, along with medical oncologists, surgeons, interventional radiologists, and other physicians. For patients that have resectable disease or locoregional disease, surgery is used as upfront therapy and is recommended with a curative intent. For patients with locally advanced but unresectable disease or stage IV metastatic disease that is not amenable for resection, we offer systemic therapies upfront as the preferred approach. Among the different types of systemic therapies, somatostatin analogues are most commonly used upfront. Most patients with pancreatic neuroendocrine tumors or any neuroendocrine tumors in general have high expression of somatostatin receptors. As such, when we use somatostatin analogues, we can delay progression in patients with pancreatic neuroendocrine tumors and we have shown that improves symptoms in those with functional neuroendocrine tumors or functional pancreatic neuroendocrine tumors. Another agent that can be used in the frontline setting is everolimus, which is also preferred for those patients that do not express somatostatin receptors. As of now, the genetics of an individual patient or the molecular pathways of an individual patient are not used to select a therapy upfront, but this may be an area of investigation down the road.

Q: Based on what we know about the molecular biology of these tumors, are there any investigational targeted agents being tested in clinical trials for subsets of patients with pancreatic neuroendocrine tumors?

DR. DE JESUS-ACOSTA: It is very difficult to do clinical trials specifically for certain molecular subtypes in patients with pancreatic neuroendocrine tumors, primarily because these types of tumors have a very low incidence. If we subclassify or subdivide patients based on the molecular profiling, we may not have enough patients to enroll them very quickly in a clinical trial. In general, clinical trials allow patients with any type of neuroendocrine tumor to participate or we can do clinical trials specifically for pancreatic neuroendocrine tumors without limiting them to specific molecular subtypes. On the other hand, there are other ongoing trials such as NCI-MATCH that can assign patients to a specific therapy based on the genetic changes found in these tumors through genetic sequences for individual patients. This approach seeks to determine whether cancer treatment based on its specific genetic changes is effective irrespective of the primary cancer type. For patients with pancreatic neuroendocrine tumors, if you know there are certain specific mutations or some pathways that are expressed based on genomic sequencing, you can refer these patients to the NCI-MATCH study, but we do not have clinical trials that are specifically for molecular subtypes in pancreatic neuroendocrine tumors.

Q: Are there other clinical trials of drugs for these patients, specific for pancreatic neuroendocrine tumors or ones that include these patients among others, that are showing some signs of efficacy or have a novel mechanism of action that you could highlight?

DR. DE JESUS-ACOSTA: Yes, there is significant interest in a drug called cabozantinib (Cabometyx). It is a tyrosine kinase inhibitor that targets multiple pathways, including VEGF and c-Met, which are important for pancreatic neuroendocrine tumors and neuroendocrine tumors in general. There is a phase II study that demonstrated promising results in pancreatic neuroendocrine tumors and other carcinoid tumors with an increase in response rate and progression-free survival. Currently, there is a large, ongoing, phase III study by the Alliance for Clinical Trials in Oncology, which is testing the efficacy of cabozantinib compared with placebo in patients with advanced neuroendocrine tumors. Patients with pancreatic primary tumors can participate in this clinical trial. Other studies are investigating additional VEGF inhibitors, such as pazopanib (Votrient); they have already completed enrollment and we are waiting for the results.

Another topic of interest is immunotherapy. We know that immunotherapy enhances the individual’s own immune system and it is an effective way to treat patients with other tumor types. For neuroendocrine tumors, the benefit of immunotherapy is still unknown. There have been a couple of clinical trials specifically using checkpoint inhibitors, such as programmed cell death protein 1 or programmed death-ligand 1 inhibitors, that have shown limited efficacy in patients with neuroendocrine tumors in general. However, there are still other ongoing trials that are testing these agents in combination with other immunotherapy drugs such as cytotoxic T-lymphocyte-associated protein 4 and those have shown some early promising results based on abstracts that have been presented at recent clinical meetings.

Key Question

What are the therapy options for relapsed and refractory disease?

DR. DE JESUS-ACOSTA: Everolimus and sunitinib can be used if they were not used in the frontline setting for these patients. Patients with somatostatin receptor expression as evidenced by an octreoscan or gallium-68 imaging can be offered treatment with peptide receptor radionuclide therapy (PRRT) using lutetium-177. This therapy is also known as Lutathera. PRRT was approved in the United States based on results from the NETTER-1 trial. This phase III clinical trial evaluated PRRT in patients with gastrointestinal neuroendocrine tumors. Results demonstrated that patients treated with lutetium-177 PRRT in combination with best supportive care had improved progression-free survival, compared with the control arm. Lutetium-177 PRRT was associated with a 79% reduction in the risk of disease progression versus single-agent octreotide treatment. As a result, the FDA approved lutetium-177 PRRT in early 2018 for the treatment of patients with pancreatic neuroendocrine tumors who have disease progression following frontline therapy.

Financial Disclosure: The authors have no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.

PERSPECTIVE

Identifying Pathway Alterations Impacts Development of Novel Systemic Therapies

Thorvardur R. Halfdanarson, MD, and Mohamad B. Sonbol, MD

Pancreatic neuroendocrine tumors (pNETs) are uncommon but their incidence in the United States has tripled in the last 2 decades, likely owing to earlier and incidental diagnoses. Multiple systemic treatment agents have activity but the optimal sequence of therapies in patients with advanced disease is not known. With the introduction of peptide receptor radionuclide therapy (PRRT), treatment has become more complicated and there is a need for comparative studies of the available agents. Although PRRT has yet to be prospectively investigated in pNETs, multiple retrospective studies suggest good efficacy with favorable progression-free survival. There is also a need for a better understanding of how genomic alterations may predict outcomes of different therapies.

Whole-genomic sequencing has shown that pNETs have a unique genetic landscape compared with their exocrine counterpart and small bowel NETs. For example, compared with small bowel NETs, pNETs carry a significantly higher frequency of mutations in MEN1, FOXO3, ATRX, and TSC2 genes. A recent study also showed that sporadic pNETs frequently harbor germline mutations, including mutations in the DNA repair genes MUTYH, CHEK2, and BRCA2. Overall, more than half of pNETs carry somatic mutations at the ATRX, DAXX, or MEN1 (A-D-M) genes, which is associated with inferior prognosis compared with A-D-M wild-type status. This A-D-M status (mutated vs wild) can potentially be helpful in stratifying the prognosis of well-differentiated pNETs and aid in counseling patients. However, results of studies published thus far have been somewhat discordant, perhaps due to the disease stage of the cohorts that were analyzed. Loss of function of ATRX and DAXX is associated with alternative lengthening of telomeres (ALT), which in turn is associated with adverse outcomes. The presence of ALT, which can be conveniently detected with fluorescence in situ hybridization, is associated with an increased risk of metastases in small pNETs, and may be a useful tool to predict outcomes following resection. Further validation of ALT as a prognosticator is needed. The somatic mutations in pNETs are commonly found in genes involved in 4 main pathways: chromatin remodeling, DNA damage repair, activation of the mammalian target of rapamycin (mTOR) signaling, and telomere maintenance.

The identification of alterations of these pathways may have important therapeutic implications in terms of developing novel systemic therapies for pNETs and improving upon currently known active drugs in pNETs (eg, everolimus with mTOR inhibition). While the reported alterations do not yet routinely inform treatment decisions, genomic predictors of outcomes should be incorporated in future prospective trials of therapies for pNETs as correlative studies.

Financial Disclosure: The authors have no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.

Dr. Halfdanarson is a medical oncologist focusing on gastrointestinal malignancies, especially neuroendocrine tumors and hepatobiliary tumors, at Mayo Clinic.

Dr. Sonbol is an oncologist at Mayo Clinic, with special interest in neuroendocrine tumors and gastrointestinal malignancies.