CE / CME
Physician Assistants/Physician Associates: 0.75 AAPA Category 1 CME credit
Nurses: 0.75 Nursing contact hour
Pharmacists: 0.75 contact hour (0.075 CEUs)
Physicians: maximum of 0.75 AMA PRA Category 1 Credit™
Released: December 30, 2024
Expiration: December 29, 2025
Introduction
In this module, Thor R. Halfdanarson, MD, and Jonathan Strosberg, MD, discuss the diagnosis, management, and treatment of gastroenteropancreatic (GEP) neuroendocrine tumors (NETs) along with current clinical evidence.
The key points discussed in this module are illustrated with thumbnails from the accompanying downloadable PowerPoint slideset, which can be found here or downloaded by clicking any of the slide thumbnails in the module alongside the expert commentary.
Clinical Care Options plans to measure the educational impact of this activity. Some questions are asked twice: once at the beginning of the activity, and once again after the discussion that informs the best choice. Your responses will be aggregated for analysis, and your individual responses will not be shared. Thank you in advance for helping us assess the impact of this education.
NETs Are Heterogeneous
Jonathan Strosberg, MD:
In addition to endocrine organs such as the thyroid gland or the pituitary gland, there are cells throughout the body that have an endocrine function. They produce hormones, communicate with the nervous system, and have microscopic similarities with neurons. NETs that arise from this diffused neuroendocrine system can be benign or malignant, often grow slowly, and have a propensity to produce hormones that can result in various unusual syndromes.1 The incidence of GEP-NETs increased by 6.4 times between 1975 and 2012 to 3.56/100,000 persons per year.2
NETs are also heterogeneous; each has unique characteristics, such as their pace and primary site of growth.3 They are distinguished as low grade, intermediate grade, or high grade based on either the mitotic rate or the Ki-67 proliferation index. Other identifying features include differentiation, with poorly differentiated cancers (meaning poorly resembling the endocrine cells of origin) often being extremely aggressive. NETs that produce hormones are referred to as functional and NETs that do not produce a clear hormonal syndrome are called nonfunctional. In addition, the level of SSTR expression has important therapeutic implications.
When considering treatments, we also can distinguish between liver-dominant tumors, where liver-directed therapy such as surgery or embolization is considered, and more widely metastatic disease, where tumors exist in other organs such as bone or in parts of the body such as the peritoneum and systemic therapies are usually considered.3
GEP-NET Grade Classification
Jonathan Strosberg, MD:
Tumor grade is a key aspect of classification.
Low-grade, or grade 1, tumors are defined as having a Ki-67 index of <3% and a mitotic index of <2 mitoses/10 hpf.4,5 Intermediate-grade (grade 2) tumors have a Ki-67 index of 3% to 20% or a mitotic rate of 2/10 hpf to 20/10 hpf. Finally, high-grade (grade 3) tumors have a Ki-67 index of >20% or a mitotic rate of >20/10 hpf. Typically, the Ki-67 index is higher than the mitotic rate, and therefore tends to be the dominant influence on grade classification.
Well-differentiated tumors can be grade 1, 2, or 3, but poorly differentiated NETs are referred to as neuroendocrine carcinomas and are always grade 3.
Diagnosing GEP-NETs
Jonathan Strosberg, MD:
Current guidelines outline that GEP-NET diagnosis and treatment should be based on key features of neuroendocrine neoplasms (NENs) such as proliferative activity, SSTR expression, tumor growth rate, and extent of disease.1 Hereditary background should be considered, especially with pan-NETs. Genetic testing should be done in patients with multiple endocrine neoplasia and in patients with a family history of NENs or associated diseases and characteristics suspicious of a hereditary disease, as well as in patients <40 years of age with gastrinoma.1
For proper pathologic diagnosis, morphology, grading, and immunohistochemical staining for chromogranin A (CgA) and synaptophysin should be carried out.1 Whole-body SSTR imaging (with 68Gallium/64Copper-SSTR-PET-CT, ideally) should be considered for initial tumor staging, and for preoperative imaging.1
Classification and Biomarkers for Diagnosing GEP-NETs
Jonathan Strosberg, MD:
The classification of NETs based on mitotic count and Ki-67 index is listed here. Higher mitotic count and Ki-67 index are associated with higher-grade NETs.1
Various biomarkers other than Ki-67 can be used in the diagnosis and treatment of GEP-NETs and they are listed here. These additional biomarkers are optional and not routinely used in clinical practice.1
Small Intestine (Midgut) NETs
Jonathan Strosberg, MD:
Now we discuss specific subtypes of GEP-NETs. Small intestine or midgut NETs typically originate in the terminal ileum.6 They are often multifocal; there is often more than 1 tumor within the same region, and they typically grow slowly but have a high propensity to metastasize, so we rarely see stage I small bowel NETs.7,8 They are likely to metastasize to the regional mesenteric lymph nodes, which often leads to desmoplastic fibrosis, where the metastases tether the surrounding small intestine and can cause bowel obstruction or ischemia.9,10
Midgut NETs produce hormones and vasoactive substances, namely serotonin and others such as prostaglandins, tachykinins, bradykinins, which can cause carcinoid syndrome and typical symptoms of flushing and diarrhea. Patients with carcinoid syndrome frequently have distant liver metastases.11 Of interest, because these tumors grow relatively slowly, patients can develop a high degree of liver tumor burden without having significant abnormalities in liver function.
Functional pNETs
Jonathan Strosberg, MD:
Pancreatic NETs account for 1% to 2% of pancreatic malignancies.12 Ten percent are functional, resulting in a characteristic constellation of symptoms determined by the hormone produced.
There are a variety of functional pNETs, such as insulinomas, caused by excessive secretion of insulin and resulting in hypoglycemia. The vast majority of these are small and are considered benign. Insulinomas can be a manifestation of multiple endocrine neoplasia type 1. They are usually sporadic with no underlying syndrome, and are easily cured with surgical resection. If they are metastatic, they can respond to peptide receptor radionuclide therapy (PRRT) and chemotherapy, which we discuss later, and diazoxide and everolimus can be used to control blood sugar levels.
Zollinger-Ellison syndrome is seen in patients with gastrinoma, which is the excessive production of gastrin that results in increased gastric acid. Symptoms include peptic ulcers, reflux, diarrhea, abdominal pain, and occasionally bowel perforations, typically gastric or duodenal. In addition to treating the tumor, it is important to treat the acid production with high doses of proton pump inhibitors such as omeprazole or pantoprazole.
Gastric NETs
Jonathan Strosberg, MD:
Gastric NETs come in 3 or possibly 4 different types.13 Most are type 1 and arise in chronic atrophic gastritis. Increased production of gastrin is thought to stimulate neuroendocrine hyperplasia and tumors. These tumors are often indolent and small (<10 cm) and can be treated with endoscopic resection and surveillance.14
Type 2 is an uncommon type and is associated with Zollinger-Ellison syndrome, and also with high gastrin levels.13 However, unlike with type 1 gastric NETs, the gastric pH is low. Typically the presentation is dominated by the features of Zollinger-Ellison syndrome and the tumors can be treated similarly as type 1 tumors.
Type 3 is the least well defined of the gastric NETs, with generally larger tumors and more aggressive, higher-grade disease. Type 3 gastric NETs are thought to require more radical oncologic resection, although there are some reports of endoscopic resections for smaller type 3 tumors.15
Last, what are sometimes referred to as type 4 gastric NETs may arise in patients who have been chronically treated with proton pump inhibitors. These are likely more benign or more indolent than some of the other gastric NETs.16
Appendiceal and Rectal NETs
Jonathan Strosberg, MD:
Appendiceal NETs are the most common tumors of the appendix and are usually found incidentally in 0.16% to 2.3% of patients who have appendectomies for acute appendicitis.6,17
Metastases are exceptionally rare and appendectomy is typically sufficient treatment for tumors that are <2 cm. For larger tumors, especially in younger patients, completion right hemicolectomy is recommended.
Rectal NET is another type of NET that is increasingly being diagnosed, probably largely because of increased screening with colonoscopies.18 They are most often found incidentally during lower endoscopy, account for 1% to 2% of all rectal tumors, and are usually small when diagnosed.13,19 If <2 cm, T1, and grade 1, they can be excised endoscopically. If they grow beyond 1 or 2 cm or invade the muscularis propria, they can metastasize to pelvic regional lymph nodes and typically require oncologic resection.20
Basics of Treatment
Jonathan Strosberg, MD:
Generally, if the disease is local, depending on the primary site, size, and invasiveness, then standard treatment would be endoscopic or surgical management.1 Advanced disease most often requires systemic treatments. However, liver-directed therapy has an important role for liver-dominant NETs. Surgery may also be considered, even in the setting of stage IV disease, for either debulking of metastases, most commonly in the liver, or resection of the primary tumor, most commonly for small bowel NETs.
Next, we discuss the major trials for metastatic, well-differentiated NETs.
SSAs: Octreotide (PROMID) and Lanreotide (CLARINET)
Jonathan Strosberg, MD:
First-line treatments for well-differentiated GEP-NETs typically consist of SSAs because they have low risks and are very well tolerated. Although they do not shrink tumors significantly, they can stabilize disease, sometimes for many years.
The first randomized phase III study of an SSA in treating NETs was PROMID, in which patients with low-grade small bowel NETs were randomized to receive long-acting octreotide or placebo.21 The left figure shows significantly longer time to tumor progression with SSA treatment, with an HR of 0.34.
The next study was the CLARINET study, in which patients with pNETs with a Ki-67 index of ≤10% and SSTR expression were randomized to receive lanreotide or placebo.22 Again, this trial showed significant prolongation in progression-free survival (PFS), with an HR of 0.47.
These drugs not only palliate hormonal syndromes associated with NETs, such as reducing incidents of flushing and diarrhea, they also delay tumor growth.23
Everolimus: RADIANT Studies
Jonathan Strosberg, MD:
Another important medication for NETs is everolimus, an mTOR inhibitor that has been evaluated in several randomized clinical trials after encouraging early-phase trial data.
The phase III RADIANT-2 trial randomized patients with advanced GEP-NETs and a history of carcinoid syndrome to receive everolimus plus octreotide vs placebo plus octreotide.24 Although this study showed a trend toward everolimus delaying PFS, the delay was relatively modest and was not statistically significant.
However, the RADIANT-3 study, which evaluated everolimus vs placebo in advanced pNETs, was a positive study showing longer median PFS, 11 months vs 4.6 months (HR: 0.35).25 These results led to FDA approval of everolimus for the treatment of patients with progressive NETs of pancreatic origin with unresectable, locally advanced or metastatic disease.
The final study with everolimus, RADIANT-4, enrolled patients with nonfunctional NETs of gastrointestinal (GI) or lung origin.26 The bottom right figure shows that this population had much more aggressive disease than the RADIANT-2 population. Median PFS was 3.9 months for the placebo group compared with 11.0 months for the everolimus group (HR: 0.48). This positive study led to the FDA approval of everolimus for adults with progressive, well-differentiated, nonfunctional NETs of GI or lung origin. More specifically, this is primarily relevant to lung NETs, gastroduodenal NETs, and rectal NETs, rather than small bowel (midgut) NETs, which are often functional.
Phase III Study of Sunitinib vs Placebo in pNETs
Jonathan Strosberg, MD:
Sunitinib is another targeted drug that is FDA approved for patients with pNETs. It is a tyrosine kinase inhibitor that targets receptors VEGF1 and VEGF2 as well as PDGFRβ. In a phase III study of sunitinib vs placebo, the median PFS was better with sunitinib, 11.4 months vs 5.5 months with placebo, which is a result similar to that seen in the RADIANT-3 trial of everolimus.27
ECOG 2211: Capecitabine/Temozolomide vs Temozolomide in pNETs
Jonathan Strosberg, MD:
Cytotoxic chemotherapy is relatively active in pNETs (less active in GI and lung NETs). In the past, the most commonly used drug was streptozocin, an intravenous alkylating agent.
However, in recent decades, temozolomide has emerged as an oral alternative with generally better tolerability. The ECOG 2211 study looked at the combination of capecitabine and temozolomide vs temozolomide monotherapy for advanced progressive pNETs.28 This figure shows significantly better median PFS with capecitabine/temozolomide, 22.7 months vs 14.4 months with temozolomide.
This drug combination is now considered a standard therapy for advanced pNETs and most appropriate for patients with aggressive disease, who have the highest rate of benefit from cytotoxic chemotherapy.1 There are other cytotoxic regimens, such as FOLFOX (folinic acid, fluorouracil, and oxaliplatin), that can also be active in aggressive pNETs, but that are associated with more risks and adverse events (AEs).29
NETTER-1: 177Lu-Dotatate vs High-Dose Octreotide
Jonathan Strosberg, MD:
An important treatment modality for advanced NETs is PRRT, which uses a radiolabeled SSA, linked to radionuclides such as 177 Lutetium (177Lu), to target SSTR-expressing tumors.30 And as mentioned earlier , most well-differentiated NETs strongly express SSTRs.
After many years of single-arm studies and retrospective series, the phase III NETTER-1 study was developed to test whether 177Lu-dotatate plus best supportive care including octreotide long-acting repeatable (LAR) could provide significantly longer PFS in patients with advanced, progressive small bowel NETs.7 These patients had disease that had previously progressed during SSA octreotide LAR treatment and they were randomized to receive 177Lu-dotatate plus best supportive care, including octreotide LAR, or high-dose octreotide LAR. NETTER-1 showed significantly greater PFS with an HR of 0.21 in the 177Lu-dotatate group, and the median PFS was not reached at the time of primary endpoint analysis. The final PFS analysis determined the median PFS was 25.0 months and 8.5 months for the 177Lu-dotatate group and the high-dose octreotide LAR group, respectively.31 The objective response rate (ORR) with 177Lu-dotatate was 18%.30
Treatment Sequencing for Metastatic Small Bowel NET
Jonathan Strosberg, MD:
Now we discuss how to sequence these therapies in patients with metastatic small bowel midgut NETs. First, providers typically consider surgical cytoreduction and resection of primary small bowel tumor to prevent obstruction, bleeding, and other complications.32 However, patients who are not considered candidates for surgery are commonly treated with an SSA as a first-line systemic therapy. Either octreotide or lanreotide is used in this case to inhibit tumor growth and control hormonal syndromes. We typically begin treating with the SSAs octreotide and lanreotide because they are well tolerated compared with most oncologic drugs and they strongly inhibit tumor growth. Although response rates are low, they have been shown to significantly increase PFS based on data from the PROMID and CLARINET studies.21,22
If a patient’s disease progresses, the next most common systemic therapy is 177Lu-dotatate. However, for patients with liver-dominant disease, we commonly consider hepatic artery embolization; this includes various modalities including chemoembolization, bland embolization, and radioembolization. There are certain contraindications and risks with PRRT. Patients with heavy-burden peritoneal carcinomatosis have an increased risk of bowel obstruction with 177Lu-dotatate, so this is often considered a contraindication.33 We also know from several studies that this agent is associated with a 2% to 3% risk of developing treatment-related myelodysplastic syndromes or leukemia.34 There also may be a further increased risk after prior cytotoxic chemotherapy.34,35
Less commonly, we consider everolimus as a second-line therapy for patients who do not meet criteria for 177Lu-dotatate or liver embolization, or as a later line of treatment. However, everolimus can have significant and chronic effects on quality of life, which are of particular concern in older or frail patients, patients with underlying lung disease, and patients with underlying diabetes.36
We also note that cytotoxic chemotherapy is generally ineffective for small bowel NETs.37 CAPTEM (capecitabine and temozolomide) or FOLFOX may be considered a last-resort treatment in rare cases of particularly high-grade or aggressive disease.
Comparing Second-line Systemic Therapies
Jonathan Strosberg, MD:
When considering second-line systemic therapies, 177Lu-dotatate is usually the preferred option rather than everolimus, particularly for small bowel NETs. Although no studies have directly compared these 2 drugs, 177Lu-dotatate likely has a higher degree of activity and is associated with a more favorable HR for PFS than everolimus.1,7
Hepatic Arterial Embolization for GEP-NETs
Jonathan Strosberg, MD:
There have been numerous retrospective studies evaluating hepatic arterial embolization for liver-dominant disease that show relatively high response rates of approximately 50%, long PFS of approximately 1.5 years, and high rates of symptomatic responses, particularly for patients with carcinoid syndrome and other hormonal syndromes.38,39 However, there have been no good-quality prospective studies to strongly support the choice of embolization for treatment. The RETNET trial results are currently being analyzed and may offer data comparing bland embolization vs chemoembolization in patients with liver-dominant NETs.40
Bland embolization, chemoembolization, and radioembolization tend to have few short-term AEs, most commonly nausea, pain, and fever. With bilobar radioembolization, there is a risk of chronic treatment-induced liver damage, but it may be particularly effective for patients with fast-growing tumors.41
Systemic Treatment Algorithm for pNETs
Thor R. Halfdanarson, MD:
pNETs have a larger range of systemic treatment options, which makes it more difficult to determine how to best sequence therapies. In most cases, SSAs are the first-line treatment for SSTR-positive disease, as observed on SSTR path imaging, and in the absence of symptoms or large tumor bulk.32 Choice of subsequent treatment depends on the presentation of the disease. The evidence and reasoning for second-line treatment are discussed in the next section.
If there is large tumor bulk and symptoms, then chemotherapy with capecitabine/temozolomide and now PRRT (since the NETTER-2 trial data became available, described below) can be considered earlier in the sequence to control tumor growth. Other options include everolimus and sunitinib with cabozantinib as a possible later-line therapy.
For the less common SSTR-negative tumors, which may represent as many as 10% to 20% of cases, SSAs are inapplicable.42 We then would consider capecitabine/temozolomide (CAPTEM), everolimus, and sunitinib, CAPTEM being particularly appropriate for symptomatic disease because everolimus and sunitinib are unlikely to result in meaningful tumor regression and symptom resolution. There are other chemotherapies that can be used, but typically in later lines of therapy, such as FOLFOX and FOLFIRI (folinic acid, 5-FU, and irinotecan). Streptozocin-based therapy was commonly used in the past but streptozocin has been in short supply recently.
Randomized pNET Studies
Thor R. Halfdanarson, MD:
Here we briefly highlight several randomized pNET studies, some which have already been mentioned. The RADIANT 3 study compared everolimus vs placebo, showing that, although response rates were low, everolimus provided better median PFS than placebo (11.0 vs 4.6 months, respectively).25,43 The Su011248 study similarly showed a benefit with sunitinib vs placebo, with a median PFS of 12.6 months vs 5.8 months.27,44
More recently, the randomized phase II ECOG 2211 trial compared capecitabine/temozolomide vs temozolomide alone, showing superiority of the combination (median PFS: 22.7 vs 14.4 months) and higher response rates than in the other trials, but not much difference in the response rates between the 2 arms.28 The small OCLURANDOM study found that 177Lu-dotatate provided substantially longer median PFS than sunitinib (20.7 vs 11.0 months, respectively).45 Finally, the SEQTOR study compared 5-fluorouracil/streptozocin vs everolimus and allowed patients to switch to the other arm after disease progression.46 Of interest, it showed little difference in PFS and AEs between the 2 arms, although response rates were higher with the chemotherapy.
When choosing second-line therapy for SSTR-positive pNETs, we can consider that chemotherapy regimens, particularly capecitabine/temozolomide and the radionuclide SSTR agonist 177Lu-dotatate, seem to provide higher response rates and likely longer median PFS than everolimus and sunitinib. There are few studies directly comparing these drugs, but capecitabine/temozolomide and 177Lu-dotatate also appear to have fewer daily AEs. For most patients, these agents should be considered earlier in sequencing than everolimus and sunitinib, but not necessarily for everyone. Another consideration is patient comorbidities, which may influence therapy choice. For example, in patients with uncontrolled diabetes, everolimus may be challenging, and sunitinib may be difficult to administer in patients with poorly controlled hypertension.
Some emerging data may show that the cytotoxic options such as capecitabine/temozolomide induce a higher mutational burden on NETs and induce disease transformation to more aggressive and higher-grade disease.47,48 There also is concern about long-term myelodysplastic syndromes and acute leukemia with 177Lu-dotatate treatment.34,49,50 Therefore, in some patients, everolimus and sunitinib may be the most appropriate options after SSAs even though they may be less effective. More studies directly comparing these options are needed to better inform treatment sequencing.
Treatment Sequencing for Non-Midgut GI NET
Thor R. Halfdanarson, MD:
In treatment sequencing for non-midgut GI NET, including rare subtypes of gastroduodenal or metastatic rectal GI NETs, we typically start with SSAs if the tumors are SSTR positive.32
In the case of progression after an SSA, there is relatively strong evidence from the RADIANT-4 study supporting everolimus as a second-line treatment.26 The trial randomized patients with nonfunctional NETs to everolimus or placebo. 177Lu-dotatate can also be considered, but most data for this treatment in this disease type are from nonrandomized studies.35 Last, liver embolization can be considered for patients with liver-dominant disease.
There is relatively weak evidence evaluating cytotoxic chemotherapy in non-midgut GI NETs compared with pNETs, so regimens such as capecitabine/temozolomide do not have a particularly prominent role in nonpancreatic NETs.
NETTER-2: First-line 177Lu-Dotatate Plus Octreotide LAR vs High-Dose Octreotide LAR
Thor R. Halfdanarson, MD:
Now we discuss the recent and ongoing NETTER-2 trial, a prospective, randomized phase III trial of 177Lu-dotatate with octreotide LAR vs high-dose octreotide LAR for patients with higher-grade SSTR-positive GEP-NETs.51 Specifically, these tumors were grade 2 with a Ki-67 index of ≥10% or well-differentiated grade 3 NETs with Ki-67 ≤55%, and the patients could have had no prior therapy; 226 patients were randomized to receive 4 treatments of 177Lu-dotatate PRRT and octreotide LAR 30 mg every 8 weeks or high-dose octreotide LAR 60 mg every 4 weeks.
The primary endpoint was PFS with secondary endpoints of ORR and quality of life. The primary analysis has been completed, but evaluation of some of the secondary endpoints is ongoing, the most important of which is overall survival (OS).
NETTER-2: PFS
Thor R. Halfdanarson, MD:
Here we see that 177Lu-dotatate had a superior median PFS of 22.8 months vs 8.5 months with high-dose octreotide (HR: 0.276; P <.0001).51
NETTER-2: Objective Responses
Thor R. Halfdanarson, MD:
ORR was also higher in the PRRT arm (53.0% vs 9.3%), with several complete responses (5.0%).51 The partial response rate was 37.7% in the PRRT arm vs 9.3% in the high-dose octreotide arm. This has confirmed a suspicion that the higher-grade SSTR-positive well-differentiated NETs may be more likely to respond to PRRT than other tumors.
Does NETTER-2 Change the Treatment Paradigm for High-Grade NET?
Thor R. Halfdanarson, MD:
There is a strong rationale for the early use of PRRT, but there are some caveats. PRRT carries more and different risks than SSA therapy and we do not yet have the OS data from the NETTER-2 trial.
There is also a question of whether octreotide is a legitimate comparator to 177Lu-dotatate. It is worth keeping in mind that other potential treatments are more active than SSAs in this population, especially capecitabine and temozolomide in pNETs. The ongoing prospective, randomized phase III COMPOSE trial may provide further information on the comparison of PRRT vs chemotherapy in patients with pNETs. First-line 177Lu-dotatate should be considered for well-differentiated GEP-NETs with the characteristics of the patients included in the NETTER-2 study, but ultimately the choice of treatment should be a joint decision between the patient and the treating healthcare professional.
NETTER-P: Pediatric Study of 177Lu-Dotatate With GEP-NETs or Pheochromocytoma/Paraganglioma
Thor R. Halfdanarson, MD:
177Lu-dotatate was also studied in the pediatric NETTER-P trial. This ongoing, international, multicenter, open-label, single-arm study included adolescent patients with locally advanced/inoperable or metastatic SSTR-positive GEP-NETs or pheochromocytoma/paraganglioma. Efficacy outcomes were also extrapolated from NETTER-1. Major safety measures of NETTER-P included absorbed radiation doses in target organs as well as AEs.52,57
Based on the NETTER-1, NETTER-2, and NETTER-P trials, 177Lu-dotatate is now indicated for the treatment of adult and pediatric patients aged 12 years or older with SSTR-positive GEP-NETs, including foregut, midgut, and hindgut NETs.58
Alliance A021602: CABINET Trial
Thor R. Halfdanarson, MD:
Now we discuss the phase III CABINET trial of cabozantinib that may affect NET treatment. Cabozantinib is a tyrosine kinase inhibitor targeting the receptors VEGF, MET, AXL, and RET, and was shown to have encouraging activity in a small, recently reported phase II trial.54 Cabozantinib is not currently approved by the FDA for the treatment of NETs.
The CABINET trial is a double-blinded, randomized phase III study in patients with previously treated well-differentiated NETs of pancreatic or extrapancreatic origin, including lung NETs.55 The enrollment criteria included patients with disease progression after concurrent treatment with SSAs and at least 1 additional FDA-approved line of therapy. Patients were stratified between those with pNETs and those with extrapancreatic NETs and were randomized to receive cabozantinib 60 mg/day or placebo. The primary endpoint was PFS; the key secondary endpoints were OS, ORR, safety, and tolerability. The study was later unblinded because better PFS was seen with cabozantinib and allowed crossover from placebo to cabozantinib after disease progression.
CABINET: PFS and OS in Extrapancreatic NET
Thor R. Halfdanarson, MD:
Here we see that the extrapancreatic cohort had longer median PFS with cabozantinib compared with placebo (8.4 vs 3.9 months , respectively; HR: 0.38).55 There was no significant difference in median OS, with 21.9 months in the cabozantinib treatment group vs 19.7 months with placebo, but it is important to remember that most patients in the placebo group were able to cross over.
CABINET: PFS and OS in pNETs
Thor R. Halfdanarson, MD:
Moving on to the pNET cohort, we see that cabozantinib provided longer median PFS than placebo (13.8 vs 4.4 months, respectively).25 The efficacy is more striking than for the extrapancreatic cohort. Again, there was no statistically significant difference in OS, but this likely is because of, at least in part, the crossover.
CABINET: Confirmed Objective Responses (Blinded Independent Central Review)
Thor R. Halfdanarson, MD:
Here we see the ORR in the extrapancreatic cohort was 5% with cabozantinib vs 0% with placebo, as expected.55 The ORR for the pNETs was 19%, nearly 4 times that of the extrapancreatic NETs. There were more frequent responses in pancreatic disease, but the figures here show that there were more patients in each cohort with some degree of regression that did not meet the criteria for formal responses.
CABINET: Adverse Events
Thor R. Halfdanarson, MD:
The AEs with cabozantinib were as expected from its use in treating other tumor types and there were no new safety signals.55 In the extrapancreatic NET cohort, 62% of patients had grade 3-5 AEs, most commonly hypertension (a class effect of tyrosine kinase inhibitors), fatigue, and diarrhea. The pancreatic cohort had similar AEs and frequencies.
Exposure
Thor R. Halfdanarson, MD:
Dose reductions were frequent, with 66% of patients randomized to cabozantinib and 10% of those randomized to placebo having doses reduced, and the median exposure to cabozantinib was 38.4 mg/day.55 Thirty-one percent of patients treated with cabozantinib discontinued the drug because of AEs.
Integrating Cabozantinib Into Treatment Plans
Thor R. Halfdanarson, MD:
We have yet to see whether cabozantinib will receive FDA approval for NETs. Of importance, it provided longer PFS treating both extrapancreatic and pNETs in previously treated patients.55 The degree of the increase was more prominent in treating pNETs, which was not surprising. The impact on OS is difficult to gauge because the CABINET study was not powered to evaluate OS differences and there was crossover.
Unlike sunitinib and everolimus, cabozantinib now has proven efficacy in extrapancreatic NETs regardless of functional status. We also have not had good data on kinase inhibitors in treating small bowel and pulmonary NETs. Phase III sunitinib data were only obtained for pNETs.27 Everolimus had positive phase III study results in treating pNETs in the RADIANT-3 trial and in treating nonfunctional extrapancreatic NETs of the small bowel and lung in RADIANT-4.25,26 However, everolimus did not show statistically significant superiority in patients with small bowel NETs and carcinoid syndrome in RADIANT-2.24
Although cabozantinib has shown PFS and HRs superior to those seen with sunitinib or everolimus, we should take into account that it is difficult to compare results across different trials. The toxicities should also be considered, particularly in older or frail patients, because most patients were unable to tolerate the 60-mg dose of cabozantinib. We might consider starting at a lower dose, especially in those patients at a higher risk for AEs. Ultimately we need a clinical trial to establish whether starting with a lower dose of cabozantinib and stepping up is a better strategy than starting with a full dose and titrating down.
Conclusions
Thor R. Halfdanarson, MD:
The optimal sequencing of systemic and liver-directed therapy is complicated. We have not had trials that compare many of these different treatments, and many of the related clinical trials lack the power to assess the impact on OS. Also, some of the trials had a crossover design, and in some instances patients were able to get the treatment later, even if outside of a clinical trial.
The NETTER-2 trial supports 177Lu-dotatate as a new first-line therapy option for grade 2 NETs and well-differentiated grade 3 NETs. Cabozantinib showed promising PFS compared with placebo in patients with previously treated progressive GEP-NETs, some of whom had received more than 2 lines of prior therapy. This is definitely a patient population in need of better therapy, but as of December 2024, cabozantinib is not yet approved for this indication.
Last, ideal and optimal management of patients with NETs should be provided within a multidisciplinary team with experience in treating these patients, and the patients need to be fully engaged at every step of treatment. Present cases at tumor boards when needed and pay attention to the goals of therapy, which typically are not to cure but which could be to prolong PFS and OS. Keep in mind that not all of the interventions that prolong PFS also reduce symptoms, so be mindful of quality of life, both in terms of symptoms from the actual malignancy and as a result of the AEs of the treatments you administer.
Why See an NEN Expert?
Thor R. Halfdanarson, MD:
Why should patients see an NEN expert? Patient advocacy groups and networks are extremely helpful.
An NEN expert may have access to services not available to community oncologists, such as expert pathologists, advanced imaging, specialized surgery, and other medical specialties, as well as a tumor board.
Not every patient with an NEN needs to see an NET expert at every visit. Collaboration with community oncologists is preferred. Intermittent visits to an NEN center, as often as needed, may be all that is required. A return to an NEN center is advisable at major forks in the road such as cancer progression and a proposed change in therapy.
The Role of the NEN Expert Center
Thor R. Halfdanarson, MD:
NEN Expert Centers can help confirm that an NEN diagnosis is correct and that the treatment plan is optimal. Up to 25% of patients with NENs have a change in the diagnosis and/or the plan after receiving an expert opinion.
An NEN expert may have access to services not available to community oncologists, such as expert pathologists, advanced imaging, specialized surgery, and other medical specialties, as well as a tumor board.
Why Consult a Tumor Board?
Thor R. Halfdanarson, MD:
Tumor boards are essential because having multiple specialties involved early may change the initial approach. There are sometimes several reasonable approaches, which can lead to a discussion of the pros and cons of each approach.
A tumor board is especially useful when surgery or other regional/local therapy (eg, liver-directed therapy) is being considered. Tumor boards may improve the outcomes of therapy and may result in fewer visits for the patient.
Back
