STELLAR-311: a randomized phase II/III study of zanzalintinib versus everolimus in advanced neuroendocrine tumors

ABSTRACT

Patients with advanced neuroendocrine tumors (NETs) experience poor outcomes despite the use of available therapies, including molecularly targeted agents. Zanzalintinib is a novel, small-molecule inhibitor of multiple kinases, including the TAM kinases (TYRO3, AXL, MER), MET, and vascular endothelial growth factor receptors, which play important roles in oncogenic processes. Zanzalintinib has shown encouraging antitumor activity and a consistent safety profile across multiple solid tumor types in clinical trials. STELLAR-311 is a global, randomized, open-label, phase II/III trial evaluating the efficacy and safety of zanzalintinib versus everolimus in patients with advanced NETs. Adults aged ≥18 years with locally advanced/unresectable or metastatic, well-differentiated grade 1–3 NETs of pancreatic (pNET) or extra-pancreatic (epNET; gastrointestinal tract, lung, thymus, or other) origin will be enrolled. Patients with pNETs or nonfunctional epNETs can be untreated or have received up to one prior line of therapy (excluding somatostatin analogs); patients with functional epNETs must have received one prior line of therapy. The primary endpoint is progression-free survival (PFS), by blinded independent central review (BICR). Secondary endpoints include overall survival; investigator-assessed PFS; BICR- and investigator-assessed objective response rate, duration of response, and disease control rate; safety; and quality of life.

Clinical trial registration www.clinicaltrials.gov identifier is NCT06943755.

PLAIN LANGUAGE SUMMARY

Neuroendocrine tumors (NETs) are rare cancers. They start from cells in the body that make hormones and other chemical signals. NETs can develop in different parts of the body, and most commonly start in the pancreas, lungs, and bowel. When new blood vessels grow in a process called angiogenesis, the tumor is supplied with oxygen and nutrients and can spread to other parts of the body. People with advanced disease have tumors that have spread or have grown to a point where they cannot be removed with surgery. Zanzalintinib is an investigational medicine taken by mouth that may reduce or stop the growth and spread of tumors. It works by blocking certain proteins that cancer cells need to develop, grow, and spread. STELLAR-311 is a study that will compare zanzalintinib with everolimus, a standard treatment for people with advanced NETs. The study will include adults with advanced NETs that started in the pancreas, lungs, stomach, bowel, or other organs. Participants will have received either no treatment or one treatment for advanced NETs, depending on where in the body the NET started to grow. Researchers will measure how long people can live without their cancer getting worse when treated with zanzalintinib compared with everolimus. They will also monitor for side effects to understand the safety profile of zanzalintinib in people with NETs. Approximately 440 participants from countries across the world will be included in the study.

1. Introduction

1.1. Disease overview and management

Neuroendocrine tumors (NETs) are rare cancers that are biologically and clinically heterogeneous [1,2]. The most common tumor sites of origin are the gastrointestinal tract (GI), lungs, and pancreas, although they can develop from neuroendocrine cells in any organ [1,3]. In recent decades, the incidence of NETs has increased [4], particularly for localized and low-grade NETs [1,4–8]; this has been largely driven by advances in detection, diagnostics, and classification practices. NETs are generally categorized as pancreatic (pNET) or extra-pancreatic (epNET) in origin, with approximately 10–40% being classified as functional (i.e., as producing and secreting hormones or other chemical signals that can result in widespread symptoms) [1,7,9,10]. The location of NETs impacts prognosis, with worse survival outcomes reported in patients with distant disease from primary lung and pancreatic NETs compared with small intestine NETs [8]. Complications related to tumor location include recurrent infections, hemoptysis, pneumonia, and dysphagia in lung NETs, as well as abdominal pain, obstruction, GI bleeding, and perforation in GI NETs [3]. Tumor-related symptoms often persist for prolonged periods before NETs are diagnosed and lead to considerable morbidity, resulting in impaired health-related quality of life (HRQoL) compared with the general population [3,11,12]. Symptoms have also been associated with increased mortality risk in patients with NETs [13,14].

Angiogenesis plays a key role in NET pathogenesis and is mediated by proangiogenic factors, including vascular endothelial growth factor (VEGF), which has been shown to promote the progression and metastasis of NETs [15]. VEGF is expressed in both pNETs and epNETs, with strong VEGF expression associated with worse survival [15]. Increased expression of VEGF receptors 2 and 3 has also been shown in GI NETs and pNETs [16–18]. These findings were validated by a meta-analysis of phase II and III trials, which demonstrated antitumor activity with anti-VEGF tyrosine‑kinase inhibitors (TKIs) in patients with NETs [19]. Preclinical data have shown that MET activation is also involved in stimulating NET growth and mediates resistance to anti-VEGF therapy, suggesting that MET inhibition may be a rational target in this disease [20–22]. Indeed, MET tumor expression has been shown to correlate with reduced overall survival (OS) in patients with pNET [20].

Management of NETs requires a specialized, multidisciplinary approach due to the complex, diverse nature of these tumors [3,4,23]. Treatment decisions are based on disease stage, disease burden, tumor characteristics (including primary tumor site, differentiation, grade, functional status related to hormone secretion, and somatostatin receptor expression), patient factors (including preferences, comorbidities, and performance status), and previous treatments [2,3,10]. These factors contribute to the biological and clinical heterogeneity of NETs and may influence treatment outcomes. Current first-line standard-of-care treatment for most patients with advanced NETs consists of somatostatin analogs (SSAs; e.g., lanreotide and octreotide) for symptom management and tumor control. Following progression on SSAs, treatment approaches include molecularly targeted therapies, radiolabeled SSAs, and chemotherapy. Options are distinct for patients with pNETs versus epNETs. For patients with progressive pNETs, treatments include the mammalian target of rapamycin (mTOR) inhibitor everolimus and the multitargeted TKIs sunitinib and cabozantinib, as well as the peptide receptor radionuclide therapy (PRRT) ¹⁷⁷Lu-DOTATATE for somatostatin receptor-positive NETs. Chemotherapy (e.g., capecitabine or temozolomide) can also be considered for pNETs. For patients with progressive epNETs, treatment options include cabozantinib and everolimus (nonfunctional GI and lung NETs), as well as ¹⁷⁷Lu-DOTATATE for somatostatin receptor-positive GI NETs.

The combination of earlier diagnosis and introduction of new therapies has improved the prognosis for patients with advanced NETs [6,7]. However, although several treatment options are currently available, they are not all suitable for every patient and can be associated with tolerability issues [10]. Furthermore, despite these advances, patients with advanced NETs continue to have a poor prognosis and most have incurable disease; hence, an unmet need for novel therapies remains [2,5,24,25]. The therapeutic landscape of NETs is evolving with the development of several multitargeted TKIs beyond sunitinib and cabozantinib [26]. The ongoing clinical investigation of agents targeting VEGF and related pathways reflects continued efforts to improve efficacy and broaden treatment options for patients with advanced NETs.

1.2. Mechanism of action of zanzalintinib

Zanzalintinib is a novel, oral, small-molecule inhibitor of multiple kinases, including the TAM kinases (TYRO3, AXL, and MER), MET, and VEGF receptors, which play important roles in oncogenic processes such as tumor cell proliferation, metastasis, angiogenesis, drug resistance, and evasion of antitumor immunity (Figure 1) [27–29]. Preclinical data have shown that zanzalintinib treatment leads to reductions in tumor-cell proliferation and angiogenesis, with dose-dependent tumor growth inhibition witnessed in xenograft models of several solid tumors [27]. Zanzalintinib has demonstrated a kinase-inhibition profile similar to that of cabozantinib; however, zanzalintinib has an improved pharmacokinetic profile, with a shorter half-life and preferential in vivo tumor distribution [27,30–32].

Figure 1.

Zanzalintinib mechanism of action [27–29].

Abbreviations: APC, antigen-presenting cell; CD8, CD8+ T cell; GAS6, growth arrest-specific 6; HGF, hepatocyte growth factor; MDSC, myeloid-derived suppressor cell; NK, natural killer cell; PD-1, programmed cell death protein 1; PD-L1, programmed cell death-ligand 1; TAM, the TAM kinases (TYRO3, AXL, and MER); VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor receptor.

1.3. Clinical data with zanzalintinib

Zanzalintinib has demonstrated antitumor activity and a consistent safety profile in clinical trials across multiple solid tumor types when administered as a single agent or in combination with immune checkpoint inhibitors [28,31,33,34], supporting further evaluation in additional tumor settings. In the phase I STELLAR-001 trial, single-agent zanzalintinib demonstrated clinical activity in patients with advanced clear cell renal cell carcinoma or metastatic colorectal cancer and was generally well tolerated [28,31,33]. The most common adverse events included hypertension, nausea, and diarrhea, which is consistent with the known safety profile of VEGFR-TKIs [19,28,31,33,35]. Significant changes in biomarkers associated with angiogenesis, tumor growth, and metastasis were also observed with zanzalintinib treatment [29]. In the phase III STELLAR-303 trial, combination treatment with zanzalintinib plus atezolizumab demonstrated a statistically significant improvement in OS versus regorafenib in patients with previously treated metastatic colorectal cancer and was associated with a manageable safety profile [34].

As patients with advanced NETs are rarely cured with currently available systemic therapies, there remains a significant unmet medical need for new treatments. Additionally, because there is a lack of clinical data comparing targeted therapies that inhibit the VEGF or mTOR signaling pathways, the relative benefits of these therapeutic strategies are not known. Based on the unmet need for additional treatment options for patients with advanced NETs, the lack of comparative data between VEGF- and mTOR–targeted therapies, the biological rationale for targeting pathways inhibited by zanzalintinib in NETs, and encouraging clinical experience with zanzalintinib in other solid tumors, the STELLAR-311 trial was initiated. The study allows enrollment of a broad population of patients with NETs and inclusion of stratification factors that align with recommendations in the 2021 National Cancer Institute consensus report on NET clinical trial design [36].

2. Methods

2.1. Study design and treatment

STELLAR-311 (NCT06943755) is a global, open-label, randomized, phase II/III trial designed to evaluate the efficacy and safety of zanzalintinib versus everolimus in patients with locally advanced or metastatic NETs. Everolimus was approved based on the phase III RADIANT-3 and RADIANT-4 studies [37,38] and was selected as the comparator in STELLAR-311 because, at the time of study design, it was widely accepted and used as a standard treatment option for patients with advanced pNET and nonfunctional epNET. Patients will be randomized 1:1 to receive either zanzalintinib or everolimus, both administered orally once daily (Figure 2). Randomization will be performed centrally via an Interactive Response Technology system using a permuted block design, with primary tumor site (pNET vs epNET), prior PRRT (yes vs no), and tumor grade (grade 1 or typical carcinoid vs grade 2 or 3 atypical carcinoid) as stratification factors. The trial is open label and thus unblinded. Patients will be treated until lack of clinical benefit, unacceptable toxicity, or other protocol-defined reasons for discontinuation. Crossover from everolimus to zanzalintinib will be allowed upon disease progression as confirmed by blinded independent central review (BICR). Following study treatment discontinuation, safety will be evaluated after 30 days, and patients will be contacted every 12 weeks to assess survival and subsequent anticancer therapy.

Figure 2.

STELLAR-311 study design and endpoints.

^aPer RECIST v1.1.

^bChange from baseline in patient-reported Global Health Status and disease-related symptoms.

Abbreviations: BICR, blinded independent central review; DCR, disease control rate; DOR, duration of response; EORTC, European Organization for Research and Treatment of Cancer; ep, extra-pancreatic; NETs, neuroendocrine tumors; ORR, objective response rate; OS, overall survival; p, pancreatic; PFS, progression-free survival; PRRT, peptide receptor radionuclide therapy; QLQ-C30, Quality of Life Questionnaire – Core 30; QLQ-GI.NET21, Gastrointestinal Neuroendocrine Tumor-Specific, 21-Item Quality of Life Questionnaire; R, randomization; RECIST v1.1, Response Evaluation Criteria in Solid Tumors version 1.1.

2.2. Eligibility criteria

Patients must be aged ≥18 years with histologically confirmed, locally advanced/unresectable or metastatic, well-differentiated grade 1–3 pNET or epNET (Table 1). Eligible patients are required to have measurable disease and documented radiographic disease progression ≤12 months before randomization (both per Response Evaluation Criteria in Solid Tumors version 1.1 [RECIST v1.1] as assessed by the investigator). Prior lines of therapy are allowed based on NET primary tumor site and functional status: patients with pNET or nonfunctional epNET may be untreated or have received up to one prior line of systemic anticancer treatment for NET; patients with functional epNET must have received one (but no more than one) prior line of systemic anticancer therapy (PRRT or chemotherapy) for NET. Prior SSA treatment is allowed and does not count toward the requirement of a prior line of systemic anticancer treatment. Concomitant SSA treatment is permitted for patients with functional NETs provided the dose has remained stable for ≥8 weeks prior to randomization. Archival tumor tissue is required, if available; alternatively, a fresh biopsy may be provided if it can be safely and feasibly obtained.

Table 1.

Key eligibility criteria^.a

Inclusion criteria
Histologically confirmed, locally advanced/unresectable or metastatic, well-differentiated grade 1, 2, or 3 NETs of pancreatic origin or extra-pancreatic origin
Prior lines of systemic anticancer therapy are allowed based on the site of NET and functional statusb Untreated or up to one prior line for pNET and nonfunctional epNET One, and no more than one (PRRT or chemotherapy), required for functional epNET
Documented radiographic disease progression per RECIST v1.1 as assessed by the investigator based on imaging assessments (CT or MRI) within 12 months before randomization
Measurable disease according to RECIST v1.1 as determined by the investigator
Functional or nonfunctional tumors
Somatostatin receptor-positive or -negative disease
Archival tumor tissue is required, if available. If archival tumor tissue is not available, a fresh biopsy may be submitted if it can be safely and feasibly obtained. Every attempt should be made to provide tumor tissue
Age ≥18 years
Exclusion criteria
Histologically confirmed neuroendocrine carcinomas (including small cell lung cancer), medullary thyroid cancer, pheochromocytoma, paraganglioma, Merkel cell carcinoma, and mixed neuroendocrine non-neuroendocrine neoplasm
Prior treatment with a VEGFR-targeting tyrosine kinase inhibitor or an mTOR inhibitor
Systemic chemotherapy and any liver-directed or other ablative therapy within 4 weeks before randomization
Systemic radionuclide therapy within 6 weeks before randomization
Radiation therapy for bone metastases within 2 weeks before randomization; any other radiation therapy, except as indicated above, within 4 weeks before randomization

^aOther protocol-defined inclusion/exclusion criteria may apply.

^bPrior treatment with an SSA is allowed and does not count toward the requirement of prior line of systemic anticancer therapy. Continuation of treatment with an SSA while on study treatment is only allowed for patients with functional NETs who have been on a stable dose of SSA for ≥8 weeks before randomization.

Abbreviations: CT, computed tomography; ep, extra-pancreatic; MRI, magnetic resonance imaging; mTOR, mammalian target of rapamycin; NET, neuroendocrine tumor; p, pancreatic; PRRT, peptide receptor radionuclide therapy; RECIST v1.1, Response Evaluation Criteria in Solid Tumors version 1.1; SSA, somatostatin analog; TKI, tyrosine kinase inhibitor; VEGFR, vascular endothelial growth factor receptor.

Patients are excluded if they have histologically confirmed, poorly differentiated neuroendocrine carcinomas (including small cell lung cancer), medullary thyroid cancer, pheochromocytoma, paraganglioma, Merkel cell carcinoma, or mixed neuroendocrine non-neuroendocrine neoplasm. Treatment-based exclusion criteria include prior treatment with VEGFR-targeting TKIs or mTOR inhibitors, systemic chemotherapy or any liver-directed or other ablative therapy ≤4 weeks before randomization, systemic radionuclide therapy ≤6 weeks before randomization, and radiation therapy for bone metastases ≤2 weeks before randomization – or any other radiation therapy ≤4 weeks before randomization.

2.3. Endpoints

The primary endpoint is progression-free survival (PFS) per RECIST v1.1 as assessed by BICR. PFS is defined as the time from randomization to the earlier of either radiographic progressive disease per RECIST v1.1 or death due to any cause. Secondary efficacy endpoints include PFS per RECIST v1.1 as assessed by the investigator, OS, overall response rate (ORR), duration of response (DOR), disease control rate (DCR), and HRQoL. Evaluation of ORR, DOR, and DCR will be per RECIST v1.1 as assessed by BICR and the investigator. OS is defined as the time from randomization to death from any cause. ORR is defined as the proportion of patients with best overall response of complete response or partial response (confirmed at a subsequent imaging assessment ≥28 days later). DOR is the time from the first documentation of objective response (confirmed at a subsequent visit ≥28 days later) to the earlier of disease progression or death from any cause. DCR is the rate of patients with a best overall response of confirmed complete response, confirmed partial response, or stable disease (as assessed by BICR and by the investigator). Safety and tolerability will be evaluated by recording the incidence and severity of adverse events throughout the study. HRQoL will be assessed by change from baseline in patient-reported global health status and disease-related symptoms using the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire – Core 30 and the EORTC Gastrointestinal Neuroendocrine Tumor-Specific, 21-Item Quality of Life Questionnaire. HRQoL assessed by the EuroQol 5-Dimension 5-Level questionnaire is an exploratory endpoint.

2.4. Statistical methods

The trial is designed with an adaptive phase II/III structure. For the phase II part, approximately 120 patients will be randomized. If the phase II study results support trial continuation to phase III based on a prespecified Go/No-Go futility analysis, an additional 320 patients will be randomized (enrollment will continue without interruption while the Go/No-Go futility analysis is conducted) to achieve approximately 440 patients in total. To ensure adequate representation of primary tumor sites, enrollment of patients with epNET will be capped at 60% of the total study population, while patients with pNET will have a target enrollment of 40% of the total study population, up to a maximum of 50%.

Time-to-event endpoints will be summarized using the Kaplan–Meier method, and treatment effects (hazard ratios) will be estimated using Cox proportional hazards models. Missing data will not be imputed. Subgroup and biomarker analyses may be conducted to explore the potential impact of the clinical and biological heterogeneity of NETs on treatment outcomes.

2.5. Ethics and dissemination

The study is being conducted in accordance with the International Council for Harmonization Good Clinical Practice guidelines and the Declaration of Helsinki, as well as local laws and regulatory requirements. The study protocol was approved by the institutional review board or independent ethics committee at each participating site. All patients must provide written informed consent prior to participation in the study. Safety will be monitored by an independent data monitoring committee. Data will be collected by the investigators and analyses will be conducted by the sponsor.

2.6. Study status

STELLAR-311 is currently enrolling patients. Active or planned study sites include the United States, Australia, Austria, Brazil, Canada, France, Germany, Hong Kong, Italy, Mexico, the Netherlands, Poland, Puerto Rico, South Korea, Spain, Taiwan, and the UK. The study sites are listed on the ClinicalTrials.gov website (NCT06943755) [39].

3. Conclusions

Despite important developments in the management of advanced NETs, current treatments have limited efficacy, and many patients continue to have a poor prognosis [2]. Zanzalintinib is a novel, oral, small-molecule inhibitor of multiple kinases, including the TAM kinases (TYRO3, AXL, and MER), MET, and VEGF receptors, and may provide a novel treatment option for patients with advanced NETs, regardless of site of origin, somatostatin-receptor expression level, and functional status. STELLAR-311 is a global, phase II/III study to assess the efficacy and safety of zanzalintinib versus a standard-of-care treatment, everolimus, in patients with advanced NETs. As the first study to randomize a multitargeted inhibitor of the VEGF pathway against an mTOR inhibitor in this setting, the results of STELLAR-311 have the potential to inform future treatment decisions by helping to refine oral treatment options for these patients.

Funding Statement

The study is sponsored by Exelixis, Inc., Alameda, CA, USA. The funder was involved in the study design, decision to publish, and preparation of the manuscript.

Article highlights

• Despite improved outcomes for patients with neuroendocrine tumors (NETs) in recent years, those with advanced disease continue to have a poor prognosis; thus, more effective treatments are needed.

• Expression of vascular endothelial growth factor (VEGF) receptors and MET is associated with aggressive disease and poor prognosis in advanced NETs.

• Zanzalintinib is a novel, small-molecule inhibitor of multiple kinases, including the TAM kinases (TYRO3, AXL, and MER), MET, and VEGF receptors, which play important roles in oncogenic processes including tumor growth, angiogenesis, and metastasis.

• STELLAR-311 is a global, randomized, multicenter, open-label, phase II/III study evaluating the efficacy and safety of zanzalintinib versus everolimus in patients with advanced NETs.

• Patients aged ≥18 years with histologically confirmed, locally advanced/unresectable or metastatic NETs of pancreatic (pNET) or extra-pancreatic (epNET) origin with measurable disease and radiographic disease progression will be included in the study.

• Patients with pNETs or nonfunctional epNETs can have received up to one prior line of systemic anticancer therapy; patients with functional epNETs must have received one prior line of systemic anticancer therapy. Prior treatment with somatostatin analogs is allowed and is not included in the count of permitted prior systemic anticancer therapy.

• The primary endpoint is progression-free survival as assessed by blinded independent central review.

• Patients are currently being recruited in North America, South America, Europe, and the Asia-Pacific region.

• Results from STELLAR-311 have the potential to change the treatment landscape for patients with advanced NETs.

Disclosure statement

Jennifer Chan reports research funding from Camurus, Exelixis, Inc., and Lilly; consulting or advisory roles for Boehringer Ingelheim, Curium, Lantheus, and Sanofi; honoraria from Ipsen and Sanofi; participation on an advisory board/data monitoring committee for Novartis; and stock or stock options in Merck.

Jonathan R. Strosberg reports research funding from Boehringer Ingelheim, Crinetics, and Exelixis, Inc., and consulting or advisory roles for Boehringer Ingelheim, Exelixis, Inc., Merck, Sanofi, and Tersera.

Changhoon Yoo reports research funding from AstraZeneca, Boehringer Ingelheim, Boryung, Ipsen, Lunit Inc., and Servier; consulting or advisory roles for AstraZeneca, Bristol Myers Squibb, Eisai, Ipsen, MSD, Roche, and Servier; and honoraria from Abbisko, ABL Bio, AstraZeneca, Boryung, Bristol Myers Squibb, CKD pharm, Eisai, Elevar Therapeutics, HLB, Ipsen, MSD, Novartis, Qurient, Roche, and Servier.

Simron Singh reports research funding from Camurus and Novartis; consulting or advisory roles for Camurus, Ipsen, and Novartis; and participation on an advisory board/data monitoring committee for Ipsen and Novartis.

Huang Huang, Manika Goel, and Christian Scheffold are employees of and hold stock/stock options in Exelixis, Inc.

Jaume Capdevila reports research funding from Advanced Accelerator Applications, Amgen, AstraZeneca, Bayer, Eisai, Novartis, and Pfizer; and consulting or advisory roles for Advanced Accelerator Applications, Advanz, Amgen, Bayer, Eisai, Esteve, Exelixis, Inc., Hutchinson Pharma, Ipsen, ITM, Lilly, Merck Serono, Novartis, Pfizer, Roche, and Sanofi. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Medical writing assistance was provided by Tamsin Williamson of Twist Medical LLC and funded by Exelixis, Inc.

Reviewer disclosure

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Ethical declaration

The study is being conducted in accordance with the International Council for Harmonization Good Clinical Practice guidelines and the Declaration of Helsinki, as well as local laws and regulatory requirements. The study protocol was approved by the institutional review board or independent ethics committee at each participating site. All patients must provide written informed consent prior to participation in the study. Safety will be monitored by an independent data monitoring committee.

Data sharing statement

Individual patient data will not be shared.