Editor’s Note: The 2024 European Society for Medical Oncology (ESMO) Annual Meeting has successfully concluded. Several studies presented during the neuroendocrine tumor (NET) oral sessions have offered new treatment strategies and molecular profiling for advanced unresectable NETs. Notably, the Phase III CABINET study on cabozantinib achieved its primary endpoint of progression-free survival (PFS), and the multi-center single-arm CVM-005 study showed impressive PFS results with the anti-vasculogenic mimicry drug CVM-1118, especially when combined with SSA (somatostatin analogs), with the median PFS for extra-pulmonary NETs reaching nearly two years. Furthermore, multi-omics profiling has provided a foundation for prognostic assessment, classification-based treatment, and biomarker exploration in NETs. Oncology Frontier invited Dr. Jie Chen from Fudan University Shanghai Cancer Center to introduce and comment on these studies.

01

Cabozantinib versus placebo for advanced neuroendocrine tumors (NET) after progression on prior therapy (CABINET Trial/Alliance A021602): Updated results including progression free-survival (PFS) by blinded independent central review (BICR) and subgroup analyses

NETs are highly vascularized tumors, and anti-angiogenic tyrosine kinase inhibitors (TKIs) have become a key treatment option for unresectable NETs. Sunitinib became the first TKI to complete a Phase III trial for advanced NETs around 2011. The second TKI to report Phase III trial results was surufatinib, independently developed in China, with the SANET-ep and SANET-p studies completed in 2020. The third TKI, reported at the 2023 ESMO Congress, was cabozantinib in the CABINET study (Alliance A021602), which demonstrated improvements in PFS based on investigator assessments. The 2024 ESMO Congress further reported PFS results assessed by blinded independent central review (BICR).

Cabozantinib is a multi-target TKI that inhibits VEGF-R, c-MET, AXL, and RET. The CABINET study was similar to previous studies, dividing patients into pancreatic NET (pNET) and non-pancreatic NET (epNET) cohorts. However, this study included patients with G1-G3 tumors (while previous studies mainly focused on G1-G2), and it allowed patients who had previously received treatments like everolimus, sunitinib, and Lu-177.

BICR results showed that in the epNET cohort, the cabozantinib group had a median PFS of 8.4 months compared to 3.9 months in the placebo group, with a 62% reduction in the risk of death or progression (HR 0.38, 95% CI: 0.25-0.59, P < 0.0001). The benefit trend was consistent across subgroups, and PFS assessed by BICR was similar to that assessed by investigators.

In the pNET cohort, the cabozantinib group had a median PFS of 13.8 months compared to 4.4 months in the placebo group, with a 77% reduction in the risk of death or progression (HR 0.23, 95% CI: 0.12-0.42, P < 0.0001). The benefit trend was consistent across subgroups, and the BICR-assessed PFS was longer than the investigator-assessed PFS, though the overall HR remained similar.

The objective response rates (ORR) for cabozantinib versus placebo were 5% vs. 0% in the epNET cohort and 19% vs. 0% in the pNET cohort. Overall survival (OS) data remain immature, with HRs of 0.86 and 0.95 in the epNET and pNET cohorts, respectively. Whether PFS will translate into OS benefit remains to be seen.

Regarding drug exposure and tolerability, 66% and 68% of patients in the epNET and pNET cohorts, respectively, required dose reductions due to drug toxicity. Additionally, 31% and 20% of patients in these cohorts discontinued treatment due to adverse events (AEs). Common grade 3-4 AEs in the epNET cohort included fatigue (13%), diarrhea (11%), and hypertension (21%), while in the pNET cohort, they included fatigue (11%), hypertension (22%), and thrombotic events (11%).

Dr. Jie Chen：The CABINET study is the third major Phase III trial showing positive results for anti-angiogenic TKIs in the treatment of unresectable NETs. This study differed from previous ones by including a small number of G3 patients, as well as patients who had already received other standard treatments like TKIs or SSAs. The results demonstrated a significant PFS benefit for cabozantinib over placebo. Overall, the PFS for cabozantinib, similar to previous TKIs, appears to be limited to within one year, suggesting this may be the ceiling for anti-angiogenic TKIs. The efficacy of cabozantinib was higher in pNETs due to their higher vascular supply. In terms of safety and tolerability, cabozantinib was comparable to previous TKIs, with a significant proportion of patients requiring dose reductions or treatment adjustments. While cabozantinib demonstrated PFS benefits, whether this will translate into OS benefits remains uncertain. These results affirm the value of anti-angiogenic TKIs in unresectable NETs but also underscore the need for more effective and safer treatments to surpass the current “ceiling” of therapeutic benefit.

02

CVM-005: Phase IIa study of CVM-1118, a novel oral anti-vasculogenic mimicry (VM) agent, in advanced neuroendocrine tumors (NET) after progression on prior therapy

As previously mentioned, traditional anti-angiogenic TKIs seem to hit a PFS ceiling of around 12 months, necessitating the development of new mechanisms of action. Targeting vasculogenic mimicry (VM) has emerged as a novel anti-cancer strategy. CVM-1118 is a new oral drug that inhibits the overexpression of TRAP1 (tumor necrosis factor receptor-associated protein 1), a mitochondrial HSP90 family chaperone that regulates tumor VM through pathways such as AKT and BRAF. In previous Phase I trials, CVM-1118 demonstrated anti-tumor activity in non-small cell lung cancer (NSCLC), colorectal cancer (CRC), renal cell carcinoma (RCC), ovarian cancer, and lung carcinoids.

The multi-center, single-arm CVM-005 study enrolled 35 patients with G1 and G2 NETs, including gastrointestinal (GI) NETs, pancreatic NETs, and pulmonary NETs. These patients had previously received multiple treatments, such as everolimus, sunitinib, PRRT, and chemotherapy. Thirty patients had received SSAs, and 14 continued SSA treatment during the study. After enrollment, patients received 200-300 mg of CVM-1118 twice daily (28-day cycles) until disease progression or death. The study aimed for a median PFS of 8 months, an extension of 3.6 to 5.6 months over historical controls.

The results exceeded expectations, with a median PFS of 10.5 months for the overall population and an ORR (CR+PR) of 2.9%. For patients who continued SSA treatment, the median PFS reached 22.4 months. In subgroup analyses, GI NET and pulmonary NET patients achieved median PFS of 22.4 months and 25.1 months, respectively. Safety results were also favorable, with most AEs being grade 1/2, and no serious treatment-related AEs (TRAEs). The incidence of grade 3-4 TRAEs was only 18%.

Dr. Jie Chen：The anti-VM mechanism of CVM-1118 represents a novel approach to NET treatment. Despite being a single-arm study, the study achieved a median PFS of 10.5 months, doubling that of historical controls (4.9 months). Notably, CVM-1118 combined with SSA resulted in a median PFS of 22.4 months, far exceeding previous SSA + TKI studies. For example, in the 2021 ASCO GI conference, the AXINET trial, which combined axitinib with long-acting octreotide, failed to show a significant PFS improvement (17.2 vs. 12.3 months, HR 0.816, P = 0.169). Additionally, GI and pulmonary NETs typically respond poorly to traditional TKI treatment, as seen in CABINET and other previous studies, yet CVM-1118 achieved a median PFS of 22.4 months for GI NETs and 25.1 months for pulmonary NETs, a remarkable 4-7 times improvement over historical controls (5.6 and 3.6 months, respectively). This impressive two-year PFS has far surpassed the PFS ceiling for traditional TKIs. Furthermore, CVM-1118 exhibited excellent safety, with a grade 3-4 TRAE rate of only 18%, far lower than that reported in the CABINET study. Although the CVM-005 study had a small sample size and was single-arm in design, its efficacy and safety results far exceeded expectations, and future expanded clinical trials are eagerly awaited.

03

Refining molecular classification of neuroendocrine tumors from diverse origins using multi-omic integration models: Unveiling novel neuroendocrine subtypes and their clinical implications

NETs are characterized by diverse tissue and anatomical distributions, with non-specific clinical and molecular features. Currently, there is no widely accepted molecular classification system to guide prognosis assessment and treatment selection. Recent multi-omics studies have focused on the epigenetic features of NETs, aiming to develop a feasible molecular classification system.

The ESMO Congress reported this study, which analyzed 194 formalin-fixed, paraffin-embedded tumor samples from NET patients across various primary sites. The clinical and tumor characteristics of the patients are shown in the table below, demonstrating a relatively diverse and balanced sample set, representing good generalizability.

Researchers primarily analyzed transcriptomic and methylation features, using Multi-Omics Factor Analysis (MOFA) to build an 8-factor model that integrates two datasets, which were then applied to clustering algorithms for classification.

The MOFA model ultimately identified three neuroendocrine subtypes (NS1, NS2, NS3), which were significantly associated with patient prognosis. NS1 had a favorable prognosis, with features like immune enrichment (e.g., high cytotoxic T-cell and monocyte lineage infiltration) and increased expression of metabolic and digestive genes. NS3 had the worst prognosis, exhibiting aggressive clinical characteristics such as poorer differentiation (G2-3), later-stage disease (diagnosed at stage IV), and higher 5-HIAA levels. NS3 also showed elevated levels of methylation and copy number variations, with higher expression of genes related to neuroendocrine function and cell proliferation (e.g., PTPRN, BEX1, and NOVA1). NS2 had an intermediate prognosis, with high NK cell levels and increased expression of endocrine function genes. External validation confirmed these findings.

Dr. Jie Chen：The anatomical and pathological complexity of NETs has long posed a challenge. Precision medicine has led to the exploration of genomic research on NETs, as seen in the 2019 WHO classification, which included genomic descriptions such as MEN1, DAXX, and ATRX mutations common in well-differentiated NETs, while TP53 and RB1 mutations were common in neuroendocrine carcinomas (NECs). However, there has not yet been a widely accepted molecular classification system for guiding clinical diagnosis and treatment. This study used a multi-omics analysis model to extract relevant factors from transcriptomics and methylation datasets, creating three molecular subtypes: NS1-3. These subtypes not only revealed common biological variations among NETs from different primary sites but also helped to distinguish different prognostic characteristics. This study has positive implications for identifying biomarkers and developing classification-based treatment strategies for NETs, and further validation and application are eagerly anticipated.

References:

1. Chan J, et al. Cabozantinib vs. Placebo for Advanced Neuroendocrine Tumors (NET) after Progression on Prior Therapy (CABINET Trial/Alliance A021602): Updated PFS by Blinded Independent Central Review (BICR) and Subgroup Analyses. ESMO 2024; 1141O.
2. Raymond E, et al. Updated Results of the Phase III Trial of Sunitinib vs. Placebo for Advanced Pancreatic Neuroendocrine Tumor (NET). Gastrointestinal Cancers Symposium, 2010, Orlando, FL. Abstract 127.
3. Xu J, et al. Surufatinib in Advanced Extrapancreatic Neuroendocrine Tumours (SANET-ep): A Randomized, Double-Blind, Placebo-Controlled, Phase 3 Study. Lancet Oncol, 2020; 21(11): 1500-1512.
4. Xu J, et al. Surufatinib in Advanced Pancreatic Neuroendocrine Tumours (SANET-p): A Randomized, Double-Blind, Placebo-Controlled, Phase 3 Study. Lancet Oncol, 2020; 21(11): 1489-1499.
5. Yen CJ, et al. CVM-005: Phase IIa Study of CVM-1118, a Novel Oral Anti-Vasculogenic Mimicry (VM) Agent, in Advanced Neuroendocrine Tumors (NET) after Progression on Prior Therapy. ESMO 2024; 1145MO.
6. Garcia-Carbonero R, et al. A Phase II/III Randomized Double-Blind Study of Octreotide Acetate LAR with Axitinib vs. Octreotide Acetate LAR with Placebo in Patients with Advanced G1-G2 NETs of Non-Pancreatic Origin (AXINET Trial). ASCO GI 2021.
7. Carretero-Puche C, et al. Refining Molecular Classification of Neuroendocrine Tumors from Diverse Origins Using Multi-Omic Integration Models: Unveiling Novel Neuroendocrine Subtypes and Their Clinical Implications. ESMO 2024; 1143O.

Dr. Jie Chen

• Chief Physician, Professor, and Doctoral Supervisor
• Chief Expert in Neuroendocrine Tumor Multidisciplinary Care at Fudan University Shanghai Cancer Center
• Director of the Neuroendocrine Tumor Center at Fudan University Shanghai Cancer Center
• Chair of the Chinese Anti-Cancer Association Neuroendocrine Tumor Committee