A phase II study of lenvatinib plus everolimus in advanced extra-pancreatic neuroendocrine tumors (epNETs): Updated results and real-world comparison.

4145 Background: Advanced epNETs are rare malignancies with limited treatment options beyond somatostatin analogs, peptide receptor radionuclide therapy, and Everolimus (E). Preclinical evidence suggests that dual blockade of VEGF and FGF is an effective antiangiogenic strategy and that concomitant inhibition of the mTOR pathway may be further synergistic. Lenvatinib (L), a multi-target tyrosine kinase inhibitor, suppresses VEGFR, FGFR, and other angiogenic pathways, while E targets the mTOR pathway. Their combination may synergistically impair angiogenesis and tumor growth. Methods: This open-label, single-center, phase II study evaluated L + E in patients (pts) with advanced, progressive, well-differentiated (a/p w-d) epNETs. The primary endpoint was objective response rate (ORR). Secondary endpoints were progression-free survival (PFS) and safety. Following a 2-stage design with H0: ORR < 5% and H1: ORR ≥ 20%, up to 32 pts were needed for type I & II error = 10%. In a post-hoc analysis, a real-world cohort of 2:1 matched a/p w-d epNET pts receiving E alone served as a historical comparator. Differences in ORR were assessed using univariable and multivariable logistic regression while those in terms of PFS were analyzed with propensity score-based inverse probability of treatment weighting (IPTW)-adjusted Cox proportional hazards regression and Kaplan-Meier method. Results: 32 pts were enrolled. The starting dose regimen was L 18 mg + E 5 mg p.o. daily with L being reduced to 14 mg p.o. daily after the first 3 pts experienced Grade 3 adverse events (AEs). Median age was 59 years (range 33 – 76), and 59% were male. Primary tumor sites included small bowel (59%), lung & thymus (16%), unknown (16%), and colorectal (9%) with the majority being G2 (69%). Median number of prior therapies was 2 (range 0 – 3) while 11 pts had carcinoid syndrome. The study met its primary endpoint: L + E achieved an ORR of 43.8% (6.3% unconfirmed), which was significantly higher than that observed with E alone (3.1%; OR 24.50, 95% CI 5.09 – 117.94, p < 0.001). This finding was supported by the multivariable analysis (OR 20.43, 95% CI 3.93 – 106.13, p < 0.001). After propensity score-based IPTW adjustment, a trend toward longer PFS favoring L + E (16.0 months [95% CI 13.0 – 23.6] vs 11.3 months [95% CI 8.6 – 24.3]; HR 0.88 [95% CI 0.51 – 1.52], p = 0.647) was observed. On trial, 23 Grade 3 AEs (11 after L dose reduction) were noted with elevated LFTs (8), hypertension (6) and thrombocytopenia (4) being the most frequent while one Grade 4 AE (hypertriglyceridemia) was reported. Conclusions: L + E demonstrated markedly superior ORR and a trend toward prolonged PFS compared to E alone with a manageable safety profile. These findings highlight its potential as a therapeutic option for a/p w-d epNETs and warrant further investigation in randomized trials. Clinical trial information: NCT03950609 .