Angiogenesis Inhibitor Slows Extrapancreatic Neuroendocrine Tumors

Progression-free survival (PFS) in extrapancreatic neuroendocrine tumors (NETs) improved significantly with the addition of an angiogenesis inhibitor to standard endocrine therapy, according to a randomized trial.

Median PFS by independent review increased from 9.9 months with the somatostatin analog octreotide (Sandostatin LAR) to 16.6 months when combined with axitinib (Inlyta). The difference represented a 29% reduction in the hazard for disease progression or death (95% CI 0.54-0.94). Four times as many patients responded to the combination, and about 80% of patients had some degree of tumor shrinkage.

The results are consistent with the improvement in PFS by investigator assessment, which was reported earlier this year, said Rocio Garcia-Carbonero, MD, of the Hospital Universitario 12 de Octubre in Madrid, at the European Society for Medical Oncology (ESMO) virtual meeting.

“The central reading was very similar to that reported by investigators, 16.6 months versus 17.2 months,” she noted. “The assessment by central reading of the placebo arm was somewhat lower (9.9 vs 12.3 months), and therefore the hazard ratio per central reading was also lower and was statistically significant, whereas per investigator (HR 0.82), the difference did not achieve this statistical significance.”

“So we can say that axitinib significantly improved progression-free survival in combination with octreotide LAR versus placebo and octreotide LAR in patients with advanced, progressive G [grade] 1-2 extrapancreatic NETs per independent radiological assessment blinded to treatment allocation,” she added. “Axitinib also significantly improved objective response … with an odds ratio of 4.58. The safety profile is consistent with the known safety profile for axitinib and octreotide. Survival data are still immature.”

The rational for the AXINET trial came from recognition that angiogenesis plays an important role in NET development and progression, Garcia-Carbonero noted. Axitinib is a selective inhibitor of VEGF receptor 1, 2, and 3 and has shown activity in vascular-dependent solid tumors, including approved indications in renal cell carcinoma.

Patients with advanced, progressive extrapancreatic NETs have limited options. The AXINET trial was designed to determine whether the addition of axitinib to octreotide LAR would improve outcomes in grade 1/2 extrapancreatic NETs.

Investigators randomized 256 to octreotide LAR plus axitinib or placebo. The primary endpoint was investigator-assessed PFS, and the primary analysis showed that axitinib did not significantly improve PFS versus single-agent octreotide LAR. Garcia-Carbonero reported findings for the secondary outcome of PFS by central radiologic review.

The patients had a median age of about 61. Patients in the axitinib arm were more likely to have grade 2 disease (77.8% vs 66.2%). About 60% of the patients had primary tumors in the gastrointestinal tract (predominantly jejunum-ileum), followed by the lung in about 30% of cases. About half the patients had undergone surgery, and a similar proportion had prior exposure to a somatostatin analog. About 45% of the patients had received no systemic therapy.

The 6.9-month difference in PFS by blinded central review achieved statistical significance in favor of axitinib (P=0.017), as did the difference in objective response rate (13.2% vs 3.2%, P=0.0045).

Comparison of central versus investigator assessment of PFS showed a concordance rate of 74.5%. Most differences related to shifts between partial response and stable disease, said Garcia-Carbonero.

ESMO invited discussant Nicola Fazio, MD, of the European Institute of Oncology in Milan, tried to clear up potential confusion between the two assessments of PFS. The AXINET study began as a phase II trial with investigator-assessed PFS as the primary endpoint. Trial organizers subsequently converted the study to a phase III design and added central review of PFS as a secondary endpoint. The addition of the secondary assessment of PFS was not widely communicated, including the initial presentation of investigator-assessed PFS earlier this year, he said.

Over the past decade, trials similar to AXINET have generally adopted central review of PFS, Fazio continued. One notable exception was a Chinese study of surufatinib for extrapancreatic NETs, which had a primary endpoint of investigator-assessed PFS. Investigators have used different criteria to assess PFS (WHO, SWOG, RECIST 1.0, RECIST 1.1) and the criteria have changed over the years.

The RADIANT trials of everolimus (Afinitor) for advanced NETs provided insight into differences between investigator-assessed and centrally reviewed outcomes, said Fazio. Collectively, the trials showed about a 30% discordance between central and local tumor response assessment. In the surufatinib study, investigator-assessed PFS for the experimental treatment was almost 2 months higher as compared with centrally reviewed PFS.

“Central versus local tumor response assessment remains a debatable issue of randomized clinical trials in NETs,” Fazio said. “Axitinib showed a relevant impact on PFS according to the blinded central review, which was a secondary endpoint. If axitinib is approved by the FDA or EMA [European Medicines Agency] on this basis, it will be the first tyrosine kinase after sunitinib [Sutent] in NET based on a Western phase III randomized controlled trial.”