Tivozanib Improves PFS Versus Bevacizumab in NRP-1 Low mCRC

Al B. Benson III, MD

Low serum neuropilin-1 (NRP-1) may indicate longer progression-free survival (PFS) with tivozanib compared with bevacizumab as a first-line treatment for patients with metastatic colorectal cancer (mCRC), according to final results of the phase II BATON-CRC study.

“The increase in PFS for patients treated with tivozanib compared to bevacizumab is both interesting and encouraging,” principal investigator Al B. Benson III, MD, professor of medicine at the Feinberg School of Medicine, Associate Director for Clinical Investigations at the Robert H. Lurie Comprehensive Cancer Center of Northwestern University, said in a statement. “I believe that these findings warrant further study in a prospectively defined trial of patients with low serum NRP-1 metastatic colorectal cancer.”

Final results of the BATON-CRC phase II trial, including results from a predefined biomarker analysis, were presented at the AACR Tumor Angiogenesis and Vascular Normalization Conference. The clinical trial randomized 265 patients in a 2:1 ratio to receive tivozanib, an oral VEGF TKI, in combination with mFOLFOX6 (n = 177) or bevacizumab and mFOLFOX6 (n = 88). Patients were stratified by lactate dehydrogenase, origin of cancer, and number of metastatic sites.

Patients in the tivozanib arm received 1.5 mg once daily for 21 days, followed by 7 days of treatment. The bevacizumab arm received 5 mg/kg every 2 weeks on days 1 and 15. All patients received mFOLFOX6 every 2 weeks of each 28-day cycle. The primary endpoint of the study was PFS.

Eligible patients were not previously treated with systemic chemotherapy, including adjuvant fluorouracil-containing therapy in the last 6 months, and had an ECOG performance status <1. Patients had not received prior VEGF therapy, including bevacizumab. Individuals with a significant thromboembolic or vascular disorder within 6 months were excluded from the trial.

In the overall population, PFS and overall response rate (ORR) were comparable between both arms. PFS and ORR were 9.8 months and 46.9%, and 9.5 months and 43.2% for tivozanib and bevacizumab, respectively.

In the predefined biomarker analysis, patients with low NRP-1 showed an improved PFS compared with those with high NRP-1. Patients with serum NRP-1 levels below the median had longer PFS with tivozanib (17.9 months; n = 52) compared with bevacizumab (11.2 months; n = 28 [HR = 0.38; P = .0075]). This demonstrates NRP-1 may be a potential biomarker of tivozanib activity relative to bevacizumab, according to the researchers.

PFS was shorter in patients with high serum NRP-1 levels (7.3 months with tivozanib versus 7.5 months with bevacizumab).

The study concluded, as a potential hypothesis, that in the presence of high serum NRP-1, VEGF-A164 is bound resulting in VEGFR-2 not being activated, making the VEGFR blockade method less significant. On the other hand, for patients with low serum NRP-1, VEGFR activation is higher, suggesting more efficacy for VEGFR-targeted therapies, like tivozanib.

“Where NRP-1 expression is low, the tumor appears to rely more heavily on the VEGF pathway and become more susceptible to anti-VEGF therapies, such as bevacizumab or tivozanib, an observation seen across several tumor types, including breast, gastric, colorectal and renal cell cancers,” said Benson. “Results from the BATON-CRC study are consistent with these observations.

The overall safety profile was comparable in both arms. The most common treatment-emergent, all-grade adverse event was diarrhea (58.2% in tivozanib arm versus 57.5% in bevacizumab arm). The most common grade 3/4 adverse event was neutropenia (39.5% in tivozanib arm versus 24.1% in bevacizumab arm).

The number of patients who discontinued treatment due to adverse events was 41.2% in the tivozanib arm and 34.5% in the bevacizumab arm. Nine patients died while on treatment or within 30 days of last dose; 7 in the tivozanib arm and 2 from the bevacizumab arm.

“These results underscore tivozanib’s activity and differentiating properties, including the potential benefit of a more complete anti-angiogenesis blockade in selected patient populations,” Michael Needle, MD, chief medical officer of AVEO, the developer of tivozanib, said in a statement. “We believe these data suggest a promising new development path forward in a significant patient population.”

Benson AB, Krivoshik A, Van Sant C, et al. Neuropilin-1 as a potential biomarker of progression-free survival benefit for tivozanib plus mFOLFOX6 versus bevacizumab plus mFOLFOX6 in metastatic colorectal cancer: post-hoc biomarker analysis of BATON-CRC phase II trial. Poster presented at the AACR Angiogeneis Meeting; March 6, 2015, Orlando, FL. Abstract 24.