Two staged phase II clinical trial of Eribulin monotherapy in advanced or recurrent cervical cancer.

ABSTRACT

BACKGROUND: Eribulin a microtubule targeting agent and analog of Halichondrin B, a natural product isolated from marine sponge H. okadai, has proven clinical efficacy in metastatic pretreated breast cancer and liposarcoma. We conducted a 2-stage Phase II study of eribulin in patients with advanced/recurrent cervical cancer to examine its clinical activity and evaluate biomarkers for predictors of response. METHODS: Women with advanced/recurrent cervical cancer after ≤1 prior chemotherapy regimen, measurable disease and ECOG performance status ≤2 were treated with eribulin (1.4 mg/m2 IV day 1 and 8, every 21 days) with tumor assessments every 2 cycles. Primary endpoint was 6-month progression-free survival (PFS6); secondary were best overall response (RECISTv1.1), toxicity (CTCAEv4.03) and overall survival (OS). Exploratory endpoints were associations of biomarkers with clinical activity. Immunohistochemistry was performed on archival tumor samples. Overexpression was defined when both intensity and distribution scores were ≥ 2. RESULTS: 32 patients enrolled from 11/2012-5/2017. 29/32 patients had prior chemotherapy with cisplatin/paclitaxel/bevacizumab (n = 12) or cisplatin/gemcitabine (n = 12) as the most common regimens. 14 patients received prior paclitaxel. 1 (3%) had a complete response, 5 (16%) had a partial response and 13 (41%) had stable disease for ORR of 19% (95% CI 8, 37). Those who are paclitaxel naïve experienced the greatest benefit with a 29% ORR (95% CI 12, 54). Patients who received prior paclitaxel responded less favorably than those who did not (p = .002) and had a shorter PFS and OS. Grade 3/4 adverse events occurring in >10% of patients were anemia (n = 12, 38%), neutropenia (n = 7, 22%) and leukopenia (n = 6, 19%). Analysis of correlative predictors of response revealed that patients who did not overexpress ßII and BAX were significantly more likely to respond to e`ribulin. PFS was significantly shorter in patients with ßII and BAX overexpression, OS was significantly shorter in those with ßIII and BAX overexpression. These associations remained after multivariate analysis. CONCLUSIONS: Eribulin shows modest activity in patients with recurrent/advanced cervical cancer with a favorable toxicity profile. Prior paclitaxel exposure is associated with decreased eribulin response. ßII, ßIII tubulin subtypes and BAX are predictors of response and survival. Eribulin may be an option for women with paclitaxel-naïve recurrent/advanced cervical cancer.