Phase II Clinical Trial of Ixabepilone in Metastatic Cervical Carcinoma

Abstract

Lessons Learned

• Accrual to cervical cancer studies remains a puzzling challenge given the lack of options and the dismal prognosis of this disease. The majority of patients referred for a trial such as this have very advanced disease that is difficult to manage.

• The observation of 4 partial responses among the 41 patients indicates that ixabepilone has some activity but not sufficient for further development without greater understanding of mechanisms of sensitivity and resistance.

Background.

Ixabepilone is a microtubule-stabilizing agent approved for metastatic breast cancer. Preclinical data have shown that ixabepilone is active in taxane-sensitive and -resistant cells. Metastatic cervical carcinoma (mCC) has a poor prognosis and no established second-line therapies. This study assessed the efficacy and safety of ixabepilone in previously treated mCC.

Methods.

Patients with histologically confirmed mCC and at least one prior cisplatin-containing regimen were treated with ixabepilone [6 mg/m² per day for 5 days] every 21 days. The primary endpoint was progression-free survival (PFS) according to the Response Evaluation Criteria in Solid Tumors (RECIST). Secondary endpoints were response rate, rate of tumor growth, overall survival (OS), and safety. Levels of glu-terminated and acetylated tubulin, markers of microtubule stabilization, and surrogates for target engagement were assessed by Western blot.

Results.

In total, 41 patients were enrolled; 34 had tumors with primarily squamous histology. The median number of prior therapies was 2 (range 1–6). Four patients (9.7%) had a partial response. Median PFS in months was 2.3 for all, 3.84 for taxane-naïve, and 2.03 for taxane-pretreated patients (p = .13). Consistent with this, we found statistically similar (p = 1) rates of growth in taxane-naive patients (0.0035 per day) and taxane pretreated patients (0.0053 per day). Median OS was 5.84 months. G1/2 toxicities included vomiting (43%), sensory neuropathy (21%), and fatigue (60%). Bowel fistulas were observed in 7% of patients. Glu and acetylated tubulin were assessed in tumor samples from 11 patients during the first cycle of treatment. Although there was clear evidence of “target engagement” and microtubule stabilization in all tumors, a correlation between the extent of tubulin stabilization and response to therapy could not be demonstrated.

Conclusion.

Ixabepilone was well tolerated but showed very modest activity in second- or later-line mCC and cannot be recommended as a therapy. Target engagement was demonstrated but was not correlated with responses, suggesting that other factors mediate drug sensitivity. New strategies are needed for refractory mCC.

Discussion

Treatment of recurrent or advanced metastatic cervical carcinoma has been based on systemic chemotherapy. In first line, cisplatin or carboplatin are frequently combined with a second chemotherapeutic agent, often a taxane, although gemcitabine, navelbine, and topotecan have shown similar rates of response. A recent phase III trial reported that the addition of bevacizumab to cisplatin plus paclitaxel or topotecan plus paclitaxel improved overall survival (OS; 17.0 months vs. 13.3 months; hazard ratio for death, 0.71; 98% confidence interval [CI], 0.54–0.95; p = .004 in a one-sided test) and led to a higher response rate (48% vs. 36%, p = .008). This led to the approval of bevacizumab for this indication by the U.S. Food and Drug Administration (FDA).

Unfortunately, no chemotherapy regimen has shown meaningful activity in second line. In this setting, single-agent chemotherapy is often used, although this approach is not supported by convincing evidence. Phase II clinical trials have reported response rates between 10% and 20% with topotecan, ifosfamide, vinorelbine, or pemetrexed. Ixabepilone (Ixempra) is a semisynthetic epothilone B analog with microtubule stabilizing properties. Preclinically, ixabepilone is more potent than taxanes and is active in cancer cell lines resistant to taxanes, in part because it is not a substrate for P-glycoprotein (Pgp). The FDA approved ixabepilone for treatment of metastatic breast cancer previously treated with an anthracycline or a taxane. We previously reported the activity of ixabepilone in metastatic cervical adenocarcinoma patients.

In the present study, we report the results of a phase II clinical trial using ixabepilone in metastatic or recurrent cervical cancer after the failure of first or subsequent lines of treatment. The overall response rate was 9.7%, with a median progression-free survival (PFS) of 2.3 months that was independent of previous taxane exposure. Additionally, estimates of the tumor growth rate, a variable we have previously shown to correlate well with PFS and OS, was not significantly different between patients with/without prior exposure to a taxane. No new toxicities were encountered, and the occurrence of bowel fistulae in three patients, all of whom had previous radiation therapy, was likewise not unexpected. We considered in these patients the possibility this complication could represent ixabepilone-induced radiation recall, as has been previously suggested with taxanes but could not prove this.

Finally, we were able to document target engagement of ixabepilone with microtubules as evidenced by an increased level of glu-terminated and acetylated tubulin (Fig. 1). However, as in renal cell cancer, the depth of the tumor response was independent of the extent of microtubule stabilization, suggesting that engagement is likely necessary but insufficient and that additional events distal to target engagement are also important.

Figure 1.

The top two panels show the fold change between the pre- and post-ixabepilone tumor biopsies in the levels of two α-tubulin modifications that occur when polymerization is induced by engagement of ixabepilone with its target. The bottom panel depicts a waterfall plot for these patients. There was no correlation between the fold change in the α-tubulin modifications and tumor response, suggesting that in addition to target engagement, other factors influence the tumor response to therapy.

Abbreviation: RECIST, Response Evaluation Criteria in Solid Tumors.

Author disclosures available online.