Interim analysis of a phase I/IIa trial assessing E39+GM-CSF, a folate binding protein vaccine, to prevent recurrence in ovarian and endometrial cancer patients

Abstract

Background: Folate binding protein(FBP) is an immunogenic protein over-expressed in endometrial(EC) and ovarian cancer(OC). We are conducting a phase I/IIa trial of E39 (GALE 301)+GM-CSF, an HLA-A2-restricted, FBP-derived peptide vaccine to prevent recurrences in disease-free EC and OC patients. This interim analysis summarizes toxicity, immunologic responses, and clinical outcomes to date.

Methods: HLA-A2+ patients were vaccinated(VG), and HLA-A2- or -A2+ patients were followed as controls(CG). Six monthly intradermal inoculations of E39+250mcg GM-CSF were administered to VG. Demographic, safety, immunologic, and recurrence rate(RR) data were collected and evaluated.

Results: This trial enrolled 51 patients; 29 in the VG and 22 in the CG. Fifteen patients received 1000mcg E39, and 14 received <1000mcg. There were no clinicopathologic differences between groups(all p ≥ 0.1). E39 was well-tolerated regardless of dose. DTH increased pre- to post-vaccination (5.7±1.5 mm vs 10.3±3.0 mm, p = 0.06) in the VG, and increased more in the 1000mcg group (3.8±2.0 mm vs 9.5±3.5 mm, p = 0.03). With 12 months median follow-up, the RR was 41% (VG) vs 55% (CG), p = 0.41. Among the 1000mcg patients, the RR was 13.3% vs 55% CG, p = 0.01. Estimated 2-year DFS was 85.7% in the 1000mcg group vs 33.6% in the CG (p = 0.021).

Conclusions: This phase I/IIa trial reveals that E39+GM-CSF is well-tolerated and elicits a strong, dose-dependent in vivo immune response. Early efficacy results are promising in the 1000 mcg dose cohort. This study proves the safety and establishes the dose of E39 for a larger prospective, randomized, controlled trial in HLA-A2+ EC and OC patients to prevent recurrence.

Keywords: cancer; endometrial; folate binding protein; immunotherapy; ovarian.

Figure 1. Patient flow for study enrollment

The prospectively followed control group also included three HLA-A2 positive patients who declined vaccine administration.

Figure 2. Maximum local and systemic toxicities appreciated within the dosing cohorts

There was a statistically significant difference between the local toxicity experienced by the 1000mcg vs <1000mcg group (p=0.04). No significant difference was noted within systemic toxicity. This difference may be due to an increased consumption of GM-CSF with higher doses of the peptide, thus leading to a smaller amount of GM-CSF to cause worsened local or systemic toxicity.

Figure 3. Immune response before and after the primary vaccination series according to dosing cohorts

The average size DTH reaction in all vaccinated patients prior to vaccination was 5.7 ±1.5 mm compared to 10.3 ±2.8 mm post-vaccination (p=0.06). The 1000mcg patients had a statistically significant increase in pre-vaccination versus post-vaccination DTH (3.8 +2.0 mm vs 9.5 +3.5 mm, p=0.03), while <1000mcg patients experienced a smaller increase in pre- vs post-vaccination, which was not statistically significant (7.8 +2.1 mm vs 11.3 +4.8 mm, p=0.28).

Figure 4. The 2-year estimated DFS for vaccinated patients was 43% (95% confidence interval (CI): 18-66%) versus 33.6% (95% CI: 13-56%) in control patients (p=0.36)

The vaccinated patients experienced a 31% reduction in relative recurrence risk regardless of dose.

Figure 5. This subgroup analysis was performed based on dosing

The 2-year estimated DFS indicated a significant survival advantage for the 1000 mcg group at 85.7% (95% CI: 54-96%) compared to controls at 33.6% (95% CI: 13-56%) and the <1000 mcg group at 20.8% (95% CI: 4-47%). Comparing the 1000 mcg and control groups, there was an 83% reduction in relative risk of recurrence.