Randomized, Double-Blind, Placebo-Controlled Phase II Study of Yeast-Brachyury Vaccine (GI-6301) in Combination with Standard-of-Care Radiotherapy in Locally Advanced, Unresectable Chordoma

Abstract

Background: Brachyury is a transcription factor overexpressed in chordoma and is associated with chemotherapy resistance and epithelial-to-mesenchymal transition. GI-6301 is a recombinant, heat-killed Saccharomyces cerevisiae yeast-based vaccine targeting brachyury. A previous phase I trial of GI-6301 demonstrated a signal of clinical activity in chordomas. This trial evaluated synergistic effects of GI-6301 vaccine plus radiation.

Materials and methods: Adults with locally advanced, unresectable chordoma were treated on a randomized, placebo-controlled trial. Patients received three doses of GI-6301 (80 × 10⁷ yeast cells) or placebo followed by radiation, followed by continued vaccine or placebo until progression. Primary endpoint was overall response rate, defined as a complete response (CR) or partial response (PR) in the irradiated tumor site at 24 months. Immune assays were conducted to evaluate immunogenicity.

Results: Between May 2015 and September 2019, 24 patients enrolled on the first randomized phase II study in chordoma. There was one PR in each arm; no CRs were observed. Median progressive-free survival for vaccine and placebo arms was 20.6 months (95% confidence interval [CI], 5.7-37.5 months) and 25.9 months (95% CI, 9.2-30.8 months), respectively. Hazard ratio was 1.02 (95% CI, 0.38-2.71). Vaccine was well tolerated with no vaccine-related serious adverse events. Preexisting brachyury-specific T cells were detected in most patients in both arms. Most patients developed T-cell responses during therapy, with no difference between arms in frequency or magnitude of response.

Conclusion: No difference in overall response rate was observed, leading to early discontinuation of this trial due to low conditional power to detect statistical difference at the planned end of accrual.

Implications for practice: Chordoma is a rare neoplasm lacking effective systemic therapies for advanced, unresectable disease. Lack of clinically actionable somatic mutations in chordoma makes development of targeted therapy quite challenging. While the combination of yeast-brachyury vaccine (GI-6301) and standard radiation therapy did not demonstrate synergistic antitumor effects, brachyury still remains a good target for developmental therapeutics in chordoma. Patients and their oncologists should consider early referral to centers with expertise in chordoma (or sarcoma) and encourage participation in clinical trials.

Keywords: Chordoma; Immunotherapy; Radiation therapy; Randomized clinical trial; Therapeutic vaccine.

Figure 1

Consort flow diagram. Note that all patients who were assessed for eligibility on this protocol were enrolled and randomized. A separate National Cancer Institute screening protocol is used for the general screening of new patients, but these data are not reported. Of 24 patients who gave signed informed consent and enrolled on study, 11 were randomized to the interventional (vaccine) arm and 13 to the placebo arm. Two patients on the vaccine arm left the study before first restaging and were considered not evaluable; these two patients were included in the intention‐to‐treat analysis. Patients who received placebo were allowed to cross over to vaccine after progression, but data postcrossover are not part of the primary efficacy analysis. Abbreviation: RT, radiation therapy.

Figure 2

(A): Kaplan‐Meier curve of progression‐free survival (PFS). In the interventional arm (vaccine), median PFS was 20.6 months (95% confidence interval [CI], 5.7–37.5 months), 12‐month PFS was 72.7% (95% CI, 37.1–90.3 months), and 24‐month PFS was 39.0% (95% CI: 10.6–67.3%). In the control (placebo) arm, median PFS was 25.9 months (95% CI, 9.2–30.8 months), 12‐month PFS was 76.9% (95% CI, 44.2–91.9%), and 24‐month PFS was 57.7% (95% CI, 25.6–80.1%). There was no difference between the arms (p = .97). Hazard ratio (PFS) for vaccine versus placebo was 1.02 (95% CI, 0.38–2.71). (B): Kaplan‐Meier curve of overall survival (OS). In the interventional arm (vaccine), median OS was 37.5 months (95% CI, 21.6–50.6 months). In the control (placebo) arm, median OS was not reached. Hazard ratio (OS) for vaccine versus placebo was 2.65 (95% CI, 0.79–8.81). There was little difference between the arms (p = .10), but results somewhat favor placebo.

Figure 3

Frequency of patients developing T‐cell responses against the TAAs brachyury, MUC1, and CEA during therapy. Immune responses were calculated by comparing the absolute number of CD4+ or CD8+ T cells producing cytokine (IFN‐γ, TNF‐α, IL‐2) or positive for CD107a per 1 × 10⁶ PBMCs plated at the start of the in vitro stimulation assay. For analysis of the development of any T‐cell response, background (obtained with the negative control peptide pool, human leukocyte antigen [HLA]) and any response prior to therapy were subtracted: [TAA after therapy – HLA after therapy] – [TAA before therapy – HLA before therapy]; the frequency of patients developing any immune response (>250 positive cells) after therapy is indicated. Multifunctional T‐cell responses, or CD4+ and CD8+ T cells expressing two or more of IFN‐γ, TNF‐α, IL‐2, or CD107a, were also evaluated. The frequency of patients developing a more than threefold increase in multifunctional T cells after therapy is shown. Abbreviations: TAA, tumor‐associated antigen.