Background: Peptide cancer vaccines may be most effective when used in earlier stage cancers or pre-cancers where systemic and tumor microenvironmental immune suppression are less profound. Nelipepimut-S (NPS) plus granulocyte-macrophage colony-stimulating factor (GM-CSF) is a vaccine comprised of a human leukocyte antigen (HLA) restricted peptide from the extracellular domain of the HER2 protein (E75) combined with GM-CSF. We have completed a randomized phase II trial of preoperative vaccination with NPS+GM-CSF vs. GM-CSF alone with the primary outcome being NPS-specific cytotoxic T lymphocyte (CTL) responses.

Methods: HLA-A2 positive, DCIS patients were enrolled and randomized to either NPS+GM-CSF vs GM-CSF alone. The patients received two vaccinations prior to surgery at 2-week intervals. The number of NPS-specific CTL was measured at specified intervals (pre-vaccination, time of surgery, 1 month (+/- 7 days) post-op, and 3 months (+/- 7 days) post-op) using a flow cytometry-based dextramer assay. Differences in NPS-specific CTL responses between the two groups and between baseline pre-vaccination and 1-month post-op were analyzed using either a two-sample t-test or Wilcoxon rank sum test, when appropriate. The incidence and severity of adverse events, graded according to Common Terminology Criteria for Adverse Events (CTCAE) version (v) 4.03, were recorded and compared between treatment groups.

Results: 45 patients were registered; 7 withdrew consent, 1 opted for surgery at an external facility, 20 were ineligible due to negative HLA-A2, and 4 failed screening for other reasons, leaving 13 patients enrolled. The 13 patients were randomized (2:1) into treatment groups, with nine patients receiving NPS+GM-CSF and four patients receiving GM-CSF alone. The two groups were well-matched for age; however, the GM-CSF alone group had higher percentages of African American (50% vs. 22%) and Hispanic (25% vs. 11%) patients as compared to the NPS+GM-CSF group. In general, vaccination was well-tolerated with similar treatment-related toxicity profiles in the NPS+GM-CSF vs GM-GSF groups (Grade 1 - 93.3% vs. 89.3%, Grade 2 - 6.7% vs. 10.7%, respectively). The mean NPS-specific CTL% in the NPS+GM-CSF group at 1-month post-op was double that of the GM-CSF alone group (0.10 +/- 0.12% vs. 0.05 +/- 0.08, p=0.70). In addition, between baseline pre-vaccination and 1-month post-op, the NPS+GM-CSF group experienced an 11-fold increase in percentage of NPS-specific CTL (0.01 +/- 0.02% vs. 0.11 +/- 0.12%) as compared to only a 2.25-fold increase of NPS-specific CTL in the GM-CSF alone group (0.04 +/- 0.07% vs. 0.09 +/- 0.15%).

Conclusions: NPS+GM-CSF is safe and well-tolerated when given preoperatively to patients with DCIS. In HLA-A2 positive patients with DCIS, a single inoculation with NPS+GM-CSF can induce in vivo immunity and a continued antigen-specific T-cell response one month post-surgery. This data provides support for further testing of NPS+GM-CSF in the neoadjuvant and adjuvant settings in an attempt to prevent invasive recurrence in DCIS.