Vaccine therapy may to destroy the current ductal carcinoma in situ (DCIS), prevent recurrence, and have little risk of long-term complications. One of the limitations for a DCIS vaccine is identifying appropriate antigens which, when targeted, may be able to treat DCIS and prevent development of IBC. In this study, we identified proteins that were overexpressed in both DCIS and IBC tumors across fifteen human Geo and Array Express data sets. From the 68 candidate proteins, twelve proteins expression was necessary for cancer cell survival across breast cancer subtypes using a high throughput siRNA screen. We chose candidates that increased apoptosis and decreased cell survival in human HER2 positive (HER2), triple negative (TN), and hormone receptor positive HER2 negative (HR) human breast cancer cell lines with decreased expression of the target protein. Twelve proteins (AURKA, KIF11, NDC80, RRM2, SDC1, UBE2C, HJURP, CENPA, CENPF, HIST2H2AA3, KRT8, and TOP2A) were the twelve antigens developed for the vaccine. All twelve targets were immunogenic in women with DCIS and IBC as compared to age matched control women. For example, autoantibodies to four of the antigens (AURKA, NDC80, KRT8, and RRM2) predicted women with DCIS with AUC 0.69 (p=0.005 95% CI 0.56 to 0.80) and IBC with AUC 0.79 (p<0.001 95% CI 0.66 to 0.90) as compared to age matched control women in a discovery set of 59 DCIS patients, 37 IBC patients, and 43 control women. In an independent validation set of serum samples of 50 DCIS, 60 IBC, and 50 control patients, autoantibodies to the four candidate antigens could predict women with DCIS with AUC 0.62 (p=0.04, 95% CI 0.51 to 0.73) and IBC with AUC 0.69 (p<0.001, 95% CI 0.60 to 0.79). Autoantibodies to the four DCIS antigens could identify each subtype of breast cancer subtype as compared to control, HR+(n=20 patients) with AUC 0.69 (p=0.01, 95% CI 0.55 to 0.83), HER2+ (n=20 patients) with AUC 0.70 (p=0.01, 95% CI 0.55 to 0.84), and TN (n=20 patients) with AUC 0.74 (p=0.002, 95% CI 0.61 to 0.87). All these were IgG autoantibodies suggest pre-existing T cells that recognize the candidate antigens because Ig class switching from IgM to IgG requires antigen-specific T cells. These twelve candidate antigens, overexpressed in human breast cancer, are also overexpressed in the TgMMTV-neu and C3(1)Tag transgenic mouse mammary tumor models of breast cancer. Vaccination with human MHC Class II predicted epitopes from KRT8, conserved between mouse and human, inhibited tumor growth in TgMMTV-neu mice by 47% and C3(1)Tag by 62%. The KRT8- specific IFN-g immune response developed was associated with lower tumor volume with mice vaccinated with the human MHC Class II Th1 KRT8 epitopes with the KRT8-specific IFN-g immune response associated with smaller tumors in TgMMTV-neu mice (R2= 0.64, p=0.02). We plan to develop these antigens in a multi-antigen DCIS vaccine that may be able to destroy DCIS and prevent breast cancer across all breast cancer subtypes.

Citation Format: Sasha E. Stanton, Jason Schlumbohm, Mary L. Disis. Candidate antigens for a ductal carcinoma in situ vaccine, essential for breast cancer cell survival across multiple subtypes, are immunogenic in DCIS and IBC [abstract]. In: Proceedings of the AACR Special Conference on Rethinking DCIS: An Opportunity for Prevention?; 2022 Sep 8-11; Philadelphia, PA. Philadelphia (PA): AACR; Can Prev Res 2022;15(12 Suppl_1): Abstract nr PR007.