Thomas Hickey ¹ Uma Mudunuri ² Heidi Anne Hempel ¹ Troy Kemp ¹ Nancy Roche ¹ Keyur Talsania ² Brian A Sellers ³ James M Cherry ³ Ligia A Pinto ^1*

• ¹ Vaccine, Immunity, and Cancer Directorate, Frederick National Laboratory for Cancer Research, National Cancer Institute at Frederick (NIH), Frederick, United States
• ² Advanced Biomedical Computational Science, Frederick National Laboratory for Cancer Research, National Cancer Institute at Frederick (NIH), Frederick, United States
• ³ Center for Human Immunology, Inflammation and Autoimmunity, National Institute of Allergy and Infectious Diseases (NIH), Bethesda, Maryland, United States

The first vaccines approved against SARS-CoV-2, mRNA-1273 and BNT162b2, utilized mRNA platforms. However, little is known about the proteomic markers and pathways associated with host immune responses to mRNA vaccination. In this proof-of-concept study, sera from male and female vaccine recipients were evaluated for proteomic and immunologic responses 1 and 6 6 months following homologous third dose vaccinationadministration. An aptamer-based proteomic assay coupled with traditional serology was leveraged to generate a comprehensive evaluation of systemic responsiveness in 66 64 and 68 healthy recipients of mRNA-1273 and BNT162b2 vaccines, respectively. Sera from female recipients of mRNA-1273 recipients showed up-regulated indicators of inflammatory and immunological responses at 1-month post--third vaccination, and sera from female recipients of BNT162b2 demonstrated up-regulated negative regulators of RNA sensors at 1-month. Sera from male recipients of mRNA-1273 showed no significant up-regupregulation of pathways at 1--month post post-third vaccination, though there were multiple significantly upregupregulated proteomic markers. Sera from male recipients of BNT162b2-vaccinated males demonstrated up-regupregulated markers of immune response to double-stranded RNA and cell-cycle G(2)/M transition at 1-month. Random Forest analysis of proteomic data from pre-third third-dose sera identified 85 markers used to develop a model predictive of robust or weaker IgG responses/antibody levels to SARS-CoV-2 spike protein at 6--months following boost; no specific markers were individually predictive of 6-month IgG response. Thirty markers that contributed most to the model were associated with complement cascade/activation; IL-17, TNFR pro-apoptotic, and PI3K signaling; and cell cycle progression. These results demonstrate the utility of proteomics to evaluate correlates or predictors of serological responses to SARS-CoV-2 vaccination.