trial-specific median), race (white vs. biomarkers, and 3C5 demographic variables. The best prediction of participants durable response positivity based on twoweek reactions rendered up to close-to-perfect accuracy; the best prediction of participants durable response magnitude rendered correlation coefficients between the observed and expected reactions ranging up to 0.91. Though prediction performances differed among biomarkers, durable immune reactions were best expected from the two-week response level of the same biomarker. Adding demographic info and two-week response levels of different biomarkers offered little or no improvement in the predictions. Conclusions For some biomarkers and for the vaccines we analyzed, two-week post-vaccination reactions can well forecast durable reactions six months later on. Therefore, if immune response toughness is only assessed inside a sub-sample of vaccine recipients, statistical analyses of durable reactions will have improved effectiveness by incorporating two-week response data. Further study is needed to generalize the findings to additional vaccine regimens and biomarkers. Conclusions Clinicaltrials.gov identifiers: “type”:”clinical-trial”,”attrs”:”text”:”NCT01799954″,”term_id”:”NCT01799954″NCT01799954, “type”:”clinical-trial”,”attrs”:”text”:”NCT00820846″,”term_id”:”NCT00820846″NCT00820846, “type”:”clinical-trial”,”attrs”:”text”:”NCT00223080″,”term_id”:”NCT00223080″NCT00223080. strong class=”kwd-title” Keywords: Binding antibody multiplex array, Immunogenicity, Intracellular cytokine staining, Regularised random forest, Statistical power 1. Intro Durable vaccine-induced immune reactions are critical to minimize waning effectiveness of vaccines against varied pathogens, including HIV-1 (e.g., [1C3]). An important objective of earlyphase HIV vaccine medical trials is to evaluate and display vaccine candidates based on immune reactions measured at time points near the vaccinations (e.g., 2C4 weeks later on) and at later time points, including at an early toughness (e.g., 6C12 weeks later on) time point [3]. Although vaccination-proximal reactions are generally assessed in every trial participant, durability reactions are sometimes only assessed inside a subset of participants because of trial cost/length considerations and loss to follow-up. Incomplete data in the durability time point potentially limit statistical power for immunogenicity profiling of candidate vaccines. One viable remedy is to incorporate immune response toughness predictors into the analysis. Fadrozole For example, based on statistical methods explained in [4,5], a 25C50% effectiveness gain (i.e., sample size savings) Fadrozole can be achieved in the assessment of toughness immune reactions between two organizations if the predictor has a correlation of 0.6C0.8 with the toughness response (Supplementary eFig. 1). To investigate whether this could be applied to the analysis of HIV vaccine-induced immune reactions, we characterized predictors of cellular and humoral immune response durability using data from randomized HIV-1 vaccine medical tests, focusing on biomarkers that have been validated and/or previously used as immunogenicity endpoints. The specific objectives of this analysis were to evaluate whether and how well a participants (1) positivity and (2) magnitude of vaccine-induced cellular and humoral immune reactions at an early durability time point can be expected by that participants immune reactions of the same biomarker in the two-week post-vaccination time point. We also evaluated whether adding baseline demographic info and two-week response data from additional biomarkers improved prediction. 2. Materials and methods 2.1. Study cohorts and immune response biomarkers We analyzed data from seven vaccine regimens evaluated in three preventive HIV vaccine tests: a phase 1 study of four DNA, NYVAC and/or AIDSVAX vaccine regimens carried out in Switzerland (HVTN 096), a phase 2 study of two DNA and/or Modified Vaccinia Ankara (MVA) vaccine regimens in the USA and Peru (HVTN 205) BMP1 [6], and a phase 3 study of a single ALVAC/AIDSVAX routine in Thailand (RV144) [7] (Table 1). Final vaccinations were given at Month 6 for those regimens. We restricted analysis to vaccine recipients who received all planned immunizations and were HIVuninfected at the time of immune response assessments. For HVTN 205 and RV144, data were available from a randomly selected subset of vaccine recipients in the original studies [6,7]. Table 1 Summary of study data. thead th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ /th th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ HVTN096a /th th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ HVTN096b /th th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ HVTN096c /th th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ HVTN096d /th th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ Fadrozole HVTN205e /th th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ HVTN205f /th th valign=”bottom” align=”remaining” rowspan=”1″ colspan=”1″ RV144 /th /thead Vaccine regimenNYVAC (M 0) br / NYVAC (M 1) br / NYVAC/AIDSVAX (M 3) br / NYVAC/AIDSVAX (M 6)NYVAC/AIDSVAX (M 0) br / NYVAC/AIDSVAX (M 1) br / NYVAC/AIDSVAX (M 3) br / NYVAC/AIDSVAX (M 6)DNA (M.