DNA vectors have been widely used as a priming of poxvirus vaccine in prime/boost regimens. Whether the number of DNA impacts qualitatively or quantitatively the immune response is not fully explored. With the aim to reinforce T-cell responses by optimizing the prime-boost regimen, the multicentric EV03/ANRS VAC20 phase I/II trial, randomized 147 HIV-negative volunteers to either 3xDNA plus 1xNYVAC (weeks 0, 4, 8 plus 24; n = 74) or to 2xDNA plus 2xNYVAC (weeks 0, 4 plus 20, 24; n = 73) groups. T-cell responses (IFN-γ ELISPOT) to at least one peptide pool were higher in the 3xDNA than the 2xDNA groups (91% and 80% of vaccinees) (P = 0.049). In the 3xDNA arm, 26 (37%) recipients developed a broader T-cell response (Env plus at least to one of the Gag, Pol, Nef pools) than in the 2xDNA (15; 22%) arms (primary endpoint; P = 0.047) with a higher magnitude against Env (at week 26) (P<0.001). In both groups, vaccine regimens induced HIV-specific polyfunctional CD4 and CD8 T cells and the production of Th1, Th2 and Th17/IL-21 cytokines. Antibody responses were also elicited in up to 81% of vaccines. A higher percentage of IgG responders was noted in the 2xDNA arm compared to the 3xDNA arm, while the 3xDNA group tended to elicit a higher magnitude of IgG3 response against specific Env antigens. We show here that the modulation of the prime strategy, without modifying the route or the dose of administration, or the combination of vectors, may influence the quality of the responses.