Influenza B viruses from 2 lineages cocirculate annually. Because the single B strain contained in trivalent vaccines may not match the major circulating strain, adding a second B virus could enhance protection. This study compared the safety and immunogenicity of an investigational quadrivalent Ann Arbor strain live attenuated influenza vaccine (Q/LAIV) with that of 2 trivalent vaccines (T/LAIV), each containing a B strain from a different lineage. This randomized, double-blind study was designed to demonstrate the immunologic noninferiority of Q/LAIV compared with T/LAIV in children 2-17 years of age by comparing postdose geometric mean titers of hemagglutination inhibition antibodies. Children were randomized 3:1:1 to receive Q/LAIV or 1 of 2 T/LAIV vaccines. Those subjects who were 9-17 years of age received 1 dose, and those 2-8 years of age received 2 doses 1 month apart. Serum immune responses were evaluated 1 month after dose 1 (dose 2 for influenza vaccine-naive subjects aged 2-8 years). Q/LAIV was noninferior to T/LAIV: upper bounds for all four 95% confidence intervals for the postdose geometric mean titer ratios (T/LAIV divided by Q/LAIV) were ≤1.5, the predefined noninferiority margin. The overall seroresponse rates (4-fold rise) were comparable between treatment groups. Safety events were comparable, except that fever was more common after dose 1 in Q/LAIV subjects (5.1%) than in T/LAIV subjects (3.1%) 2-8 years of age. The immunogenicity of Q/LAIV was noninferior to that of T/LAIV in children aged 2-17 years; safety was also comparable. Q/LAIV may broaden the protection against influenza B strains provided by current trivalent influenza vaccines.