CPG 7909, a 24-mer B-Class CpG oligodeoxynucleotide (ODN), was tested for safety, tolerability and its ability to augment the immunogenicity of a commercial trivalent killed split influenza vaccine (Fluarix^® containing A/Beijing/262/95, A/Sydney/5/97 and B/Harbin/7/94; SmithKline Beecham) in a phase Ib blinded, randomized, controlled clinical trial. Sixty healthy volunteers were recruited in two consecutive cohorts of 30 subjects, who were randomly assigned to receive Fluarix plus 1mg CPG 7909 or Fluarix plus saline control (15 subjects each). Vaccines were administered by intramuscular injection on a single occasion with subjects in the first cohort receiving a 1/10th dose of Fluarix and those in the second cohort receiving the full-dose. All safety measures including physical evaluation, laboratory blood assays, and assays for DNA autoimmunity were within normal values except for transient and clinically inconsequential decreases in total white blood cell counts in groups receiving CPG 7909. All vaccines were found to be generally well tolerated with similar frequency and intensity for most adverse reactions for groups receiving CPG 7909 as controls. Exceptions were injection site pain and headache, which were reduced in frequency in subjects receiving the 1/10th Fluarix dose without CpG, compared to the frequency in all other groups. There was a lack of pre-existing immunity, defined as hemagglutinin inhibition (HI) activity ≤20, for all subjects to the influenza strains A/Beijing/262/95 and B/Harbin/7/94 and for some subjects to A/Sydney/5/97. Post-vaccination humoral immune responses, as determined 2 and 4 weeks later by assay of HI activity and ELISA to detect antibodies against hemagglutinin (anti-HA) were similar for both full and reduced Fluarix doses but the cellular immune responses (measured as PBMC antigen-specific IFN-γ secretion) were reduced in the 1/10th Fluarix dose group. Humoral responses were not significantly enhanced by the addition of CPG 7909, except in individuals with pre-existing immunity to A/Sydney/5/97 strain (baseline HI activity titre >20), where there was a trend to higher HI activity with CPG 7909 (P=0.06). The addition of CPG 7909 to the 1/10th dose of Fluarix did however result in significantly higher levels of IFN-γ secretion from peripheral blood mononuclear cells recovered at 4 weeks and restimulated ex vivo with A/Beijing/262/95 (P=0.048) and B/Harbin/7/94 (P=0.0057), restoring these to the level seen with full-dose vaccine. These results suggest that addition of CPG 7909 to Fluarix may allow the use of reduced vaccine doses without reduced immunogenicity.