Background. @nbsp; To protect a naive global population against pandemic influenza, pandemic vaccines should be effective at low antigen doses, because of limited manufacturing capacity. Methods. @nbsp; In a multicenter, randomized, blind-observer phase 1 trial, groups of 50 healthy young adults received 2 doses, 21 days apart, of influenza A/Vietnam/1194/2004 NIBRG-14 (H5N1) vaccine containing 1.9, 3.8, 7.5 or 15 mug of hemagglutinin with oil-in-water emulsion adjuvant or 7.5 mug of hemagglutinin without adjuvant. Safety was monitored to day 42. Homologous hemagglutination-inhibition (HI) and microneutralization titers were determined after each vaccination. Cross-reactivity against A/Indonesia/05/2005 RG2 was tested after the second vaccination. Results. @nbsp; No vaccine-related significant or serious adverse events occurred. Injection site reactions, but not systemic reactions, were more frequent with adjuvant than without. Even with only 1.9 mug of hemagglutinin plus adjuvant, 72% of subjects had HI titers >/=1:32 after 2 doses. This proportion was 81%-89% with higher adjuvanted doses but was only 34% without adjuvant. Adjuvanted vaccine induced cross-neutralizing antibodies in 39%-65% of samples, versus 7% without adjuvant. Conclusions. @nbsp; The emulsion-adjuvanted pandemic influenza vaccine candidate was safe, immunogenic, and induced cross-reactive antibodies. This adjuvanted 1.9-mug candidate is the lowest effective dose tested to date. This could have a major impact on prepandemic vaccination strategies with stockpiled batches of vaccine. Trial registration. @nbsp; ClinicalTrials.gov identifier: NCT00457509 .