BACKGROUND: Vaccination against hepatitis A virus (HAV) is unaffordable to many developing countries. Substantial reductions in cost occur when vaccines are administered intradermally at low doses. Aluminum-free HAV vaccines are considered more suitable for intradermal use than traditional vaccines which can cause long-lasting local reactions. Thus, we compared the immunogenicity and safety of an aluminum-free virosomal HAV vaccine (Epaxal) administered by different routes: intradermal (i.d.), subcutaneous (s.c.), and intramuscular (i.m.). METHODS: Two open pilot studies were conducted as sub-studies of a large lot consistency trial. Healthy subjects aged 18 to 45 were enrolled. Study 1 compared two i.d. regimens of a lower dose of Epaxal [0.1 mL (4.8 IU), one or two injection sites] with i.m. administration of the standard dose [0.5 mL (24 IU)]. Study 2 compared the s.c. with the i.m. administration of the standard dose. At month 12, subjects in study 1 received a booster dose of 0.1 mL i.d. or 0.5 mL i.m.; subjects in study 2 received 0.5 mL via the respective route (s.c. or i.m.). Serum was tested for antibodies at baseline, 2 weeks (study 1), and 1 and 6 months after the primary vaccination as well as prior and 1 month after the booster dose. Incidences of solicited and unsolicited adverse events were recorded. RESULTS: Seroprotection rates (anti-HAV geometric mean concentration of > or =20 mIU/mL) after 1 month ranged from 93.2% to 100% in all groups and remained high until month 12 (range 85.2&-90.2%). Complete (100%) seroprotection was achieved by all subjects in all groups after booster vaccination. All routes of administration were well tolerated. Local reactions were more common in subjects vaccinated i.d. and s.c. than i.m. CONCLUSIONS: The aluminum-free virosomal HAV vaccine Epaxal is highly immunogenic and well tolerated when administered either via i.d., s.c., or i.m. Vaccination via the i.d. route may confer significant cost savings over the conventional i.m. route.