BACKGROUNDWe assessed the 10-year efficacy, immunogenicity and safety of two doses of a combined measles-mumps-rubella-varicella vaccine (MMRV) or one dose of a monovalent varicella vaccine (V) in children from Czech Republic, Lithuania, Poland, Romania and Slovakia.METHODSThis was a phase IIIB follow-up of an observer-blind, randomized, controlled trial (NCT00226499). In phase A, healthy children aged 12-22 months from 10 European countries were randomized in a 3:3:1 ratio to receive two doses of MMRV (MMRV group), one dose of MMR followed by one dose of V (MMR + V group), or two doses of MMR (MMR; control group), 42 days apart. Vaccine efficacy (VE) against varicella (confirmed by viral DNA detection or epidemiological link and clinical assessment) was calculated with 95% confidence intervals using Cox proportional hazards regression model. Immunogenicity was assessed as seropositivity rates and geometric mean concentrations (GMCs). Solicited and unsolicited adverse events (AEs) and serious AEs (SAEs) were recorded.RESULTSA total of 3705 children were vaccinated (1590, MMRV group; 1586, MMR + V group; 529, MMR group). There were 663 confirmed varicella cases (47, MMRV group; 349, MMR + V group; 267, MMR group). VE ranged between 95.4% (Lithuania) and 97.4% (Slovakia) in the MMRV group and between 59.3% (Lithuania) and 74% (Slovakia) in the MMR + V group. At year 10, seropositivity rates were 99.5%-100% in the MMRV group, 98%-100% in the MMR + V group and 50%-100% in the MMR control group, and the anti-VZV antibody GMCs were comparable between MMRV and MMR + V groups. The occurrence of solicited and unsolicited AEs was similar across groups and no SAE was considered as vaccination-related. No new safety concerns were identified.CONCLUSIONSOur results indicated that two doses of varicella zoster virus-containing vaccine provided better protection than one dose against varicella and induced antibody responses that persisted 10 years post-vaccination.