Few data exist on long-term persistence of measles antibodies after vaccination of West African infants. The data that do exist indicate that the antibody titres decline very rapidly. Our data would be the first to describe the persistence of measles antibodies after two doses of measles vaccine, and the study would allow us to identify unprotected children and offer them revaccination. Since persistence of measles antibodies is of crucial importance to measles control, the study will contribute significantly to the existing knowledge and might have important implications for future eradication programmes.

Objective To determine persistence of measles antibodies among children who received either one or two doses of Edmonston-Zagreb (EZ) or Schwarz (SW) measles vaccine.

Background The World Health Organization has targeted measles for eradication by the year 2010, and although measles incidence has fallen drastically in many parts of the world, several factors could hinder the eradication goal.

First, there is no indication of improved measles immunisation coverage world wide, vaccination coverage fell in all regions of the world from 1997 to 1998, except in the Western Pacific, so the general measles immunisation coverage came down to 72%. Very high vaccination coverage of ≥ 95% is needed to interrupt transmission of the highly contagious measles virus.

Primary and secondary vaccine failure constitute another significant problem to measles control, and the HIV pandemic contributes to increased vaccine failure, and permits transmission of measles virus despite high rates of immunisation coverage.

Sub-clinical measles has been found to contribute to measles immunity by boosting of the antibody level, and with less circulation of wild measles virus secondary vaccine failure may represent a special problem in terms of waning immunity, a problem which is probably more pronounced among those vaccinated early, but there are still few data relevant to this problem.

Thus, although effective measles vaccines are available, there is still a need to find the optimal way of immunising in different epidemiological settings. In areas with high measles transmission early two-dose measles vaccination schedules have been recommended for prevention of measles in the age group below the normal age of vaccination, and as a means of raising vaccination coverage in general. We conducted such a two-dose trial with two different strains of measles vaccine in Guinea-Bissau, and found that the risk of not being vaccinated was lower in the two-dose group than in the one-dose group, and the relative efficacy of a two-dose versus a one-dose schedule was high among children below the normal age of vaccination. Further, we found that both one or two doses of the EZ vaccine resulted in 1% of the children being unprotected at 18 months of age, while one or two doses of SW resulted in 3% and 9% unprotected, respectively. The EZ vaccine, but not the SW vaccine, was able to boost the antibody response significantly after revaccination at 9 months of age in children with moderate levels of antibodies.

We therefore propose to study long-term persistence of measles antibodies in this cohort where we found rather striking differences in protection at 18 months of age according to vaccine strain and number of doses to find out whether the measles antibody level is maintained over time or waning immunity is a problem, since this could have important implications for global measles control and elimination.

Methods With 85% seroconverters in the one-dose group we would be able to detect a difference of 10% in number of non-seroconverters with a sample size of 400 children to be blood sampled in each group.

Among 6,900 children with an 18 months blood sample (10% loss to follow-up between 6 and 18 months of age), we will include all the children who received the EZ vaccine to be blood sampled at 6-7 years of age. Available are 450 samples from 18 months of age, and with 20% loss to follow-up we will have a total of 360 children in this group.

Among children who received the SW vaccine we plan to follow up half of the children at 6-7 years of age, and the other half 3 years later at 9-10 years of age. Including the first 2,000 children with an 18 months blood sample we will be able to include about 1,600 children in the study with 20% loss to follow-up. Thus we will have 400 children in each arm of the study (one dose / two dose) at 6-7 and at 9-10 years of age.

Children with unprotective measles antibody level will be offered revaccination.

Inclusion Criteria: Children participating in the Two-dose trial (Non-Specific Effects of Standard Titre Measles Vaccination, Protocol ID: IC18-CT95-0011-Twodose1)

Exclusion Criteria: Severe illness requiring hospitalisation