We have now repeated our analyses on the correctly matched data. With
one exception, there were no errors in the case/control assignment of the
samples. Several dates of blood collection, however, were erroneously assigned,
obscuring important temporal variations in antibody titers among the cases.
Overall, analyses of the correct data still support the strong and highly
significant association between anti–EB nuclear antigen titers and risk
of MS as we originally reported, but it has become apparent that this association
is strongly modified by age. Antibody titers of cases and controls were virtually
identical in samples collected before age 20 years. However, while titers
in controls remained constant, antibody titers of the cases increased, reaching
a maximum between ages 25 and 29 years, followed by a plateau. As discussed
in the article based on the corrected data,2 this
is a new and important finding that may be critical to the understanding of
the relation between EBV and MS.