We analysed the role of dosage, route and frequency of administration of cl
inical grade interferon-beta (IFN-beta) preparations in inducing anti-IFN-b
eta antibodies (IFN-beta -Abs) in 5 groups of relapsing-remitting multiple
sclerosis (RRMS) patients who were respectively treated as follows: 1) week
ly intramuscular (i.m.) injections of 30 mug of recombinant IFN-beta 1a (Av
onex (TM)), 2) subcutis (s.c.) injections of 250 mug IFN-beta 1b (Betaferon
(R)) every other day, 3) weekly i.m. injections of 250 mug IFN-beta 1b (Be
taferon (R)), 4) s.c. injections of 22 mug of IFN-beta 1a (Rebif (R)) three
times a week, and 5) i.m. injections of 22 mug of IFN-beta 1a (Rebif (R) )
twice a week. IFN-beta -Abs were determined by ELISA, IFN-beta 1b was more
immunogenic than IFN-beta 1a not only when administered s.c. every other d
ay, but also when administered i.m. at a lower weekly dose; i.m. injection,
however, significantly delayed the appearance, and induced lower serum lev
els of IFN-beta -Abs. In patients treated with s.c. IFN-beta 1b, Ab levels
peaked 3 to 9 months after therapy initiation, and then slowly, But progres
sively, declined to pre-therapy levels that in some patients were reached a
fter three years. Patients treated with i.m. or s.c. IFN-beta 1a only rarel
y developed IFN-beta -Abs, and then at very low titers, Overall, the i.m. w
eekly administration of IFN-beta 1a was the less immunogenic treatment. In
IFN-beta 1b-treated patients, a wash-out period of two/three months was suf
ficient to bring the IFN-beta -Ab levels below the cut-off. Our findings su
ggest that the immunogenicity of IFN-beta 1a is low, regardless of the rout
e of administration and the dosage, while that of IFN-beta 1b is high, and
is significantly, but not completely reduced by i.m. administration. As IFN
-beta -Abs are cross-reactive, a wash-out period is suggested when the prep
aration is changed from IFN-beta 1b to IFN-beta 1a in order to maintain the
clinical benefits of the therapy.