The effect of scheduled antibody testing on treatment patterns in interferon-treated patients with multiple sclerosis

Table 5 Relationship between antibody level throughout the study and targeted side effects for both arms combined (mITT cohort) and individual arms*

Category	Antibody level (units)	Odds ratio	95% confidence interval	p-value (odds ratio)	p-value(linear trend)
Targeted side effects for both arms combined (mITT cohort)
Flu-like symptoms in past month	NAb ≥100	0.33	(0.22, 0.48)	<0.001	<0.001
	NAb 20 to <100	0.41	(0.28, 0.60)	<0.005
	NAb <20	0.92	(0.71, 1.18)	0.50
	BAb(−)	(Ref)	—	—
Injection site reactions in past month	NAb ≥100	0.34	(0.25, 0.45)	<0.001	<0.001
	NAb 20 to <100	0.59	(0.41, 0.83)	0.003
	NAb <20	0.79	(0.62, 1.01)	0.062
	BAb(−)	(Ref)	—	—
Depression in past month	NAb ≥100	0.70	(0.50, 0.99)	0.046	0.083
	NAb 20 to <100	0.99	(0.68, 1.44)	0.95
	NAb <20	0.93	(0.71, 1.22)	0.60
	BAb(−)	(Ref)	—	—

*Repeated measures multivariable model utilized all data for which BAb/NAb testing and adverse events were assessed at the same visit. Models were adjusted for age, sex, study arm, multiple sclerosis type, current interferon, time from onset of symptoms, and centre size. Patients who were BAb(−) served as the referent, with an odds ratio calculated to represent the odds that a particular NAb level group would experience the event, compared with the referent. The linear trend p-value tested for the presence of a linear trend across BAb/NAb groups. Targeted events that did not show a statistically significant effect are not included in the table (seizures, menstrual irregularities, abnormal liver function tests, cytopenia, and decreased hemoglobin).
BAb, serum binding antibody; mITT, modified intention-to-treat; NAb, serum neutralizing antibody.

ISSN: 1471-2377