Abstract

Objective: To prospectively validate MRI activity and neutralising anti-interferon antibody (NAb) during the first 6 months of interferon β treatment as response indicators in multiple sclerosis (MS).

Methods: Patients with relapsing–remitting MS were followed during the first 2 years of treatment. Neurological assessments were performed every 3 months or when a relapse was suspected. MRI scans performed at baseline and at 3, 4, 5 and 6 months after the start of treatment were assessed centrally for disease activity: new T2 or gadolinium enhancing T1 lesions. NAb were assessed using the MxA protein assay; positivity was defined as two consecutive titres ⩾20 NU/ml. We evaluated the predictivity of an active scan, NAb positivity, or both, during the first 6 months of treatment, on the occurrence of clinical disease activity in the following 18 months.

Results: 147 patients were assessed at 16 centres. Predictivity parameters (with confidence intervals) were as follows: active scan, sensitivity (SN) 52% (34–69%), specificity (SP) 80% (65–91%), negative predictive value (NPV) 73% (58–77%), positive predictive value (PPV) 62% (42–79%), p = 0.002; NAb positivity, SN 71% (45–88%), SP 66% (55–76%), NPV 92% (82–97%), PPV 29% (16–45%), p = 0.01; active scan and NAb positivity, SN 71% (38–91%), SP 86% (73–94%), NPV 94% (86–98%), PPV 50% (29–70%), p = 0.0003.

Conclusions: MRI activity and NAb occurrence during the first 6 months of interferon β treatment were reliable predictors of long term clinical response, particularly when combined. Patients with negative predictors showed a less than 10% risk of developing clinical activity. Patients with positive predictors showed a 50% risk of further clinical activity. These patients need to be followed carefully with further MRI and NAb tests.