Cause of death in MS: long-term follow-up of a randomised cohort, 21 years after the start of the pivotal IFNβ-1b study.

Objectives: Compared with controls, multiple sclerosis (MS) patients die, on average, 7-14 years prematurely. Previously, we reported that, 21 years after their participation in the pivotal randomised, controlled trial (RCT) of interferon β-1b, mortality was reduced by 46-47% in the two groups who received active therapy during the RCT. To determine whether the excessive deaths observed in placebo-treated patients was due to MS-related causes, we analysed the causes-of-death (CODs) in these three, randomised, patient cohorts.

Design: Long-term follow-up (LTF) of the pivotal RCT of interferon β-1b.

Setting: Eleven North American MS-centres participated.

Participants: In the original RCT, 372 patients participated, of whom 366 (98.4%) were identified after a median of 21.1 years from RCT enrolment.

Interventions: Using multiple information sources, we attempted to establish COD and its relationship to MS in deceased patients.

Primary Outcome: An independent adjudication committee, masked to treatment assignment and using prespecified criteria, determined the likely CODs and their MS relationships.

Results: Among the 366 MS patients included in this LTF study, 81 deaths were recorded. Mean age-at-death was 51.7 (±8.7) years. COD, MS relationship, or both were determined for 88% of deaths (71/81). Patients were assigned to one of nine COD categories: cardiovascular disease/stroke; cancer; pulmonary infections; sepsis; accidents; suicide; death due to MS; other known CODs; and unknown COD. Of the 69 patients for whom information on the relationship of death to MS was available, 78.3% (54/69) were adjudicated to be MS related. Patients randomised to receive placebo during the RCT (compared with patients receiving active treatment) experienced an excessive number of MS-related deaths.

Conclusions: In this long-term, randomised, cohort study, MS patients receiving placebo during the RCT experienced greater all-cause mortality compared to those on active treatment. The excessive mortality in the original placebo group was largely from MS-related causes, especially, MS-related pulmonary infections.