FDA Summary Review: Avonex (Interferon beta-1a) (PLA: 95-0979; ELA: 95-0975)
-”Interferon beta-1a (AVONEX) is indicated for the treatment of relapsing forms of multiple sclerosis to slow the accumulation of physical disability and decrease the frequency of clinical exacerbations”
-”…the primary endpoint includes 123 placebo patients and 151 interferon patients for time to confirmed progression. There were 60 confirmed progressions during the study; 36 in the placebo, 24 in the interferon group. The analysis was performed with Kaplan-Meir survival curves and the Mantel-Cox statistic with a log-rank test…These curves show a decrease in the number of progressions in interferon group (p = 0.024)”
-time to first exacerbation: Exacerbation within the first three months on-study occurred nearly equally in both arms (26% placebo, 24% interferon)
-Distribution of Exacerbations per Patient: Although there was a modest trend to fewer exacerbations in the interferon group in the first-year-on-study date, there was no statistically significant difference between the groups”
-For the subset of patients with at least 2 years on study (85 interferon, 87 placebo) there were more with no exacerbations in the interferon group (38%) than the placebo group (26%), and fewer with more than one exacerbation (32%) than the placebo group (3%). This was statistically significant (p = 0.027), Mann-Whitney rank sum test)
- The sponsor suggested a lag time between initiation of treatment and onset of clinical benefit. Thus exacerbations are more appropriately assessed over a full 2 years. However, this analysis discards considerable patient on-study experience. The two-years-on-study analysis discards all patients who did not reach 2 full years on-study, ie, all patients enrolled in the later portion of the study. An advantage of these analyses is an equality of patient weighting due to equal time contributed.
- To address this, the data were reanalyzed in groups that still maintained the equality of patient followup time, but allowed an exploratory comparison of the effects of treatment. The data were divided into three subset, each describing patient experience over a one-year period, with a full year of experience for each patient included (Table 2)
-Distributions of Exacerbations per Patient
- 1 year patient subset: p value: 0.664
- 2 year patients during year 1: p value: 0.112
- 2-year patients during year 2: p-value: 0.096
-There is no indication that benefit from interferon is delayed after treatment initiation
-Exacerbation Rate: Calculation of exacerbation rate on-study was not a prospectively defined endpoint, it was requested by CBER at the time of a pre-PLA meeting, ,due to the widespread use of this value for entry criteria or for an endpoint in MS clinical trials
-annualized exacerbation rates: 2 years: p value: 0.002
-steroid treatment course
- Year 1 patient subset: p value: 0.295
- Year 1 and 2 Patient subset: p value: 0.010
-MRI Gadolinium Lesion Endpoints
- Number of Gadolinium-enhancing lesions per patient
- Year 1 patient subset: p value: 0.024
- 2 Year patient subset: p value: 0.051
- Volume of Gadolinium-enhancing lesions per patient, mm cubed
- Year 1 patient subset: p value: 0.020
- 2-Year patient subset: p value: 0.010
-T2-weighted MRI Lesion Volume
- Baseline: p value: 0.143
- Year 1: p value: 0.023
- Year 2: p value: 0.355
-These numerous analyses (of limb functional tests?) failed to show statistically significant differences between arms except for the time to sustained progression in the 25 Foot Walk time
-the neuropsychologic batteries did not show differences in the treatment arms
-the primary endpoint was progression of disability, as measured by a 1.0 change in the EDSS, sustained for at least 6 months to lessen effects of variability from disease fluctuation and intrarater variablity
-MRI T2 lesion volume showed statistically different effects at the end of year but not year 2 scans (bold and underline mine)
-Exacerbation rates per year per patient were similar. They were statistically different for the 2-year patient subset but not for the 1-year patient subset. When all patients on-study time was analyzed there was a treatment effect (bold and underline mine)
-The distribution of number of exacerbations per patient improved with interferon treatment in the subset of patients on-study for at least two years but was not significant when analyzed using the larger number of patients on-study for at least 1 year (bold and underline mine)
-MRI T2 lesion volume showed statistically different effects at the end of year 1 but not year 2 scans (bold and underline mine)