Sola fails to wow, but does sustain hope
For all the hype that has surrounded the release of long-term data on solanezumab today, the actual event has been less than thrilling. At best, the “delayed-start” analysis of patients in a safety extension trial shows that immediately beginning treatment with the Alzheimer’s disease agent is better than waiting, but it does so inconsistently across measures and time points.
The best that can be said is that this analysis ought not to discourage those who believe that clearing soluble beta amyloid could slow the progression of Alzheimer’s. However, these data do not come close to foretelling success in the Expedition 3 trial.
One has to look very hard at the data to see the glimmers of hope, but they are there. The analysis looked at the subgroup of mildly impaired patients who completed the placebo-controlled portions of the failed Expedition 1 and 2 trials, half of whom had been in the placebo arm and half on solanezumab.
Placebo patients were crossed over to solanezumab treatment, to determine whether differences seen at the end of the Expedition trials would be sustained, and thus demonstrate an effect on disease progression. This type of study is a new development in neurodegenerative disease, having been used in an attempt to show a disease-modifying effect for Teva’s Azilect in Parkinson’s disease.
On two measures – ADAS-Cog14 and ADCS-IADL – a numerical difference between the early and delayed starts was preserved at most time points 108 weeks after the beginning of the trials, or 28 weeks after the delayed-start patients began, meeting non-inferiority criteria. This was the primary endpoint of the analysis; however, these treatment differences faded with passing time as the analysis included data to 184 weeks.
Two other measures – the mini-mental state examination (MMSE) and clinical dementia rating-sum of boxes (CDR-SB) – did not meet significance at 108 weeks, but did at later time points. On CDR-SB, the early-start arm met non-inferiority at 108, 160 and 184 weeks, but the investigators acknowledged that this was not a robust finding because there was no significant difference at the end of the placebo-controlled portion of the trial.
The longer-term data were limited by the number of dropouts, providing fewer patients across which to measure an effect. Just 581 of the more than 1,300 patients who had entered this study were still actively taking solanezumab at the 184-week time point.
And, of course, the analysis is limited by the fact that there was no pure placebo comparator.
Science vs traders
For those deep in the science of Alzheimer’s disease, this was a classic Rorschach test – supporting whatever belief one might have, positive or negative, on solanezumab and indeed on the beta amyloid hypothesis.
For those trading on the data, however, the signal was different. The market had driven Lilly’s shares to a 15-year high of $88.80 earlier this month, but it looks like cagey investors sold their pumpkin futures before Halloween. The stock was down 3% to $82.72 in early trading today.
The damage was equally bad for Biogen, whose aducanumab reported mixed and inconsistent phase I data. Its shares were off 3% to $396.90 in early trading.
In any case, the rationale for buying Lilly shares based on solanezumab has now evaporated until Expedition 3 reads out around the end of 2016. Even then, those looking for answers to the Alzheimer’s puzzle will probably find only more questions.