At an industry conference yesterday, Biogen (BIIB) said the company has decided to add 510 patients to the Phase 3 aducanumab trials after seeing more “variability” than expected in a sample size re-estimation, sending its shares into negative territory. William Blair analyst Matt Phipps, however, argued that an enrollment increase is not unusual, with his peer at Piper Jaffray recommending investors buy Biogen shares amid the weakness.

INCREASE ‘PRUDENT, NOT UNUSUAL’: Commenting on Biogen’s announcement, William Blair’s Phipps told investors that the increase in variability in and of itself does not preclude trial success. Further, the analyst said he believes increasing enrollment to ensure appropriate powering and confidence interval error margin is “prudent and not unusual.” The trial is sufficiently powered to detect a benefit significantly below what was observed in the Phase Ib trial, he contended. Phipps also pointed out that he sees “very limited” downside at current price levels if aducanumab were to fail, and added that success of Biogen’s growing earlier stage pipeline has the potential to provide additional upside. The analyst reiterated an Outperform rating on the stock. Also keeping an Overweight rating on Biogen’s shares, Piper Jaffray analyst Christopher Raymond recommended using the selloff in the stock as a buying opportunity. While it is “a bit unnerving,” particularly on the heels of another Phase 3 Alzheimer’s failure for Merck’s (MRK) verubecestat, the study remains blinded and variability on the primary endpoint can stem from a host of reasons, not all of them bad, Raymond told investors in a research note of his own. The analyst pointed out that variability in CDR-SB could be caused by a number of factors, such as a heterogeneous patient population, which would not necessarily doom it to failure either, or potentially a larger effect size, among other possibilities. Overall, Raymond does not think the news says much about the chances for an interim analysis or about aducanumab’s efficacy profile.