If Biogen’s aducanumab reports even a 0.2–0.3 effect size in the Phase III EMERGE study primary endpoint, it could help assuage efficacy reservations in Alzheimer’s disease (AD).
However, it is currently hard to gauge if aducanumab is able to do so as data detail is scarce, with the company only releasing the trial’s clinical dementia rating — a sum of boxes (CDR-SB) endpoint’s percentage changes.
The investigators interviewed by this news service said they’ve only been shown publicly available percentage results, and so are calling for additional baseline information, subgroup data and statistical model information to put percentage data into context. Biogen announced it will be present at the Clinical Trials on AD (CTAD) meeting on 4–7 December in San Diego, California.
While a 0.2–0.3 effect size is considered modest, it could still be enough of an efficacy signal to support US Food and Drug Administration (FDA) approval, should it be statistically significant, due to unmet need in AD, investigators added.
Biogen announced on 21 March it discontinued its Phase III EMERGE and ENGAGE trials following a futility analysis that showed they were unlikely to meet their primary endpoints. However, the company made a U-turn on 22 October, announcing plans to file for aducanumab’s BLA in early 2020. The 1,638-patient EMERGE (NCT02484547) was reclassified as a positive trial after further review. While the 1,647-patient ENGAGE (NCT02477800) was still not positive, subgroup analysis of the high-dose group supported its inclusion in the BLA, along with data from earlier trials, as per a media release.
Biogen’s share price jumped 40% premarket following the 22 October announcement. Two analyst reports laud aducanumab’s approvability based on available data, estimating sales of $2.4bn to $3.6bn by 2028. Two other reports had more sceptical takes, recommending a wait-and-see approach until CTAD, but did not detail what additional information is desired. Biogen, which has a $51.9bn market cap, declined to comment.
Although the FDA loosened the approval pathway for AD therapies like aducanumab, the four experts are mixed if ENGAGE actually helps in the therapy’s approval prospects. ENGAGE’s data has inconsistencies that provide little insight as to how aducanumab would fare in AD, on top of also only having publicly available percentage data, they said. Nevertheless, the investigators are hoping for aducanumab to be approved as it would open doors for combination research, which is likely how the therapy would be used moving forward.
EMERGE percentage data lacks clarity of impact
CDR-SB effect size as low as 0.2–0.3 would be FDA-approvable should it be statistically significant, the four aducanumab investigators agreed. Effect size measures the magnitude of the effect, and in statistics an effect size of 0.2 is typically considered small. The approval bar in AD is low as no current treatment has been able to reliably improve symptoms, thus aducanumab achieving a small but statistically significant effect size in CDR-SB would cross that bar, the first investigator said. However, it’s hard to tell whether aducanumab can deliver this effect size in EMERGE given efficacy had only been presented in percentages, and baseline information is not known, the first investigator said.
In EMERGE, 547 patients in the high-dose, expanded intent-to-treat (ITT) group prior to futility analysis experienced a 23% reduction in CDR-SB (p=0.010). On the other hand, in the 340-patient, high-dose expanded ITT group after futility analysis, 23% CDR-SB reduction was also reported (p=0.31), as per a 22 October corporate presentation.
But purely reporting percentages does not translate into what that means in point changes in the primary endpoint, the second investigator added. For example, patients that have a lower baseline will show lower point improvement with the same percentage improvement, the investigator explained. The third investigator said presenting efficacy as percentages probably makes more sense to the public, though he agreed as a clinician he wanted hard numbers.
Additional detail on how each AD severity group fared—from mild to severe cognitive impairment—will also yield a clearer picture of whether the improvement was unilateral, or if certain disease severities fared better, the fourth investigator said. A broad percentage descriptor fails to capture this nuance, he explained.
Further data on how the placebo arm fared, apolipoprotein E (ApoE) carrier status and dropout rates are also needed for a better grasp of EMERGE’s efficacy, the first and fourth investigators added. ApoE data would likely instruct whether aducanumab might receive a narrow label for carriers only, since current data seems to suggest a lack of efficacy in non-ApoE4 carriers, the first investigator said. Carriers of the ApoE4 allele tend to have earlier disease onset with more severe progression than noncarriers, the first investigator explained.
Insight on what statistical model was used in EMERGE, as well as the error bars, would also be welcome to put some colour on the statistically significant primary and secondary endpoint data, the second investigator noted. However, Susanne May, PhD, professor, Department of Biostatistics, University of Washington, Seattle, noted that while more information about the statistical models would be helpful, it would not explain why expanding the data to include ITT patients would drastically reverse the results of EMERGE.
ENGAGE’s role as supportive data uncertain
Meanwhile, while ENGAGE is a negative trial, it could still harbour information to support aducanumab’s approval, the second and fourth investigators said. However, since ENGAGE’s available data is also presented in percentages, it also needs hard numbers and effect sizes to show if it could be useful supporting data, the second and fourth investigators agreed, adding that parallel improvements in CDR-SB in ENGAGE and EMERGE would support that case.
ENGAGE’s protocol changes led to inadequate patient exposure in the high-dose group, hence the failed trial, as per the company presentation. May and the first investigator said, while this may be the case, it would still not account for why ENGAGE’s low-dose arm performed better than the high-dose arm, among other inconsistencies. ENGAGE may not accurately represent aducanumab’s efficacy trends, and any pooling of ENGAGE with EMERGE might be solely done for registrational purposes only, the first investigator added. In fact, aducanumab’s positive 197-patient, Phase Ib trial, PRIME (NCT01677572), showed efficacy trends more consistent with EMERGE than ENGAGE did, he noted.
Nonetheless, the four investigators agreed, the FDA might be amenable to accepting just one positive Phase III trial in EMERGE plus compelling supporting evidence from ENGAGE. This is in spite of the expected upcoming advisory committee meeting that will skew conservative as aducanumab has not met the two requisite positive Phase III trials for approval, the first, second and fourth investigators said. This is based on the FDA’s loosening of approval for therapies targeting early-stage AD with its new guidance drafted in February 2018, they added. Still, it is not clear whether ENGAGE counts as having strong supporting data, the first and second investigator noted. The second investigator added that both ENGAGE and EMERGE being discontinued trials would further affect confidence in the data.
That said, the four investigators said they are rooting for aducanumab’s FDA approval, as it would be the first AD drug approval since 2003. Even if it delivers modest efficacy, at least researchers can start exploring its amyloid-beta inhibitor mechanism in combination with already-approved or up-and-coming therapy classes, the first investigator said. Combination therapy is probably the way to go for the amyloid beta inhibitor treatment class, but until a drug is approved, combination research cannot move forward, he added. Physicians are unlikely to turn down aducanumab even if its benefit is modest, given the lack of any therapy that is truly efficacious, the third and fourth investigators said.
by Shuan Sim in New York
Shuan Sim is a Senior Reporter for Clinical Trials Arena parent company GlobalData’s investigative journalism team. A version of this article originally appeared on the Insights module of GlobalData’s Pharmaceutical Intelligence Center. To access more articles like this, visit GlobalData.