The year 2019 had its share of disappointing drug trials in Alzheimer disease, leading many to question whether anti-amyloid therapies were the optimal strategy. But that view may have shifted with controversial developments later in the year.
In January, the phase III CREAD-1 and -2 trials of crenezumab, an anti-amyloid monoclonal antibody that primarily targets soluble oligomers, were halted. An interim analysis showed no effect on cognitive decline in patients with prodromal or mild AD. CSF and PET substudies showed a nonsignificant decline in total tau, p-tau181 and amyloid load. A prevention trial in patients with the Presenilin1 E280A mutation is ongoing in Columbia.
Other casualties during the year were four beta-site amyloid precursor protein-cleaving enzyme (BACE)-1 inhibitors. A trial of verubecestat in prodromal AD was stopped early after an interim analysis found that placebo performed better than active therapy on the estimated rate of AD-associated progression (Egan et al. N Engl J Med 2019;380:1408-1420). A prior trial in mild to moderate AD was also stopped early (Egan et al. N Engl J Med 2018;378:1691-1703).
Worsening cognition was also seen in the GENERATION II/III trials of umibecestat, and in the MISSION trials of elenbecestat. Then in November, the AMARANTH and DAYBREAK-ALZ trials of lanabecestat were terminated when treatment failed to slow cognitive or functional decline in patients with early or mild AD (Wessels et al. JAMA Neurol 2019; epublished November 25, 2019).
With the repeated failure of amyloid-directed therapies, there was little surprise when the phase III development program was halted for aducanumab. An interim analysis of the EMERGE and ENGAGE phase III trials indicated that the MAb was unlikely to meet its primary endpoint. That decision was reversed in October when the manufacturer, Biogen, announced to investors that a re-analysis had demonstrated efficacy, and this would be the basis for an FDA submission.
EMERGE and ENGAGE randomized patients to placebo or aducanumab 6 mg/kg (3 mg/kg if an ApoE4 carrier) or 10 mg/kg (6 mg/kg if an ApoE4 carrier) administered every 4 weeks for 78 weeks. A protocol amendment midway through the trial allowed ApoE4 carriers to receive the 10 mg/kg dose. The planned sample size was later increased from 1,350 to 1,650 for both trials. The futility analysis of completers (n=1748 by December 26, 2018) was conducted in March 2019 and showed that treatment was unlikely to reach the primary endpoint of a significant change in Clinical Dementia Rating-Sum of Boxes (CDR-SB) score.
Two new analyses were then done. The first included 318 patients who completed the trials from the cut-off date for the interim analysis until trial termination (December 2018 to March 2019). With this larger sample size, there was a 23% difference in CDR-SB with aducanumab versus placebo in the EMERGE trial; the ENGAGE trial did not demonstrate statistical significance. Aducanumab also slowed functional decline by 40 percent in EMERGE. The positive results in EMERGE were attributed to greater cumulative drug exposure. Patients were enrolled earlier in ENGAGE, so fewer participants could receive the higher dose permitted with the mid-trial protocol amendment.
The second analysis compared all patients in the two trials who received the highest treatment exposure (i.e. 10 mg/kg over 78 weeks, n=263) versus placebo (n=1093). High-dose aducanumab achieved significance versus placebo for the primary endpoint in both studies. Thus, the planned FDA submission would include one positive trial result (EMERGE) and supporting evidence from a subgroup analysis. This attracted its share of discussion and debate (For an editorial see Schneider L. Lancet Neurol, epublished December 4, 2019 www.thelancet.com/action/showPdf?pii=S1474-4422%2819%2930480-6).
Additional details were presented at the 12th Clinical Trials on Alzheimer’s Disease conference in San Diego in December, and reported by AlzForum (www.alzforum.org/news/conference-coverage/exposure-exposure-exposure-ctad-aducanumab-scientists-make-case). The median cumulative aducanumab exposure was 116 mg/kg for patients enrolled prior to the protocol amendment compared to 153 mg/kg for those enrolled afterward (31% increase). Overall, 29% of patients in EMERGE and 22% of patients in ENGAGE received the maximum treatment exposure (10 mg/kg over 78 weeks). For the subgroup analysis, the mean change from baseline in CDR-SB score with aducanumab versus placebo was reduced 30% in EMERGE (1.23 vs. 1.76) and 27% in ENGAGE (1.31 vs. 1.79). Data on secondary endpoints (MMSE, 13-item ADAS-cog, ADCS ADL MCI) were not presented. Biomarker studies showed a dose-dependent effect on plaque reduction, and reductions in total tau and p-tau. Some have speculated that the results may indicate that there is a therapeutic lag between plaque removal and clinical and functional benefit.
The incidence of amyloid related imaging abnormalities (ARIA) was 35.1% with 10 mg/kg, 25.8% with 6 mg/kg and 2.3% with placebo for edema. In addition, 17.5% developed ARIA-H (microhemorrhages) versus 6.8% of controls. Overall, 40.7% of patients on high-dose treatment developed some form of ARIA compared to 31.7% on 6 mg/kg and 10% of controls.
While the decision to proceed with aducanumab has been controversial, it may serve to rekindle interest in anti-amyloid strategies. Other MAbs currently in phase III testing are BAN2401 and gantenerumab. The Scarlet RoAD trial of gantenerumab was terminated early (Ostrowitzki et al. Alzheimers Res Ther 2017;9:95), but a recent PET analysis reported substantial plaque removal at two years (Klein et al. Alzheimers Res Ther 2019;11:101). Two phase III trials of high-dose gantenerumab (GRADUATE 1 and 2) are ongoing.