A recent study published in Brain Communications discusses critical questions regarding the evaluation of anti-amyloid immunotherapies for the treatment of Alzheimer’s disease.
Study: Key questions for the evaluation of anti-amyloid immunotherapies for Alzheimer’s disease. Image Credit: Nefedova Tanya / Shutterstock.com
Challenges to new Alzheimer’s drugs
The accelerated approval of amyloid-lowering antibodies lecanemab and aducanumab by the United States Food and Drug Administration (FDA), assuming the decrease in amyloid-β will provide clinical benefits in Alzheimer’s disease, has set a precedent for this class of drugs. Furthermore, the breakthrough therapy status granted to donanemab and gantenerumab signaled that the approval of their biologics license applications (BLAs) could be expedited.
The decision by the U.S. Centers for Medicare and Medicaid Services to limit coverage of FDA-approved anti-amyloid immunotherapies suggests that their path to market might not be straightforward. In the present study, the authors discuss three central questions in evaluating the clinical safety and potential benefit of BLAs for anti-amyloid antibodies.
Were statistical analyses in trials appropriate and did they corroborate safety?
Peer-reviewed publications and independent evaluation of clinical trial data are not requisites for FDA approval. However, ensuring that relevant statistical analyses are performed is vital to support efficacy claims.
The primary outcomes of the lecanemab and aducanumab phase III trials have been inconsistent with previous studies. Importantly, amyloid-related imaging abnormalities (ARIAs) are adverse events reported following treatment with amyloid-lowering immunotherapies.
The extent to which ARIAs result in functional unblinding and differences in loss to follow-up should be assessed. Recent phase II or III trial publications on amyloid-lowering antibodies included little details of the statistical analyses that were used to address the potential confounding due to functional unblinding. In phase III trials on aducanumab, around 41% of patients in one of the dosage groups exhibited ARIAs.
In a phase III trial on lecanemab, ARIA with cerebral edema/effusion (ARIA-E) or microhemorrhages/hemosiderosis (ARIA-H) occurred in 12.6% or 17.3% of lecanemab recipients as compared to 1.7% or 9% of placebo subjects, respectively. Moreover, over 25% of lecanemab recipients developed infusion-related reactions relative to 7.4% of placebo recipients. Sensitivity analyses were not performed in these studies to account for unblinding.
The phase II trial on donanemab revealed that 27.5% and 30.5% experienced ARIA-E and ARIA-H, respectively, which required additional magnetic resonance imaging (MRI) surveillance until resolution. Likewise, sensitivity analyses were not performed to evaluate the risk of unblinding among subjects with ARIA.
Do treatment effects outweigh safety concerns?
ARIAs are clinically benign, as they typically resolve spontaneously or after a brief suspension of the treatment. Nevertheless, some individuals develop symptoms, whereas others may experience serious complications, particularly with high therapeutic doses, which could be associated with long-term outcomes. Regulators must review ARIA-associated clinical outcomes and clinical/radiologic severity data for comprehensive safety assessment.
The accelerated loss of brain volume is a less known effect of anti-amyloid drugs that have been reported for aducanumab, donanemab, lecanemab, and other anti-amyloid therapeutics. Beyond trials, accelerated changes in brain volume in Alzheimer’s disease have been attributed to rapid neurodegeneration.
ARIAs appear to be a putative cause of changes in brain volume but have not been thoroughly studied. The protocol for phase III aducanumab trials specified that changes in brain volume would be explored; however, these results have not been fully reported. Despite extensive trial data analyses, public evidence that the volumetric data were reviewed before the approval of aducanumab is lacking.
Brain volume changes due to aducanumab were first reported nine months after approval. Similarly, phase IIb lecanemab trial data reveals accelerated loss of brain volume relative to placebo.
Regulators should consider such volumetric changes related to these drugs when reviewing BLAs. Furthermore, safety data should be regularly updated and made public as a condition for regulatory approval.
Do data indicate disease modification beyond trial duration?
A leading hypothesis in Alzheimer’s disease, specifically regarding disease-modifying therapies, is that continued therapy would result in clinical benefits. However, clinical trials on anti-amyloid immunotherapies are not designed to evaluate disease-modifying versus symptomatic properties. Thus, future trials should address these potential benefits using staggered or delayed-start withdrawal designs.
A reversible symptomatic effect might appear to delay or slow cognitive decline, thereby resulting in a difference in time-to-event. However, assumptions of disease modification should not invalidate the relevance of minimal clinically important differences (MCIDs).
Marginal between-group effects of existing amyloid-lowering drugs are lower than empirical estimates of MCIDs, thus leading to ambiguities over what constitutes clinically meaningful effects.
A new class of drugs to treat Alzheimer’s disease represents a milestone for patients, clinicians, and caregivers. Nevertheless, data from clinical trials have been limited, and questions regarding differential dropout, functional unblinding, clinical efficacy, costs, and safety persist.
Regulatory authorities and payors should address these important considerations before making decisions on approval. To this end, attention to the three questions discussed in this study could guide evidence-based considerations for these drugs.