Rationality of Neurological Clinical Trials: A Challenge that the FDA must Resolve

1. Introduction

Author could summarize the list of failed antibodies along with the phase 3 study details (key I/E criteria), key findings/outcomes and conclusions in a table. In a separate table, author could list the antibodies and present the approval status in US, EMA, UK, Japan, China, Australia and Canada.

On 23rd October 2014, the New York Times front page reported its finding “What Drugmakers Did Not Tell Volunteers in Alzheimer’s Trials,” which disclosed serious concerns about the clinical trials conducted for Alzheimer’s drugs like Eisai's Leqembi (lecanemab-irmb) and Eli Lilly’s Kisunla (donanemab-azbt )[1]. It describes how drug companies failed to disclose critical genetic test results to participants, especially regarding the APOE4 gene variant, which increases the risk of brain injuries in patients receiving these drugs [2]. By 2021, about 2,000 participants enrolled in the Leqembi trials, among them a high-risk subset predisposed to brain bleeding and swelling [3].

Despite informing volunteers about potential genetic testing, companies withheld individual results, a decision now seen as a violation of informed consent principles by bioethicists. Two high-risk volunteers died, and over 100 experienced brain bleeding or swelling, some with severe consequences. Eisai's Leqembi received FDA approval in 2023 under an accelerated approval process despite its safety concerns, which should not have shown only a modest cognitive benefit [4]. In 2024, the FDA had called a meeting to discuss whether it should receive traditional approval. Despite many serious concerns, including the author, the FDA granted traditional approval [3]. However, regulatory agencies in the European Union [5] and Australia [6] rejected it due to its temporary and limited efficacy compared to its risks.

2. Testing Antibodies to Treat Neurological Disorders

The risks associated with amyloid-beta, tau protein, and alpha-synuclein-targeting drugs extend beyond immediate brain injuries, with studies suggesting a higher mortality rate among those treated with antibodies than untreated Alzheimer’s patients. There is also emerging evidence that these drugs may accelerate brain shrinkage, further compounding concerns about their long-term safety. Considering these findings, the clinical trials exemplify the ethical and medical complexities of advancing treatments for neurodegenerative diseases.

The relationship between amyloid-beta, tau protein, and alpha-synuclein-targeting drugs and mortality rates among treated patients is an emerging area of research, particularly in neurodegenerative diseases like Alzheimer's disease and Parkinson's disease.

Many antibodies targeting amyloid-beta (e.g., aducanumab, lecanemab) have been investigated for their ability to clear amyloid plaques in Alzheimer's patients. Some studies have reported an increase in mortality or serious adverse events in patients treated with these antibodies compared to controls, particularly in those with more advanced disease or certain comorbidities [7]. Similar observations were reported with treatments targeting tau protein [8] and the alpha-synuclein-targeting antibodies (e.g., prasinezumab) [9]. Many of the concerns about mortality have emerged from post-hoc analyses of clinical trial data, emphasizing the need for ongoing monitoring and long-term studies. This concern is now further highlighted in the recent NY Times investigation.

While companies continue to pursue the development of these drugs, scientists suggest a broader focus on alternative therapeutic avenues, such as reducing inflammation or enhancing blood flow, as the amyloid theory alone appears insufficient to address neurological disorders multifaceted nature. Table 1 lists the antibodies for which clinical trials were conducted and failed or withdrawn after approval for lack of efficacy or high toxicity. It is important to note that antibodies are effective against neurological disorders like Multiple Sclerosis if the antibody need not cross the blood-brain barrier. However, many of these antibodies have also been withdrawn.

Currently, there are 30 studies listed in clinical trials to treat Alzheimer’s disease (https://clinicaltrials.gov/search?cond=Alzheimer%27s%20Disease&term=Monoclonal%20Antibody), none for Parkinson’s disease, and 72 for Multiple Sclerosis (https://clinicaltrials.gov/search?cond=Multiple%20Sclerosis&term=Monoclonal%20Antibody), using monoclonal antibodies.

3. Future

While the flaws in clinical trials of neurological disorder treatments with antibodies have begun to rise, a more significant issue, whether these drugs even qualify to be tested in the first place based on their lack of bioavailability in the brain, has been widely ignored [21]. The known bioavailability of these antibodies is 0.01-0.1% [22] requiring high dosing to capture the minute effects; it shows that these antibodies are highly effective, but their inevitable high dosing makes them a significant clinical concern. There is also a possibility that these effects are placebo driven, as commonly found in such treatments [23,24,25,26].

The issue of ensuring bioavailability becomes more relevant since there are known mechanisms to improve the entry of antibodies into the brain. Antibodies can be readily made effective and safe by engineering them to go through transcytosis [27] such as by binding the antibodies with transferrin protein or its N-methyl lobe using a cleavable linker to avoid exocytosis, increasing the half-life of these antibodies in the brain [28].

Binding antibodies with transferrin protein or N-methyl lobe with a cleavable linker is the most optimal engineering. To demonstrate that such binding does not adversely impact the efficacy of these antibodies, we employed bioinformatic modeling to investigate the impact of non-cleavable linkers (G4S)n) on antibody binding. Our examination, centered on steric hindrance, sought to understand the effects of the length of the linker and its influence on the binding avidity with the Aβ protein.

In addition to the clearance pathways, the behavior of antibodies in the brain can vary depending on their affinity for specific target antigens. For instance, antibodies designed to bind high-affinity receptors, such as those on transport mechanisms like the transferrin receptor or therapeutic targets such as amyloid-beta in Alzheimer’s disease, may experience enhanced retention due to these interactions [29].

The complex formed between Aducanumab and Aβ has already been documented in the Protein Data Bank (PDB ID: 6c03)[30,31]. Leveraging this information, we conducted modeling experiments focusing on Aducanumab, an Alzheimer's disease treatment recently withdrawn from the market. Our methodology involved pre-processing and standardizing the antibody structure using UCSF Chimera, protein structure prediction using the AlphaFold2, followed by docking analysis using HADDOCK, both before and after linkage with transferrin via the (G4S)n linker. Subsequently, we evaluated binding affinity and interaction patterns through the PRODIGY server, paying particular attention to the linker's length. The binding parameters calculated included (Figure 1):

Retention of antibodies within the brain can be further influenced by receptor recycling pathways, such as those involving the neonatal Fc receptor (FcRn). Such mechanisms are more prominent in peripheral circulation, where FcRn is critical to antibody longevity. Still, their presence in the CNS is limited, and their effect on half-life is less pronounced [32].

The exocytosis of antibodies can be reduced or prevented by linking the transcytosis agent with a cleavable linker that would break when the conjugate enters the brain [33].

Antibody engineering may also involve removing the Fc region; however, this approach has several limitations. When the Fc (Fragment crystallizable) region of an antibody is removed, such as in antibody fragments like Fab (antigen-binding fragment) or scFv (single-chain variable fragment), the half-life within the brain can be significantly shortened [34].

5. Conclusion

Current clinical trials of antibodies intended to treat neurological disorders are unethical, unless there is sufficient proof of their reaching into the brain. While the FDA must answer the New York Times report with evidence that it has followed the GAO recommendations in qualifying the IRBs, the FDA must revise clinical trial guidelines requiring proof of sufficient bioavailability at the target site, where known, to qualify exposing humans to such trials. The developers must adopt antibody engineering to enhance the transcytosis and preventable exocytosis of antibodies targeting the brain tissue. Additionally, such trials should take into consideration the possibility of significant placebo effects, common for these drugs [28]. It is worth suggesting that the clinical effects reported with a minute of drug reaching out in the brain may be due to placebo effects since, they all end up with high toxic effects.

The need for effective treatment of neurological disorders is critical; developers chasing multibillion dollar markets are eager to test antibodies and finding they are failing and harming patients. The FDA must take actions to reduce human abuse in these trials and encourage development of scientifically rational products, which none of the current antibodies are.