As a layperson I have been studying ALS by reading scientific publications for several years, it started after the second time someone went ill in my family. This blog is the result of that effort. Scientists claim to find cures for most chronic diseases every week by targeting one or the other obscure molecule. We know that most of these articles are of very poor quality. What feels important to me are articles that show a good knowledge of human anatomy and physiology. I don't care about molecules in mice.

Nearly one century ago, medical doctors made a parallel between spinal cord injury and spinal-onset ALS. In both cases, there is important muscle wasting. And this is what kills people, so it's important to understand how this develops and what are the consequences. Obviously, a spinal cord injury will sever the link between the brain motor area, some upper motor neurons, and corresponding lower motor neurons and muscles. Yet it does not stop there, which is particularly interesting when we have ALS in mind. In spinal-onset ALS the disease starts in a very localized muscular for example a muscle in the thumb, and it spreads, often until respiratory muscles fail.

Similarly, a spinal cord injury has consequences that are far from being limited to some motor neurons and diseases. For example, pulmonary infection is a leading cause of morbidity after spinal cord injury. Bizarrely this happens because spinal cord injury causes atrophy and dysfunction in the lymphatic system, an organ system in vertebrates that is part of the immune system.

Now scientists from Ohio, describe systemic wasting that affects innervated non-paralyzed skeletal muscles. The muscles that are affected in ALS are also skeletal muscles. This wasting appeared within 1 week after experimental spinal cord injury in mice and in three weeks half of the muscle mass disappeared. This waste affects the whole body. Skeletal muscle fibers are broadly classified as "slow-twitch" (type 1) and "fast-twitch" (type 2). This muscle wasting affects fast type 2 muscles preferentially, and became exacerbated after a third thoracic vertebra (T3) paraplegia compared with low (T9) thoracic paraplegia. The T3 vertebra is situated at the level of the shoulder, while the T9 is at the height of the sternum.

Markus E. Harrigan, Jan M. Schwab et al. remark that the wasting of nonparalyzed muscle and its rapid onset and severity cannot be explained by disuse, so it implies unknown systemic drivers. Mechanisms underlying systemic muscle wasting (including fully innervated non-paralyzed muscles) early after paraplegic SCI would imply the presence of biological signaling which can quickly reach muscles of the entire body. Knowing the cause of this widespread muscle wasting after a T3 paraplegia in spinal cord injury might illuminate the similar phenomena in ALS.

The authors found that muscle transcriptome and biochemical analysis revealed a glucocorticoid-mediated catabolic signature early after T3 spinal cord injury. They generated an inducible skeletal muscle-specific glucocorticoid receptor (GR) knockout mouse model in order to test that hypothesis.

Spinal cord injury-induced systemic muscle wasting was mitigated by (i) endogenous glucocorticoid ablation (adrenalectomy) and (ii) pharmacological glucocorticoid receptor (GR) blockade and was (iii) completely prevented after T3 relative to T9 spinal cord injury by genetic muscle-specific GR deletion.

These results suggest that hypercortisolism contributes to a rapid systemic and functionally relevant muscle wasting syndrome early after paraplegic spinal cord injury in mice. Indeed in humans, hypercortisolism induces central muscle weakness, adipose tissue redistribution, skin fragility and unusual infections. Hypercortisolism has also been implicated in ALS, maybe it would be interesting to test glucocorticoid inhibitors in a clinical trial. There are studies (here or here) that show that glucocorticoid inhibitors ameliorated the health in an ALS mice model.

It's been known for more than 20 years that neprilysin inhibitors such as sacubitril could foster Alzheimer's disease. Yet in 2015, Novartis decided to ask the FDA to authorize Entresto (sacubitril/valsartan) for heart failure. A clinical trial was set up and failed to meet the primary objective (heart failure), nevertheless (usual story) the drug was authorized by FDA.

At that time it was well known that this drug could foster Alzheimer's disease, so the FDA mandated that another clinical trial was also designed (PERSPECTIVE; NCT02884206) to test if there was a risk to cognitive functions. There were also other clinical trials with the same goal, for example in Korea. These clinical trials didn't show any aggravation of cognitive functions.

More recent studies have suggested that this class of drugs (neprilysin inhibitors) could have both positive and negative effects on the development of Alzheimer's disease. . (source: Naif H. Ali and Hayder M. Al-Kuraishy)

A re-analysis of the results of another clinical trial by scientists from the University of Gothenburg, in collaboration with colleagues from the University of Glasgow, is less optimistic about the effects of sacubitril on cognitive functions.

There are many other drugs to manage heart failure and it is not even clear if Entresto brings substantial benefits to the patients. Why was Entresto authorized in the first place in 2015, and why is it still prescribed given the abundant literature about the effects of sacubitril on Alzheimer's disease?

Un article important est publié par Poul F Høilund-Carlsen et des collègues à travers le monde, sur la révision annoncée des critères de diagnostic de la maladie d’Alzheimer. Nos précédentes publications sur ce site, avertissaient déjà que cette révision aurait pour conséquence principale que les essais cliniques de médicaments seraient majoritairement approuvés, alors que la totalité des essais cliniques (324 de phase III) sur la maladie d’Alzheimer (y compris les médicaments récemment autorisés) se sont soldés par des échecs, et parfois par des effets secondaires dramatiques (ARIA).

Aussi bizarre que cela puisse paraître, les critères diagnostiques de la maladie d’Alzheimer ont subi de nombreux changements depuis 40 ans.

Initialement, elles reposaient principalement sur une évaluation clinique (l’état du malade). En 2011, l'Institut national américain sur le vieillissement et l'Association Alzheimer (NIA-AA) a approuvé, à des fins de recherche, un diagnostic de maladie d’Alzheimer préclinique basé sur l’imagerie médicale quand elle montre la présence de plaques d’amyloïde dans le liquide céphalo-rachidien. Pourtant on a démontré via des cohortes qu’un tiers des personnes âgées de plus de 75 ans peuvent avoir des plaques (et autres assemblages moléculaires) d’amyloïdes sans avoir de perte cognitive.

Cette proposition de diagnostic de 2011 ne concerne pas les médecins, elle est uniquement destinée aux scientifiques. On peut y voir une nouvelle illustration du principe que les outils distordent la perception du monde. En l’occurrence tester un malade avec un PET-amyloïde nécessite un quart d’heure et la présence d’une zone réagissant à un biomarqueur ainsi que d’une atrophie cervicale à l’imagerie, alors que tester la cognition est compliqué (y compris quand le patient ne veut pas coopérer) et toujours discutable par les proches.

En 2018, une série d'auteurs ont créé un nouveau cadre de recherche NIA-AA mettait un accent crucial sur « A », les biomarqueurs des plaques amyloïde-bêta (Aβ), et « T », les biomarqueurs de la protéine tau. En revanche, les biomarqueurs de la neurodégénérescence « (N) », notamment l’hypométabolisme et l’atrophie étaient indiqués entre parenthèses, indiquant un rôle diagnostique moindre.

La dernière révision (2023) proposée est non seulement dépourvue d’évaluation clinique, mais elle repose uniquement sur des molécules biomarqueurs, dont les rôles pathogènes n’ont jamais été prouvés. Ce qui est très grave c’est que les proposants souhaitent l'approuver non seulement pour la recherche mais également pour la pratique clinique et surtout pour les essais cliniques.

Les auteurs de cet article illustrent un résultat hypothétique de ce scénario: Celui ou un nouveau médicament serait efficace sans pour autant affecter la présence d’amyloïde et de tau dans le cerveau.

Ce médicament hypothétique, bien qu’améliorant l’état des malades, serait considéré comme un « échec » dans les essais cliniques. Pire les personnes qui présentent une amylose cérébrale et aucune démence seraient également identifiées comme des patients atteints de maladie d’Alzheimer.

Ce que les auteurs ne disent pas, ce qui est indicible dans notre société, c’est que les sociétés pharmaceutiques seraient les grandes gagnantes de ce changement. En effet la plupart des médicaments proposés depuis 5 ans sont efficaces contre les plaques amyloïdes, pour autant aucun n’est capable de montrer une amélioration de l’état de santé des malades.

Poul F Høilund-Carlsen et ses collègues proposent une procédure de diagnostic basée sur une évaluation clinique et des preuves in vivo d'une neurodégénérescence accrue qui est essentiellement la façon actuelle de diagnostiquer la maladie d’Alzheimer par les médecins.

Peut-être que les scientifiques devraient davantage fréquenter les Epahd, la maladie d’Alzheimer est quelque chose de beaucoup plus complexe que ce qu’ils pensent et enseignent.

On pourrait penser que le vieillissement affecte surtout les souvenirs anciens, c’est le contraire qui semble se passer. Pourquoi l’existence d’un conjoint ou d’un enfant est-elle occultée ?

On pourrait aussi penser que le trouble, une fois installé serait persistant. Pourquoi un malade se pense-t-il, se vit-il à une période de son enfance pendant quelques minutes puis sans trouble apparent est capable de se situer dans le temps présent?

On sait bien aussi que beaucoup de malades d’Alzheimer ont des troubles moteurs de type Parkinsonien (freeze). Il ne s’agit sûrement pas seulement d’un problème de mémoire ou de perte de cognition.

Mais le monde de la recherche médicale est, depuis la financiarisation de cette activité dans les années 80, gangrené par l’avidité mercantile. Il lui faut de l’argent rapidement et facilement, on est loin d'un idéal académique.

The article discussed today is a more philosophic one, as there are currently few research news in our field.

Frustratingly it starts great but quickly loses its way. This article is quite representative of a current wave of attacks on the pragmatic aspect of medicine by scientists frustrated by their lack of progress. As they say in Silicon valley "Fake it till you make it".

The authors are indeed right that drug discovery in neurodegenerative diseases is currently in a conceptual crisis. A clear illustration of this can be seen in hundreds of failed clinical trials of Alzheimer’s disease, ALS, vascular dementia, dementia with Lewy bodies, and Parkinson’s disease dementia, despite countless hypotheses on possible mechanisms.

To date, we have accumulated many negative randomized clinicals in these diseases. These results have not prompted a major reconsideration of the toxic proteinopathy hypothesis of causality. Imperfections in trial design and execution (incorrect dosage, insensitive endpoints, too-advanced population) but not in the underlying hypotheses have prevailed as explaining the failures. Surprisingly, the huge cost of these trials has not interrogated the managers of the pharmacological industry.

Many existing drug development programs for those diseases have been based on the “proteinopathy” concept, whereby proteins in a misfolded aggregated state cause the disease. As a consequence, it is understood that removing them from the brain should cure these diseases.

The cumulative lessons of more than 40 anti-Aβ Alzheimer’s trials should have taught us otherwise: 15 monoclonal anti-amyloid antibodies significantly reduced amyloid and, surprisingly, significantly worsened patients’ outcomes compared to placebo. Yet none of these trials was interpreted as a rejection of the hypothesis. Instead, the anti-amyloid antibodies, lecanemab and donanemab , which met the statistical threshold in the opposite direction, have been taken as a confirmation of the hypothesis, although they never translated into clinically relevant benefits. Instead, reduction in soluble Aβ levels, as measured in cerebrospinal fluid, is harmful to humans, and removal of insoluble Aβ may lead to microhemorrhages, brain atrophy, and death. The toxic Aβ hypothesis has become virtually unfalsifiable. Moreso, scientists now propose to define and diagnose these diseases by molecular biomarkers. This means that any drug that can reduce those diagnostic biomarkers (and many can) must be authorized by the FDA!

Collective evidence should have sufficed to consider amyloid rather as a downstream consequence in cellular pathophysiology, a sign of a range of biological stressors, not their cause.

The scientists call for a move from the current medical methods towards a divergent, organ-agnostic, and mechanism-based disease nosology (the branch of medical science that deals with the classification of diseases).

They complain that the focus has always been on the brain. This brain-centric approach has been maintained by the majority of clinicians and researchers to date. Yet a single pathology in the brain is the exception rather than the rule. In clinical practice very rarely do doctors find patients with only dementia or only metabolic, sensorial, or motor symptoms.

Real-world aging individuals, with or without neurodegenerative disease, have mixed manifestations of biomarkers in their brains, with almost 80% of these individuals presenting with at least two of such neuropathologies.

In contrast with the model of brain-centricity, these “dementia specific” markers can be often found in organs outside the brain, for example, Aβ aggregates documented in the skin, intestines heart, and pancreas. This should not surprise us, all cells in our body share the same DNA, and most cellular receptors are not specific to a single organ.

Semantic wandering

The authors are in favor of "network medicine". Network medicine is the application of bioinformatics concepts towards identifying, preventing, and treating diseases. It focuses on using network topology and network dynamics to identify diseases and develop medical drugs. Disease networks, which map relationships between diseases and biological factors, also play an important role in the field.

Sadly but in line with this network medicine approach the authors propose a new way to characterize diseases that is the current proposed way with just another name. The authors propose to use endotypes, but endotypes are synonymous with biomarkers! They curiously use an example of the failure of their own "modern" approach to promote it: They recall the recent accelerated approval of the anti-amyloid monoclonal antibodies aducanumab and lecanemab by the US Food and Drug Administration became the culmination of this paradigm, despite the failure of their clinical trials.

Revising disease and drug concepts

They also propose to use the concept of disease module instead of phenotype (observable characteristic or trait of a disease) to classify and name diseases. For the authors, disease modules are rather small localized multi-protein signaling networks, typically distinct from curated canonical signaling pathways. As not two scientists can agree on subtle pathways or bioinformatics networks, good luck to those tasked to define disease modules.

The authors do not stop there, they also want to change the meaning of the word "drug". For them, as the current drugs are inefficacious, we must accept no drugs approach such as lifestyle interventions. In another publication, the authors attack the clinical trial's current way of interpreting results and propose a minimal set of rules that facilitate the interpretation of negative clinical trials as falsifying the driving hypotheses, in particular, if the desirable change in surrogate endpoints has been achieved.

Conclusion

It's not clear why the authors wrote this article, it does not create new knowledge, instead, it's a semantic play on existing concepts. As for “network medicine”, it remains to be seen whether it will provide a solid framework and pragmatic results in the future, or whether it only constitutes a high-tech version of another alternative medicine.

There are many hypotheses about the etiology of ALS, and most of them are probably correct. They can be classified in several ways, but one of them is usually to separate a genetic origin from a "sporadic" origin. But even for a genetic origin, it is unlikely that the disease will wait 50 years before striking. The etiology certainly proceeds from multiple causes and stages, and the genetic aspect concerns only one of these stages towards the disease.

Spinal muscular atrophy (SMA) is a genetic disease that closely resembles ALS, but like most diseases resulting from a deleterious variant of the genetic heritage, it strikes toddlers. Infants with this disease have a defective variant of their SMN1 gene and have a very short life expectancy. Children born with type 1 SMA, until recently, died before their second birthday. Although SMA is much more common than ALS, we don't hear much about this disease because most patients die before the age of 3 and there is no significant effort to raise awareness. A few years ago, some therapies appeared, Spinraza, Zolgensma and others. Although hailed as life-saving drugs when they first appeared, they have serious side effects. Drug development continues, often aiming to increase SMN2 production to compensate for the lack of healthy SMN1.

Among the many hypotheses concerning ALS, there is a very minor one: In aging people, senescent motor neurons, exhausted by numerous stresses, reenter the cellular life cycle, perhaps as the result of a mechanism of adaptation to stress. Since it is difficult to imagine how a very elongated motor neuron would divide, they quickly die in this attempt. Usually, cells resume their life cycle after signaling by CDK proteins. CDK proteins constitute a family of proteins involved in the regulation of the cell cycle. Dysregulation of CDKs, particularly cyclin-dependent kinase 5 (Cdk5), is seen in many neurological disorders, including Alzheimer's disease (AD) and Parkinson's disease (PD). Cdk5 is a unique member of the CDK family because it does not play a critical role in cell cycle progression and is not activated by a cyclin. Instead, Cdk5 is normally activated by the regulatory protein p25. Cdk5/p35/p25 activity is normally an important regulator of the proper development of the mammalian central nervous system.

Scientists from Northwestern University have identified, in a mouse model of SMA, an unexpected role of Cdk5 signaling in the appearance of mitochondrial defects and selective degeneration of motor neurons.

The scientists report that Cdk5 activity is significantly increased in their mouse and pluripotent stem cell (iPSC) models of SMA. The increase in Cdk5 activity occurs before the appearance of SMA phenotypes, suggesting that it may be an initiator of the disease.

The article does not clearly show what causes the transition from the p35 subunit to the p25 activator. In vitro studies have already suggested that aberrant activation of Cdk5 by an endogenous truncated version (p25) of p35 could be a key event in the process of neurodegeneration.

An enzyme responsible for cleaving p35 to form p25 is calpain, a calcium-activated protease implicated in neuronal cell death and notably ALS in the past. There is also evidence that hyperactivation and redistribution of Cdk5 by p25 plays a critical role in the phosphorylation of “pathological” substrates (such as tau which is implicated in Alzheimer's disease).

As inhibition of Cdk5 signaling inhibits the degeneration of motor neurons derived from SMA mice and human iPSC models of SMA disease, this suggests that reducing aberrant Cdk5 activation could potentially improve SMA disease symptoms and benefit patients. patients. This could also have implications for other motor neuron diseases, such as ALS.

From there, we can adopt two attitudes on how to use this new knowledge. - The first is for scientists to learn more about the long chain of molecular events that lead to disease, in the hope that it is not too complex for our limited human cognitive abilities. - The other is that of doctors, it is more pragmatic, for example, it could lead to efforts to develop a Cdk5 inhibitor capable of targeting the pathway to slow down the degeneration of motor neurons. However, thousands of clinical trials on Alzheimer's disease demonstrate every day that a pragmatic approach is rigorous but very ineffective.