The approach in this paper is, of course, relevant particularly to Parkinsons, but also possible Alzheimers. In essence it argues for a thesis that general cellular failure is the cause and too much concentration on the symptoms won’t find a cure.
As people know I think the key mechanism is through lower nuclear acetylation, but as this arises from mitochondrial failure and insufficient autophagy I don’t disagree with this paper in a big way though I don’t think getting stressed about NAD is much use.
chatGPT:
Summary
This Cureus paper is a narrative hypothesis framework for ALS, not an experimental paper or clinical guideline. Its central argument is that ALS symptoms should be understood at the symptom/circuit level, rather than only as motor-neuron death. The author proposes that different symptom domains—limb weakness, bulbar decline, respiratory failure, fine-motor impairment, fatigue, mood symptoms, and cognitive change—reflect different neural circuits crossing thresholds of biological failure.
The model links six main processes:
1. Microglial pruning pressure
The paper proposes that excessive or mistimed complement/activity-dependent microglial pruning may reduce synaptic reserve. This is extended to explain possible overlap between ALS and depression, with mood symptoms in some ALS patients hypothesised to reflect limbic/prefrontal circuit vulnerability rather than only psychological reaction to diagnosis.
2. Mitochondrial/OXPHOS reserve
The author argues that circuits with insufficient oxidative phosphorylation capacity may be less able to maintain synaptic plasticity, compensate for damage, and resist progression.
3. NAD+ as a compensatory buffer
NAD+ is treated as a central reserve molecule supporting mitochondrial function, redox balance, and repair. The suggested transition is: adequate NAD+ allows compensated plasticity; strained NAD+ marks fragile plasticity; NAD+ exhaustion contributes to collapse.
4. Autophagy/proteostasis failure
ALS-linked genes such as TBK1, OPTN, and C9orf72 are used to support the idea that impaired autophagic clearance can convert synaptic/bioenergetic stress into progressive network failure.
5. Aging as loss of adaptive bandwidth
Aging is framed not as a separate cause, but as a narrowing of the reserve margin: lower OXPHOS/NAD+ reserve, more senescence/inflammation, and weaker autophagy.
6. ALS–FTD pathway divergence
The paper proposes that ALS and FTD may differ in pathway architecture: ALS showing higher PI3K-AKT-mTOR and lower OXPHOS signals, while FTD is framed more around PGC-1α/mitochondrial biogenesis failure and SIRT1-linked metabolic changes.
The paper condenses this into the inequality:
P + G + Q > E + N + A
where P = pruning pressure, G = glutamatergic/excitability burden, Q = aging/inflammatory stress, E = OXPHOS capacity, N = NAD+ reserve, and A = autophagy/proteostasis capacity. Symptoms emerge when stress exceeds reserve.
Novelty
The main novelty is not that the individual components are new. Microglia, mitochondria, NAD+, autophagy, mTOR, aging, ALS-FTD overlap, and NfL are all already part of ALS biology. The novelty is the integration.
The paper’s distinctive contributions are:
First, it tries to shift ALS stratification from “ALS as one disease label” toward symptom-domain biology: limb, bulbar, respiratory, fine-motor, cognitive, mood, and fatigue domains might each have different biological state profiles.
Second, it proposes a three-state model: compensated plasticity → fragile plasticity → network collapse. This is useful conceptually because it separates early compensatory disease from later irreversible failure.
Third, it makes NAD+ a phase-dependent compensatory buffer, rather than simply another metabolic marker.
Fourth, it links microglial pruning to mitochondrial and autophagy failure: pruning alone may be tolerable, but pruning plus weak energy reserve plus poor clearance becomes destructive.
Fifth, it attempts to generate testable biomarker predictions, including NfL trajectories, NAD+ metabolites, OXPHOS signatures, autophagy markers, complement markers, and ALSFRS-R subdomain slopes. Figure 2 explicitly presents the author’s four interlinked studies as a hypothesis-generating chain and calls for iPSC motor-neuron systems, animal models, longitudinal biomarkers, and replicated TWAS analyses before clinical interpretation.
Critique
The paper is intellectually ambitious and well aligned with the idea that ALS is a systems-level disorder rather than a simple motor-neuron cell-loss disease. Its strongest feature is that it tries to make ALS heterogeneity biologically interpretable. It also repeatedly warns that the framework is not clinically validated, which is important.
However, the main weakness is that the framework is too internally dependent. The paper synthesizes four studies by the same author, three of which are described as preprints. The paper itself acknowledges that these works are structurally interdependent and should not be treated as independent evidence. That creates a real risk of circular reinforcement: one hypothesis paper, one TWAS, one simulation, and one ALS-FTD comparison may fit together because they were designed around the same conceptual architecture.
A second problem is that much of the model depends on associative inference. TWAS can prioritize pathways, but it cannot prove that predicted gene-expression differences cause ALS progression. The paper notes that the TWAS results were not independently reprocessed, gene-set definitions were not audited, and colocalization was not tested. That is a major limitation if the model is to support pathway-specific intervention.
Third, the microglial pruning argument is plausible but under-proven in ALS. Complement-mediated pruning is well established in development and implicated in other neurological contexts, but showing that it drives ALS symptom progression would require direct ALS-specific evidence: longitudinal complement markers, synaptic-density imaging, microglial activation measures, and correlation with circuit-specific decline.
Fourth, the NAD+ framing may be too central. NAD+ biology is relevant to aging and mitochondrial stress, but ALS-specific clinical evidence remains limited. The paper is careful on this point, but the overall model may still give NAD+ more causal weight than the current evidence can support.
Fifth, the computational model is useful for generating predictions, but it risks creating a false sense of mechanistic precision. Variables such as pruning pressure, NAD+ reserve, autophagy capacity, and circuit collapse are not yet measured in a standardized way in ALS patients. Until they are operationalized, the equation is more of a conceptual diagram than a quantitative model.
Sixth, the proposed multi-target therapeutic logic is premature. A regimen targeting pruning, glutamatergic burden, NAD+, and autophagy is biologically attractive, but ALS has a long history of plausible mechanisms failing in human trials. The paper correctly states that such approaches would need pharmacological definition, dose justification, interaction review, safety monitoring, and target-engagement endpoints before human testing.
Overall assessment
This is a hypothesis-generating framework paper with a useful systems-level structure. Its value is in proposing a way to connect ALS symptoms, circuit failure, mitochondrial reserve, NAD+ dynamics, autophagy, aging, and ALS-FTD divergence. Its weakness is that it is not yet independently validated and leans heavily on author-derived, mutually reinforcing work.
The best next test would be a longitudinal ALS cohort measuring ALSFRS-R subdomains, FVC, NfL, NAD+/NADH metabolites, OXPHOS signatures, complement/microglial markers, and autophagy markers, ideally stratified by SOD1, C9orf72, bulbar-onset, respiratory-onset, and lower-motor-neuron-predominant phenotypes. Until then, the paper is best read as a research scaffold, not as evidence for treatment selection.