Ozempic Doesn't Seem All That Mysterious To Me

GLP-1 is a metabolic hormone, a famously high-impact molecule type

Post inspired by the ever-careful and well-researched Scott Alexander of Astral Codex Ten: in this piece published this week, he mused about the efficacy of Ozempic for conditions that seemingly don’t have much to do with weight or diabetes, the two primary conditions GLP-1 agonists were developed to treat.

It’s not surprising that an intestinal hormone could treat both diabetes and obesity. When you eat a big meal, your body needs to deal with the sudden blood sugar spike. And it also needs to signal to the brain to be full. At least in retrospect, all of this makes sense.

But Ozempic and other GLP-1 drugs appear to be a promising treatment for alcoholism, smoking, stimulant addition, opioid addiction, and maybe even behavioral addictions like shopping. Why?

Scott goes through the better-understood functionality of Ozempic: a hormone that works primarily in the brain to regulate signaling for both fullness and hunger, providing the following possible pathway:

He muses on the possible reasons that Ozempic could work on things like addiction, specifying some narrow receptor pathways in the brain, citing existing literature on GLP-1 specifically tamping down an interest in high-reward food over standard food.

I confess that I started writing this post when he got to the Alzheimer’s and Parkinson’s section, in particular due to his reaction to this diagram (source: this paper).

He notes that dementia is improved with Ozempic use in patients with and without diabetes; spends some time on the ambiguous boogeyman that is “inflammation” in modern nutrition, health, and wellness parlance; again throwing in some narrow, pathway-specific hypotheses for Ozempic as an anti-inflammatory.

He goes on: Ozempic is great for obesity-related cancers, why? What about obesity is connected with all of these disorders? How is Ozempic assisting with diseases that are not universally caused by overweight or obesity, like heart disease? How is it improving outcomes for addicts, given the lack of connectivity between eating food and taking heroin?

Are all of these outcomes just standard promiscuous pharma marketing [Try the new wonder drug! For everything from the common cold to improved libido]? Evidence would suggest otherwise, at least for now.

I think there are some obvious explanations for these questions, and I’m partially publishing this from a place of confusion for myself: why are the explanations that seem pretty parsimonious and obvious to me not the first hypotheses I’m seeing from Scott or the people in my circle?

I acknowledge that I don’t have evidence for every assumption included below, to say nothing of the fact that rigorously proving these intuitions is considerably more difficult than speculating on them.

But to walk through my assumptions and model with some quick and dirty sourcing:

• Your metabolism is controlled at a cellular level by mitochondria and at an organ and organism level by your hormones

• GLP-1 is a hormone. Hormones affect both your metabolism and your reproductive system and a number of cross-over features of the same: they manage blood sugar, fat distribution, muscle development, reward pathways (including motivation and obviously satiety)

• The metabolism is implicated in the function of several organs directly: pancreas, GI tract, liver, kidney, this is not an exhaustive list. The function of the metabolism depends on their macro-level function, and their macro-level function will be impaired in the presence of improper metabolism

  • Why is it less surprising that GLP-1 agonists affect pancreatic function - an input-management, metabolically-relevant organ - than that they affect kidney function - an output-management, metabolically-relevant organ?

• The metabolism includes your reward system; a healthy metabolism cannot exist in the presence of a dysfunctional food reward system

  • Your body actually is designed with some safeguards to prevent both overweight and obesity. Your reward system is supposed to reward you both for seeking food you need *and for stopping when you’re full*. It’s supposed to reward you for consuming *and expending* energy. You can’t get overweight or obese without something going *wrong* with that system. [You also can’t be underweight without the same issue going the other way, and I suspect that being underweight may disrupt reward and satiety in a way that increases risk of being overweight if certain conditions are met]. In other words, even if you gained a lot of weight by solely by eating too much because it was fun, or ate too little for too long, a side effect is that your metabolism now does not work quite the way it’s supposed to, and your reward system - at minimum the portion that controls satiety - is not maximally functional either

    • This implies that if, for example, something weird happened, like a bunch of people gaining a lot of weight all at the same time, there’s a real possibility that some feature of their environment was fucking with their food reward system in a way that fucked with their metabolism by extension.

    • Some people are true believers that fat only happens to lazy bad people, leaving little mental space for more complicated models than “fat people eat too much.” This limits even basic explanatory thinking like “you don’t get fat unless something went wrong with your body’s standard functioning,” because their model is rooted in whatever maximally implicates fat people in their own fate.¹

  • Your reward system is obviously implicated in addiction - your body is also supposed to tell you to avoid substances that impair function/cause damage (alcohol, drugs), and to cease repetitive/compulsive behaviors where they inhibit survival and reproductive functions (porn, gambling, shopping, etc). It seems reasonable to suppose that addictions and compulsions in some way subvert the system that is working in healthy people who safely engage in drinking, shopping, and so on. You’d further expect that one form of reward dysfunction (underweight/overweight/obesity) would be comorbid with other forms of reward dysfunction (alcohol abuse, compulsive porn usage, compulsive gambling, etc).

    • I would expect this loop is tighter for anything that directly harms metabolically relevant organs - alcohol abuse, for example, damages the GI tract and liver over time, and it would be surprising if this didn’t exacerbate the breakdown of proper metabolic feedback in people who drink to excess

• Cells with outsize number of mitochondria - neurons, muscle (these two are themselves a pretty tightly connected network), etc - should show outsize response to anything affecting metabolism, which at a cellular level is largely completed by mitochondria

• Dysfunctional metabolism of what you ingest results in more harmful waste products (both quantity and quality) than a functional metabolism

  • Inflammation: yes, inflammation sounds like a bogus catch-all term, because way too many people use it that way. Yes, what you eat has some direct effect on baseline inflammation. [What, you thought you were eating anti-oxidants for fun? Of course you did, you were eating dark chocolate and drinking red wine.] But just as importantly, we’d expect *how well you metabolize food* to affect how much inflammation your diet causes.

    • Every time you eat something, your body breaks it down via metabolism into what scientists have very cleverly called metabolites: byproducts of what you eat that have a useful function (e.g. amino acids), neutral waste products, and in some cases, negative byproducts (part of why drinking alcohol is bad for you is that it produces destructive metabolites, including carcinogenic acetylaldehyde, when processed by your body). One would expect that the more fucked-up metabolites your body produces - the bastard molecular children of something you either shouldn’t have ingested or that your body did not competently process - the louder and more dramatically your body will call down the cavalry of your immune system to deal with them. One would expect more fucked-up metabolites in anyone with any kind of metabolic disorder, including but not limited to, obesity. From there, one would expect more unproductive inflammation in anyone with any kind of metabolic disorder.

      • Messed up metabolites almost definitely also aggravate the helpful bacterial colony in your gut which plays a physiologically significant role in digesting your food. Data on the microbiome is still in its infancy - any given paper is not guaranteed to be high quality bc it’s such a hot topic - but the existence and relevance of a functional microbiome for overall metabolism appears to be well-established, and I therefore feel reasonably confident speculating that more weird byproducts of metabolism will increasingly disrupt the proper function of your microbiome, potentially with a negative feedback-loop mechanism.

  • Heart disease: heart disease is honestly a bunch of diseases wearing a trench coat which is not as clear as the medical establishment could ideally be. Heart disease is not always associated with obesity but it is consistently associated with metabolic dysfunction,

    • Heart disease is sexually dimorphic, specifically responding to hormonal profile: Cis men are at greatest risk for contracting and dying from heart disease [What’s wild is that I can’t find evidence that giving people testosterone will definitely increase your risk profile]. I mention this mostly to indicate that it’s not exactly novel to find that hormonal profile affects risk profile for metabolic disorders.

    • This is the part where I call to mind your hazy memory that cholesterol is something your body produces and something the medical establishment spent years telling you not to ingest nor to produce too much of. A subset of circulatory disorders - including stroke - appear to be occurring due to a build-up of fatty deposits left behind in your blood vessels like silt on the bank of a meandering stream. These plaques are made up of "cholesterol, fat and blood cells", ergo doctors for decades felt safe advising you to both not ingest the cholesterol of other animals (yes you’ve heard about eggs) and telling you to take action to reduce the cholesterol produced by your own metabolism [cholesterol is basically a way for your body to package up excess lipids - fats - for blood stream transit]. The story is weirder than that but I thought I’d make a note of a particular metabolite - cholesterol - and how it may interact with certain forms of circulatory dysfunction.

  • Diseases of the brain such as Parkinson’s and Alzheimers:

    • let’s come back to the Parkinson’s figure, the one that Scott referred to as being trolled by god:

• - We’ve covered the fact that inflammation is reasonably likely to increase in the presence of dysfunctional metabolism due to increased presence of damaging metabolites.

• - We’ve discussed the fact that mitochondria is the central player in the cell-level metabolism of nutrients. Let’s non-snarkily return to anti-oxidants: those are supposed to reduce oxidative stress, just like the GLP-1 enhanced mitochondria are reported to do in the above figure, which also means lower inflammation.

• - Apoptosis and autophagy: these are both basically waste disposal systems. They take malfunctioning cells and/or cells with screwy components and package them into your body’s version of safe biohazard bins, ones that a standard, low-intensity immune response can easily manage. Apoptosis is where the cell kills itself, sometimes using the mitochondria directly, splitting off into self-contained bubbles that can be safely cleaned up by the immune system (macrophage cells, specifically). This is less messy and causes less immunological havoc than your cells just bursting and dumping their innards everywhere, but it still means that cells are needing to off themselves in some way (bad if it’s happening like a lot).
  

In autophagy, the cell puts any malfunctioning molecules in their own bubble and then merges that bubble with existing waste bubbles in the cell - behold:

• -Apoptosis and autophagy have some related functions and pathways, and themselves interact with mitochondrial function, according to both the Parkinson’s figure and from within the literature. I’ll speculate that this is another case of mess with the metabolism, mess with stuff that uses mitochondria (apoptosis) and stuff that uses its byproducts (autophagy).

• -Protein folding is the least convenient of these to explain because of all the elements on Scott’s god-trolling figure, medicine knows the least about the messed up proteins and plaques implicated in Alzheimers and Parkinsons. My best assumptions are that given that both Parkinsons and Alzheimers are known to be dependent on the presence of certain lipids (fats, basically, that can only be broken down or built up from the fats you eat), one would expect that dysfunctional metabolism would lead to dysfunctional lipid metabolism which would lead to a higher likelihood of developing a disorder reliant on having weird lipids in your brain cell membranes.

• Putting it altogether: every biological pathway on this Parkinson’s paper figure has likely or proven connections to metabolism at a cellular and systemic level

And for the grander summary: GLP-1 agonists appear to improve both metabolic dysfunction directly (obesity and diabetes), conditions which are probably caused by downstream effects of metabolic dysfunction (heart disease, stroke, Parkinson’s), and conditions that you would expect to be comorbid with certain kinds of metabolic dysfunction (addiction).

Again, the actual in vitro, in vivo, and clinical validation of my model would represent an enormous research undertaking, in part because many of these pathways are difficult to tease apart. In terms of hypotheses, it seems most obvious and straightforward of all possible explanations.

I’m interested in any analysis that would indicate something to the contrary - let me know in the comments.

1. These people are being morally indulgent and topically stupid.