Season 9, Episode 2 Reflections on Hair Loss Associated with GLP-1 Drugs

A big welcome back everyone. This is the Evidence-based hair podcast. This is season nine, episode number two. This is a podcast series I've launched where I speak about emerging topics in the field of hair loss. If we haven't met before, I'm Dr. Jeff Donovan. I'm a Canadian dermatologist, specializing in hair loss. I'm also clinical associate professor at the University of British Columbia. When I'm not in the clinic, I'm usually teaching or writing about something hair-related. I direct the ebhf or the Evidence-based hair fellowship, which is a two-year training program for practitioners around the world. I also work with a team of Canadian and international researchers. Each month on this podcast, we'll take a look at some important topics in hair loss. We'll explore what we know, where the evidence is weak, where the debates continue, and where the research is heading next. I'm really glad that you're here. Thanks for joining me today. I want to remind you that the Evidence-based hair podcast is for educational purposes, and it's not a substitute for medical advice. Before we start the podcast, I want to introduce you to Deandra. Deandra is not a real person. She's a patient, fictional, that I've made up for this episode, but her story is typical, and she's a composite of all the conversations that I hear every day in my clinic. Let's say Deandra is 44. She teaches high school chemistry, and she's struggled with weight loss for most of her life. Last year, her family doctor told her that she had prediabetes, and so she started to put her cepatite once weekly, as an injection. Her dose climbed every four weeks from 2.5 to 10 milligrams, and she's super pleased with the results. Over 11 months, she lost 21 kilograms. Her blood sugars came back to normal, and when she went up and down those stairs at school, her knees no longer ached. But around month four, a new problem appeared. She developed hair loss. She started noticing hair in the shower drain, and on her pillow, and by month six, her ponytail was thinner, and she could see the scalp when she looked directly in the mirror. She tells us that she skips lunch because she's not hungry. She rarely eats meat. She tells you that her mother's hair started thinning in her 50s. And Deandra does what many people do. She looks online for topics related to hair loss from these medications. She types in ozempic hair loss. She reads hundreds of posts, and she's really frightened because some posts claim that hair does not come back. And so she arrives in the clinic with three questions. Did my medication do this? Is the cepatite responsible? Will my hair regrow? And should I stop this drug? And then she adds a fourth question. My 15-year-old daughter started some agglotide last month. Is she going to lose her hair, too? We'll come back to Deandra later in this episode. So why should we care about this topic? Why am I so glad that you're here? We should care because the number of people taking these drugs is enormous, and it keeps rising. Kote and colleagues report in the journal Pediatric Dermatology in 2026 that about one in eight American adults have used these GLP-1 related drugs by 2024. And more than 30,000 American adolescents have received one in 2023. And the trends are similar in other parts of the world. In Europe, Piratini and colleagues describe a 56.6% rise in sales volume between 2021 and 2025 with some agglotide alone rising 171.7%. And the point not to be missed here is that with numbers like those, even though it's an uncommon side effect, an uncommon side effect can become a real common reason for people to show up in the clinic. We should care about this topic because regulators have flagged this as a safety signal. In 2023, the US FDA listed hair loss as a potential signal of serious risk. Since then, the FDA went back and updated the labeling for several drugs to include hair loss. And for others, they concluded that the labeling was adequate like forewagovian and zepbound. But by December 2025, the FDA adverse event reporting system contained more than 3,000 reports of hair loss from GLP-1 users. We should care because when people experience hair loss from GLP-1 drugs, it affects how they feel. And it affects their decision to stay on these drugs in a major way. It affects self-esteem, it affects quality of life. Piratini and colleagues remind us of this almost 20% rise in reddit posts about GLP-1 experiences over one year and describe the emotional stress that comes with these changes. We should care because this field is changing fast. NARLA and NARLA in clinics and dermatology in 2026 describes several new drugs. We now have oral somaglutide for weight management. That was approved in late 2025. We have or for GLP-1, which is approved in April 2026. And now I'm reading about this triple agonist, Richard Trutide, which in phase 2 trials has produced a 25% weight loss. We don't see many reports of hair loss with these drugs. And that's probably because this field is moving along with this new information and hair loss is lagging. We should care about this topic because it teaches us a whole lot about the biology of the hair follicle. We know there's these GLP-1 receptors around hair follicles in most skin. GLP-1 signaling can influence insulin signaling, growth factors, stress hormones, and even the fat layer that supports hair follicles. And so if we understand more about how these drugs contribute to hair loss and hair growth in others, we're going to learn a whole lot about hair follicle metabolism. We should care about this topic finally because if you can connect patients with good advice, accurate advice, you're going to have a whole lot more patients stay on these drugs. You're going to have a whole lot more patients benefit from the heart benefits and the metabolic benefits that these drugs bring. Much of the hair loss is nonscarring and reversible. And so when clinicians understand more about these drugs, what to expect, how to support patients, we're going to have fewer patients scared, we're going to have fewer patients stop, and we're going to have more patients benefit. And so these are the reasons why I'm so glad that you're here. These are the reasons why I'm so glad to sit down with you today and talk about these topics in this podcast. The benefits of these drugs are large. The hair side effects are real, but our job is to understand both. Let me come back to Deandra. Her story touches almost every one of these reasons. She takes a drug from one of the most prescribed classes of medications in the world. She lost weight quickly while the eating very little protein. She's frightened from the story she reads online. And the information about her daughter adds a pediatric question. And as we'll see, research has barely started to answer and address those issues of GLP-1 related hair loss in adolescence. But to help Deandra, we need to understand more about the evidence that exists in the present day. If you've been following the literature over this past decade, you know that hair shedding after losing weight has been described long before anyone heard of some adletide in these GLP-1 agonists. What has changed is the amount of data. In the last three years, we've gained data from randomized trials, pharmacovigilant signals. We now have health record studies that include hundreds of thousands of patients. So let me turn to some of these facts and let me turn to the evidence. Let's take a look at what we actually know in the present day. So what do we know? So what are these GLP-1 receptor agonist drugs? And how do they work? Well, glucagon like peptide-1 receptor agonists or GLP-1 receptor agonists have transformed the treatment of diabetes and obesity. To understand how these drugs work, we have to understand the increase in system. Increases are hormones. They get released from the gut after eating. They help manage nutrients entering the bloodstream. There are two major ingredient hormones. GLP-1 and glucose-dependent insulinotropic polypeptide or GIP. So the two major ingredients are GLP-1 and GIP. GLP-1 is produced primarily by cells in the intestine and they get released after food is consumed. It acts on GLP receptors in many organs. In the pancreas, GLP-1 increases insulin secretion when the blood glucose levels are elevated and it suppresses glucagon. secretion, and by doing so, it reduces glucose production by the liver. The effect on insulin secretion is glucose dependent, and that means that the activity diminishes as the glucose levels fall. GLP1 also slows gastric emptying so that the nutrients entering the small intestine and the bloodstream are slowed after a meal. GLP1 has an important effect on the brain as well. So, not only the effects on the pancreas, the effects on the liver, the effects on gastric emptying, but the brain as well. It acts on pathways involved with hunger, satiety, and even food reward systems, and so it helps people feel full sooner. And so these combined effects, the reduction in appetite, the increased satiety, the slower gastric emptying, the improved glucose regulation. These can substantially reduce caloric intake and promote weight loss. Naturally occurring GLP1 in the body has an extremely short half-life, because it's rapidly broken down by the enzyme dipeptidal peptidase 4, DPP4. And so much of the native GLP1 disappears from the circulation within minutes of its release. At GLP receptor agonist medications were designed to reproduce these actions of GLP1, but remain in the body longer. And drugs like loragutide and somagutide activate the GLP1 receptor, but have been modified in ways that they resist rapid degradation, and they can exert their effects for hours or even days. The newer medications expand that concept even further, TREZepatite, for example, activates GIP and GLP1 receptors, and it's therefore a dual-increotin receptor agonist. And so, although we often call them GLP1 drugs, they actually belong to a broader family of incretin-based therapies. Their effects extend much beyond simply lowering blood glucose, they affect appetite, gastric emptying, insulin, and glucagon secretion, energy intake, body weight. And together, these effects produce widespread metabolic changes in the body. And the rapid increase in their use for diabetes and obesity has consequently raised interest in effects outside of these endpoints, including effects on the skin and the hair. So, what do we know about these drugs with respect to hair loss? Hair loss has been reported as an adverse effect of these drugs in randomized trials. And we've learned that the rates are pretty low. The clearest example comes from Surmount 1, the TREZepatite obesity trial, which was published in the New England Journal of Medicine in 2022. About 5% of participants on TREZepatite reported hair loss at each of the three doses, and that was compared to about 0.9% on placebo. In 2026, Chang and Chang combined seven randomized trials and two non-randomized studies in a publication in diabetes research and clinical practice. And people on GLP1 drugs had a roughly three times increased risk of hair loss compared to controls, with a relative risk around 3.25. Yet the pooled rate of hair loss was around 3.9%, and the authors put the overall incidence at less than 5%. We know that much of the hair loss associated with GLP1 drugs is a tealogen effluvium. And certainly the link between rapid weight loss and shedding is not new. Blackburn and colleagues described hair loss with rapid weight loss in 1976 in a letter to JAMA. So, the concept that we lose hair when we lose weight is not new. Let's take a look at the biology in simple terms. As you know, most hair follicles in the scalp are an antigen. A metabolic stress can push some of those hairs into cation and tealogen, and then two or three months later, those resting hairs can shed. Real-world data now helps us measure that risk. New-bowering colleagues in the Journal of the American Academy of Dermatology in 2025 studied more than 360,000 adults with obesity. GLP1 users had a 39% higher risk of tealogen effluvium compared to people treated with other weight loss medications or bariatric surgery. Over the years, we've come to learn that the risk appears to track pretty closely with the amount of weight loss and the speed of weight loss. I think that's a really important message today. In the Canadian monograph for Wagovi, hair loss occurred in about 5.3% of people who lost more than 20% of their body weight compared to around 2.5% of those who lost less. K&K colleagues in 2024 studied 140 patients with tealogen effluvium after weight loss. On average, those patients lost about 15% of their weight at a rate of around 3.5 kilograms per month. Women and older adults developed shedding with even smaller doses. Bariatric surgery seems to be at the extreme end. A meta-analysis by Zang and colleagues in 2021 found hair loss in almost 60% of patients after surgery. Bridges and Jackson note that hair loss on GLP1 drugs is less common and that may reflect differences in the degree of caloric restriction and also the difference in malabsorption with malabsorption being a more common phenomenon with obesity surgery. So good research is teaching us a whole lot. Over the years, we've come to learn that the hair follicle suffers when nutrition falls. These antigen follicles on the scalp divide and divide at a very high rate and they are extremely sensitive to nutritional stresses and metabolic stresses. Bridges and Jackson in 2026 remind us that when using these drugs, there's a reduction in 16 to 39% energy intake. Now we know that protein supplies some of the raw materials for keratin, iron acts as a cofactor for DNA synthesis, zinc regulates some enzymes involved in protein synthesis and cell division, and vitamin D receptor signaling helps start new cycles. And we know that some of these are altered in patients on GLP1 receptor agonists. Vitamin D, yes, iron, yes, protein, yes, and maybe some of the other factors as well. And over the years, we've been learning not only about telogen effluvium, but antigenetic alopecia as well. And for me, this is one of the most fascinating parts of understanding GLP1 related hair loss and the research that surrounds this field. Over the years, we've learned that antigenetic alopecia is a pretty frequent diagnosis in these patients and that hair shedding that takes place in these drugs can make antigenetic alopecia more apparent. And in some studies, antigenetic alopecia can be among the more common diagnoses in patients on these medications. Gupta and colleagues in 2026 in science progress argue that the reason we're seeing antigenetic alopecia is partly because there's a telogen effluvium from these drugs that unmasks patterned hair loss that was previously just subclinical. And this entire field is quite speculative as to what exactly is the contributions from antigenetic alopecia. Do these drugs cause antigenetic alopecia or do these drugs cause telogen effluvium, which unmasks antigenetic alopecia? The latter hypothesis tends to be the favorite one. It certainly could be that GLP1 drugs contribute directly to AGA, but most people feel that the AGA comes from simply becoming more visible during a shed and that it was present some clinically before and the telogen effluvium makes it more visible. What about alopecia Ariata, fascinating research about androgenetic alopecia, fascinating research about telogen effluvium, what about alopecia Ariata? Well, over the years, we've come to learn that alopecia Ariata can occur in people using these drugs, but to make a case for a true increase is still pretty weak. Elzeronian colleagues and Chad Case reports in 2025 described a 23-year-old woman who developed patches of alopecia in her third month on somagletide, shortly after her dose increased from 0.25 to 0.5 milligrams. She improved when she stopped the drug and with cortical steroids. I've been tracking this alopecia Ariata data pretty closely. One study I'm fascinated by is a study by sodium colleagues. An early 2025 preprint by SOTE and colleagues reported an increased risk of alopecia Ariata with these drugs, but the authors later revised the study in this preprint online to change the wording from alopecia Ariata to hair loss, and so the study was revised to examine hair loss more. generally rather than focusing on alopegiereata. And so the actual contribution of alopegiereata seems pretty unclear. The large Trinet X studies don't show a consistent increase risk of alopegiereata with these drugs. Herrera and Bordeaux found a higher risk in some of their work. Whereas other colleagues like Vidalon at Al found that alopegiereata actually occurred less than it did in controls. Other authors have found no association. And so an important message today is that there's not a lot of great evidence that there's a significant increase risk of alopegiereata with these drugs. So we're learning a ton about these drugs and how they affect the hair. We're learning a lot about what they do in adults, but we're learning a lot about what they do in children and adolescents. And we're learning that children and adolescents also lose hair, whether they do so more often than an adult is still not clear. Couté and colleagues published a very nice paper in pediatric dermatology in 2026. And they remind us about a study step teams, which is the adolescent somaglutide trial, which was published in the New England Journal in 2022, where hair loss occurred in about 4% of teens on somaglutide and none in the adolescents on placebo. In a small luraglutide study in the Journal of Pediatrics in 2017, 1 in 14 adolescents developed hair loss compared to none of the seven on placebo. And so exactly the statistics for hair loss and teens is still not super clear. And whether they develop hair loss more commonly than adults isn't clear. The real world data is still pretty sparse. But it is clear that children can lose hair too. We've learned a lot from the pharmacovigilance data. And yes, I'm still a fan of these kinds of databases. These kinds of hypothesis generating databases, which make us sit up straight and say, is there or is there not a signal that we should be concerned about? Over the years, we've come to learn that pharmacovigilance databases detect signals early, but they can't measure risk very well. Godfrey and colleagues published a very nice paper using the FDA Adverse Event Reporting System. They published this in the JEA-DV in 2025. They found that hair loss was more frequently reported than expected for somaglutide with a reporting odds ratio of around 2.46 and for transepatide at 1.73. Most of those reports in that FDA Adverse Event Reporting system came from consumers rather than coming from physicians. I've also been involved in research looking at this evidence more carefully. In 2026, Dr. Miranda Ranuski, who is now a dermatology resident at the University of Toronto, and a Canadian group published a systematic review of 10 studies in the International Journal of Dermatology. And among 626,000 GLP-1 users with hair outcomes that were reported about 9,900 or 1.6% had hair loss. We also identified 52 patients that had hair regrowth. The pharmacovigilance studies showed only a week association with hair loss. It was the strongest for somaglutide and transepatide. And our team concluded that these data don't prove causation exactly. So what else have we learned? Well, over the years, we've come to learn that hair loss with these GLP-1 drugs is usually temporary. It's a non-scarring form of hair loss, but we really don't have good long-term follow-up. We don't have evidence of any kind of permanent, flicular damage. And many clinicians now recommend continuing these medications and monitoring patients. But there's two caveats that deserve mention. Spendler and colleagues note that recovery after weight stabilizes has not been directly studied in GLP-1 users. And Puritinian colleagues remind us that nobody has really confirmed that hair regrowth when the GLP-1 drug is stopped. And so we don't have great long-term data on the outcomes when these drugs are continued and weight stabilizes. And what happens in those who choose to stop the truck? So we need longer-term data. So those are some of the key facts that really form the foundation about what we know about these GLP-1 drugs. Most of the evidence has been published in 2024. The last two years, in fact, has really changed the quality of evidence that we have. Studies are larger, comparison groups are better, and we've now been able to start asking which drugs carry the most risks. So what are the newest and latest findings in this field? Well, let's take a look. One of the most recent discoveries relates to the size of the risk when researchers choose a more appropriate comparison group. In 2026, Tang and colleagues from the University of Pennsylvania published a target trial emulation in the BMJ. A fascinating study, design. And so that particular phrase describes an observational study that's built to mimic a randomized trial. The authors compared adults with type 2 diabetes who started a GLP-1 drug with adults who started an SGLT-2 inhibitor or a DPP-4 inhibitor. That kind of design matters because both groups had the same disease and they used similar health care at similar rates. Each comparison included about 12,000 GLP-1 users. And after the statistical waiting, the rate of hair loss was about 6.91 per 100,000 person years on the GLP-1 drug and 5.04 on the SGLT-2 inhibitor, a hazard ratio of 1.36. And compared against DPP-4 inhibitors, the hazard ratio was 1.68. The signal came from what the authors called other non-scarring alopecia, which includes tealogen afluvium. Androgenetic alopecia, an alopecia area that did not show a clear increase in that tangen colleague study. And so the main result held across sensitivity and subgroup analysis, although the authors tell us that calibration with negative control outcomes reduced the strength of the association. And so the key point here is that the absolute risk seems low. And so by my math, that gap compared to SGLT-2 inhibitors is about 2 extra diagnoses per 1,000 person years. So it's a pretty small risk overall. Another really recent discovery relates to the risk of being diagnosed with Androgenetic alopecia on these drugs. Now, several health records from 2025 and 2026 report AGA more commonly in people using GLP-1 drugs. Singalan colleagues in the International Journal of Dermatology compared people with type 2 diabetes on GLP-1 drugs with people on metformin. The hazard ratios for AGA were 2.34 with loraglutide to 5.67 with tersepotide, although these groups were matched only for demographics. Herrera and Bordeaux in the JAD in 2026 matched 576,000 with some alkutide or tersepotide users with 576,000 metformin users and reported a relative risk of 2.41. Vidal and colleagues in JAD International in 2026 found an adjusted odds ratio of 1.62 at 6 months and 1.64 at 12 months. And so there are studies emerging suggesting increased risks of AGA diagnoses and patients on these drugs. Here's where I think our thinking has become a little more sophisticated. A five-fold increase risk sounds pretty alarming, but the absolute numbers are pretty small. In the Singal study, 0.138% of tersepotide users received a new AGA diagnosis and a genetic alopecia diagnosis compared to 0.043% of metformin users. So we have to be careful when we look at the description of risk. Lalk and colleagues in the JAD in 2026 found an absolute increased risk of 0.02 to 0.04%. And with active comparators, the tank study showed no increased risk of antigenetic alopecia, at least no signal. And so the explanation that fits best is probably that shedding unmasks patterned hair loss in people that are already predisposed. But we still have more to learn about this area. Another discovery, which is really important to think about, relates to which drug carries the strongest risk of hair loss. Two treatments stand out in the trial data, the cohort data, and that's tersepotide, number one. In the Chang and Chang meta-analysis, the relative risk of hair loss versus placebo was 5.4 for tersepotide and 2.31 for some appetite with luragutide showing no increased risk. Catrogatta and colleagues in 2026 compared drugs directly in the trial data. adx cohort. At one year, tersepatide users with type 2 diabetes had a 1.87 times the odds of tealogen effluvium compared to the older GLP1 drugs, and somatotide didn't differ from the older drugs. Something quite special about tersepatide maybe. Lane Heart in the Archives of Dermatologic Research in 2026 reported relative risks of 1.68 for tersepatide, 1.43 for somatotide versus metformin, with no increased risk of dilavutide or lyraglutide. Now, what's important to remember is we try to interpret this data as tersepatide produces some of the largest amounts of weight loss. In the NewBower study, BMI fell by 4.3 units on tersepatide and by 2.6 on somatotide, and tersepatide carried the hazard ratio of 2.53 for tealogen effluvium. The other thing to think about is high dose oral somatotide. In the Oasis I trial of oral somatotide 50 milligrams daily, which was published in the Lancet 2023, 7% reported hair loss more than in the placebo group. Training colleagues in the chat in 2024 noted that injectable somatotide, 2.4 milligrams, only one of 407 people reported hair loss. Now, because these are separate trials, we can't conclude that oral somatotide carries a greater risk of hair loss than injectable somatotide, but certainly the entire field of hair loss wonders about this concept. The evidence seems to be pointing in that direction that maybe just maybe oral somatotide has an increased risk. One of the most recent discoveries relates to what these drugs do to nutrient status. And researchers are now starting to look at this more directly. Scott Bush and colleagues in 2025 analyzed claims for 461,000 adults who started GLP1 drugs. Within 12 months, 22.4% had a new diagnosis of nutritional deficiency, most often vitamin D deficiency. A pilot study by Melissa and colleagues in 2025 tested iron absorption in 51 patients with type 2 diabetes. After 10 weeks on somatotide, iron absorption fell by a median of 13%. And nine patients had a drop of 30% or more. NARLA and NARLA in their 2026 continuing education article in JADD showed that fat free mass accounts for 25 to 39% of the weight loss on these drugs. And so protein intake, iron intake, may be declining at the same time as appetite. What's really not clear is how much these contribute exactly and how much replacement fixes the problem. There's a fascinating new research that I like to share about what these drugs might be doing in the skin. There's three emerging ideas that are very interesting. The first relates to fat around the follicle. Now, the lower part of each hair follicle in your scalp sits in a layer of fat. That's the dermal white adipose tissue. And that fat releases a protein that acts on the follicle. In 2021, NICU and colleagues grew scalp hair follicles in the lab next to this fat. And the fat releases a protein called a powder site growth factor. Powder site growth factor increases cell division in the hair matrix. It also increased wind beta catan signaling, which is a really important pathway that supports hair growth. And when researchers blocked the powder site growth factor with an antibody, those growth effects stopped. Now, the GLP1 question is this, these drugs shrink body fat. If the fat around the hair follicle is altered during weight loss, those follicles could lose some of that hepatocyte growth factor that supports growth. Now, that's not directly been shown with these drugs. But an important concept is that fat layer is being altered and that might be affecting growth. Another emerging concept relates to stress hormones. In 2021, Choi and colleagues studied corticosterone. That's the mouse version of the stress hormone cortisol. In mice, corticosterone kept hair follicle stem cells at rest. It did this by reducing the activity of a gene called gas six in the dermal papilla. The gas six protein helped stem cells start a new hair. But mice under chronic stress had more corticosterone and their follicles stayed in the resting phase longer. Now, here's the GLP1 related question. These drugs also act on GLP1 receptors in the brain and may affect stress hormone signaling in humans. And in theory, higher stress hormones from these drugs could keep more hair follicles in the resting stage. So those are things that have to be studied further. These drugs increase stress signaling. And finally, another really interesting emerging area relates to IGF1. DHT, as you know, is an important antigen involved in pattern hair loss. In 2023, Lee and colleagues reported that when DHT activates the antigen receptor, hair follicles make more of a microRNA called MIR221. A microRNA is a short piece of RNA that reduces how much of a certain protein a cell makes. Now, MIR221 reduces IGF1, which is a growth factor that supports hair growth. And people with patterned hair loss higher MIR221 levels went with lower IGF1 levels. Here's the GLP1 question. Rapid weight loss and caloric restriction might alter IGF1 signaling. So a person with patterned hair loss who loses weight fast on one of these drugs might have influences on the IGF1 pathway at the same time, particularly lower IGF1 levels. And so the studies behind these ideas related to IGF1 related to stress hormone signaling and related to the fat layer. They're really studies that have come from the lab or from mice. We don't really have great studies in humans yet. But in practice, these may be very relevant as these drugs can affect a multitude of different pathways in the scalp. Finally, another really important discovery relates to timing. Classic tealogen effluvium starts two or three months after some kind of trigger. With GLP1, the triggering factors can continue across many months because the dose often climbs and weight keeps falling. So it's not like when you have a flu. You have a flu in April and you shed in June or July. The flu resolved in a week, but the shedding takes place two or three months later. With GLP1 drugs, it's different. These drugs continue. These drug doses increase and the weight loss continues and the magnitude of weight loss increases. In the Tang study, hair loss diagnoses clustered in the first year of treatment and they peaked shortly before the 12-month mark. Vidalon colleagues found no increase risk of tealogen effluvium at six months, but a 76% increase at 12 months. Herrera and Bordeaux saw a tealogen effluvium signal emerge by one year. Catrogata and colleagues added one more clue. Patients who developed hair loss had a lower BMI at a follow-up than those who did not and the difference was largest for tealogen effluvium and in the heaviest patients. So shedding may appear at month four or at month 11. So we need to keep asking about hair loss throughout the entire first year and maybe beyond. And if you've started these drugs and you haven't lost hair by month six, you're not out of the woods as the expression goes, you may still develop hair loss. So those are some discoveries which show recent data as to how fast this field is moving. They also remind us of real disagreement. I see papers drawing opposite conclusions from similar data. That's pretty normal for a young field. So let me turn now to some controversies. Let me turn now to some of the questions that would start a pretty lively discussion if we put 10 experts in the same room. So there's certainly a lot we agree about, but there's some things that are still not clear. So what are the controversies? What are the unknowns? Well the first controversy I'd like to review asks whether the drug actually acts on the follicle or whether weight loss alone explains everything. Two trynetics studies in 2026 reached different conclusions. Lane Harden colleagues found a higher risk with some aglutide and terzapatide and no increased risk with lyrglutide and lyrglutide. They argue that the direct drug effect should appear across the entire class of drugs. And because the risk appears only with the two most effective weight loss drugs, they favored that this is a tealogen effluvium simply related to weight loss. Herrera and Bordeaux looked at BMI. In their GLP1 group, the mean BMI fell from 37.2 to 34.8, a change that they call modest. Yet the risk rose for every type of hair loss they studied. And they concluded that weight loss alone might not explain the risk. Gupta and colleagues also favor weight loss as an explanation, but they add a subtle point. The caloric deficit itself, more than the drop in weight, might be important because this might be what leaves the hair matrix short of energy. But neither side really has the data to support this. Most databases don't record changes in weight over time or protein intake. And tang and colleagues can't test whether weight loss is the reason for their findings. I think this is a super interesting debate. And I think it's likely a bit of both as we'll see in a moment. Most of it is probably explained by the speed of weight loss, but some might be due to the caloric deficit, and some might be due to the drug itself. All of these probably share the blame, but not equally. Now, as these GOP1 drugs have arrived on the market, suddenly an entire field of medicine, an entire field of hair loss suddenly thinks they understand everything about nutrition and hair, when in fact this area has been unexplored for a long, long time. We still have a long way to go to understand not only the speed of weight loss, but truly the impact of protein on hair and the entire concept of relative energy deficiency or reds. That's something that hasn't even been touched in the hair loss world, not even an ounce. But my personal feeling is that reds are relative energy deficiency, which affects many of my recreational and professional athletes in my practice is a contributor to hair loss. That's a topic for another time. The second controversy relates to which drug actually carries the most risk. As I mentioned earlier, most trials and cohort studies point towards trezepotide. The pharmacovigilant databases tell a somewhat different story. In 2026, Lee and Kim analyzed fairs reports from 2022 to 2025 in the journal Diabetes Metabolism Research and Reviews. Semaglutide was the only GLP-1 drug with a positive hair loss signal with an adjusted reporting odds ratio of 1.23. Trezepotide didn't show a positive signal with an adjusted reporting odds ratio of 0.78. Lee and Kim concluded that hair loss may be a molecule specific phenomenon. So clearly we have debate. A Vigie-based analysis by Kim and colleagues in 2025 also found a hair loss signal for Semaglutide only. Why the disagreement? Well, reporting systems count reports. They have no denominator and publicity matters as well. And Lee and Kim report what's known as the notoriety effect in which massive media attention to Semaglutide can increase the number of reports that appear in these databases. And bridges injection call this discrepancy a methodological artifact. Our own team took a pretty cautious line in that paper I reported by Dr. Miranda Brandewski. We wrote that Semaglutide and Trezepotide show the strongest associations, but without direct comparison nobody really knows whether their risk exceeds that of other GLP-1 drugs. Another controversy I'd like to review if you is this. How big is the risk of hair loss with GLP-1-gip agonists? Well, Laucan colleagues reported that for the drug class as a whole, the absolute risk increase was about 0.35%. Pretty small. Dr. Heyman in a jab commentary concluded that the benefits of these drugs outweigh the risks. And he puts it pretty bluntly. GLP-1's are winners. The trials at another layer, Godfrey and colleagues wrote that hair loss was not identified in trial data for Semaglutide and Trezepotide. And Camino Salvador and colleagues said that the major trials didn't report hair loss. Sermount one did report hair loss. Pretty amazing, isn't it? Here we're talking about hair loss today. The world is talking about hair loss non-stop. The online forums and blogs are talking about this non-stop. Yet those trials said very little, especially the early ones. Pirachinian colleagues explain another reason for this discrepancy. Many trial papers listed adverse events with an incidence of 5% or more. And so of hair loss was under that cutoff, it didn't get reported. Gupton colleagues remind us that hair loss is often not a pre-specified outcome in these trials. Even the labels differ. The European label for Trezepotide lists hair loss as common, but the European label for Semaglutide for diabetes doesn't put hair loss at all. Another controversy that keeps coming up is whether women face a greater risk of hair loss than men. And several sources suggest that they may. Gupton colleagues cite product trial data for Zepbound, which is the Trezepotide brand for weight management, with alopecia in 7.1% of women and 0.5% of men. A separate pre-print by Saudi and colleagues reported roughly double the hazard of hair loss with Semaglutide in women with no increase in men. In the Lee and Kim fairs data, most reports are in women. Lanehart tell us something different. The risk rose in both sexes and the estimates for men were at least as high as those for women. What are the reasons for these discrepancy as well? It could be that women notice hair loss more often and seek care more often as gupta and colleagues report. Women may also lose more weight on these drugs than men according to meta-analysis. And some studies have shown that women develop tealogen afluvium even at much slower rates of weight loss. My own view is that both of these factors contribute that there are probably a multitude of reasons why women are likely more affected with hair loss by these drugs. Something else that remains controversial is whether the reason that a patient is undergoing treatment with GLP-1 drugs matters or not. Does a patient treated for diabetes face the same risk of hair loss as a patient treated for obesity? Well, the papers seem to disagree. Sayah Jafarian colleagues wrote that hair loss is rarely reported in patients with diabetes or insulin resistance, and that treating high insulin levels may even improve hair density. But in affairs analysis by Danielle and colleagues, alopecia was more often seen in diabetes than in weight management. Leigh and Kim found a somatotide signal in both diabetes and obesity only studies. But some of the trial data does lean towards obesity and that the use of these drugs in obesity carries an increased risk. Chang and Chang found a relative risk of 3.59 in the obesity trials. And the European label for surge hepatitis says that hair loss mainly occurs in people with overweight or obesity related concerns. And so it's controversial. But some favor that hair loss is more common in those using these drugs for weight loss rather than diabetes. The dose might explain some of this. The approved weight management dose of injectable somatotide is somewhat higher than the diabetes dose. And Gupta and colleagues found that somatotide below 2 milligrams weekly is rarely implicated in hair loss. And so clearly our future studies of GLP1s and hair loss need to focus on men versus women. They need to focus on indication. Is it for diabetes? Is it for weight management? And highlight the dosing. And this brings us to one of our final controversies. What do we do when shedding starts? We absolutely have no clue what to do. But yet everyone seems to be an expert on that. The papers in our literature offer a variety of answers. Bridges and Jackson favor continuing and monitoring with a dose reduction or a pause only in severe cases. Sia Jafarian colleagues say that stopping is not required unless the hair loss is severe or persistent or causes market distress. Spindler and colleagues suggest that slowing the titration or holding the dose and considering a switch within a class is reasonable. Singalan colleagues suggest that in patients that already have AGA that maybe just maybe arrive you tired or do you like it? have the tide is the way to go. Peirichinian colleagues write that those reductions a reasonable thing or withdrawal can be considered, but nobody has really tested whether slower titrations can help prevent hair loss. Nutrition advice is all over the place. Spend here in colleagues recommend at least one to 1.2 grams of protein per kilogram per day. That's pretty standard, but it does vary. Some authors recommend correcting vitamin and mineral deficiencies, rather than routinely supplementing, but that's controversial. It was really interesting to speak with our EBHF group, our 120 or so fellows, about a clinical trial. That's underway, titled efficacy and safety of a drinkable nutraceutical for hair loss and patients using GLP1-gip receptor agonists. And you can see this clinical trial on clinicaltrials.gov. The clinical trial ID is NCT07484061. A clinical trial is underway of a nutraceutical in women 45 and over who are on these GLP1 drugs. And they are given this drinkable supplement, which contains vitamins and amino acids and minerals, plant extracts, and they're comparing this supplement against placebo. So it's one of the rare nutraceutical-based clinical trials that's underway for users of GLP1-gip drugs. But we really don't fully understand how best to manage these patients. When to stop, what should be the threshold for stopping? Should we change drugs? Should we lower doses? Should we-- how should we support nutrition? Should we routinely correct deficiencies before they even start? Should we try to prevent deficiencies? How often should we be doing blood work? These studies aren't clear. The other thing I'd like to mention before we close is we've spoken a lot about hair loss. But what's interesting is these drugs can help hair loss in some studies as well. Gordan and colleagues and Chad Case reports in 2024 published really a wonderful study describing a 57-year-old man with balding and insulin resistance. And within six months of starting tersepatide, his hair density improved. And his insulin-resistant scores, the HOMA, the HOMA scores, fell from 3.2 to 1.7. And so without any other kind of treatment, tersepatide helped us here. More set in colleagues in Curious 2024 described a man with fliculitis decalavans who improved with tersepatide. His year-end colleagues in Chad in 2025 described 81 women with central centrifugal secretorsial alopecia, CCCA on each GLP1 drugs, and showed that there could be a possible benefit in some cases. And so these controversies are important. But patients like Deandra, who we mentioned in the beginning, really need answers. So let me come back to her one final time. And let's apply what we've learned and then conclude. So when we met Deandra last time, she had these three questions and a fourth one about her daughter. Let me answer them one at a time. Did her medication do this? Well, the medication is most likely the trigger, but probably through rapid weight loss and low protein intake. She lost about 20% of her body weight in 11 months on tersepatide. And this is the drug with the strongest trial signal. Her shedding began around month four, roughly three months after the weight started to fall. And that fits well with the telogen effluvium. And her family history and her own widening central part raises the possibility that she has female pattern hair loss that's being unmasked by this telogen effluvium. Now, in the clinic, I would confirm that with an examination, a careful history, a pull test, I would examine with tracoscopy to see these miniaturized hairs that are suggestive of this component of antigenetic alopecia. I would order blood tests, CBC, TSH, for thyroid, ferritin for iron, vitamin D. And in many cases, also be 12 zinc, and sometimes even others like folate. I would ask her to aim for 1.2 grams or more of protein per kilogram of adjusted body weight spread across meals throughout the day. And I have a low threshold to get a dietician involved if we need it. And if a ferritin is low, below 40, 45, then we'll certainly replace iron. I'm not going to replace iron if the ferritin is 60, 70, 80. I'm not going to risk further GI upset. I'm not going to risk further constipation. If the evidence doesn't point to that being needed, is your hair going to grow? Well, for the telogen effluvium component, I expect improvement once her weight stabilizes. Within three to six months, once her weight stabilizes, but until her weight stabilizes, we do not set the clock and weight. If her tricoscopy shows, indeed, that there is antigenetic alopecia, we're going to consider treatment. I might discuss monoxidil, either topical or oral monoxidil. And in many cases, if oral monoxidil, I might start 0.6 to 5 milligrams daily, and then increase to 1.25 milligrams after many months. But that depends on each patient, whether to start topical or oral, whether to consider other treatments for antigenetic alopecia. Should she stop the drug? Well, for now, my answer's probably going to be no. Her metabolic health has improved. We might want to consider holding the current dose for a few months so the weight loss slows. In general, stopping probably is only a reasonable thing if the shedding is severe or it persists. But again, we need more guidelines to help us with these tough decisions. And what about her daughter? Well, 4% of teenagers in that step teens study head-air loss compared with none on placebo. So again, I'll encourage good nutrition. Family will need to follow shedding, follow persistent shedding. And so Deandre's case draws on most of what these papers teach us. And so before we say goodbye, let me finish with five take-home messages. So message one is that the risk is real and modest. The randomized trials show us that a few percent of people taking these drugs get hair loss. Roughly three times the rate of those taking placebo. And in one of the strongest real world studies from Tang and colleagues, there was an extra two diagnoses per 100,000 person years compared to SGLT2 inhibitors. Much of this comes from a tealogen effluvium related to weight loss. The relative risk sounds frightening, but the absolute risk is what we need to be thinking about. The absolute risk of hair loss seems small. The second message is these drugs probably differ in their risk. Tersepatide and high dose oral somagglutide probably show some of the strongest hair loss signals. Luragglutide and Luragglutides show weaker signals. But direct comparisons are pretty scarce. So choosing a drug for hair loss reasons is still a matter of clinical judgment. Some authors are suggesting that for those who have AGA already and worry about shedding, that maybe Luragglutide and Luragglutide are the way to go. But we really need better studies before we figure out if that approach is the right way. The third message is that we need to counsel about early hair loss, as well as delayed hair loss. We need to tell patients before their treatment starts that they may get shedding, often a few months after weight loss begins. But sometimes the hair loss doesn't start until closer to the one-year mark. We need to ask about antigenetic alopecia in the family. And we need to be really aware of the possibility that shedding can unmask patterned hair loss. The fourth message is we need to check blood tests. We need to check iron, the ferritin test, thyroid vitamin D, and others like B12 zinc. We need to assess protein intake and we need to replace when these are low. If hair loss seems to come from a dose escalation, we may need to pause that escalation, although that really hasn't been formally tested. We need to consider monoxidil if patterned hair loss or shedding persists and even stop if hair loss is persistent or severe. But we have to be careful about that because these medications bring real benefits for the heart and for other metabolic function. And the final take-home message is we need to stay curious about all of these unanswered questions. The alopecia area, a question, is still a little unresolved. We don't really know whether teenagers are more affected or whether women are truly more affected. Certainly seems that women are. What are the benefits of these drugs in CCCA and scarring alopecia in general? And what about in those with insulin-resistant androgenetic alopecia, especially PCOS? And we need to make sure that hair loss becomes a formal endpoint in these prospective trials and not just chasing these concepts as an afterthought. And so if there's one thing I'd like listeners to remember, it's that hair loss associated with GLP-1 therapies is not one phenomenon. For many patients, rapid weight loss, reduced caloric intake, and inadequate protein can trigger a classic telogen afluveum. For others, that shedding can uncover endogenetic alopecia that was already there. And remember that telogen afluveum and endogenetic alopecia are not mutually exclusive, they can occur together. And in a small number of patients, there can be hormonal, metabolic, or even direct flicular effects that we still don't understand. But much of what we see fits a weight loss model. And the pattern resembles really what we've long seen after bariatric surgery, although on a smaller scale. We expect shedding to improve once weight stabilizes, although this really needs more study. And hair loss does not automatically mean stopping a medication with important metabolic and cardiovascular benefits. We need to support good nutrition, correct deficiencies, look for underlying endogenetic alopecia, and consider slowing the pace of weight loss and slowing those escalations whenever appropriate. But I certainly hold this view loosely until we get a lot more evidence. We've got a lot of unanswered questions, don't we? Including the effect of these drugs directly on the hair follicle. Independent of their effects on weight loss and speed of weight loss. But ongoing studies with nutritional assessments and longitudinal follow-up are really going to help us answer many of these questions about what's really happening. I want to thank you so much for joining me for the second episode of season nine. If you've enjoyed today's episode and you've enjoyed today's discussion, I'd be delighted if you subscribed wherever you listen to these podcasts. And if you'd like to join the evidence-based hair fellowship program, simply email our team [email protected] and we'll send you application material. It's quite a remarkable journey that we share together. Our group right now runs 2026 to the end of 2027. And a brand new group starts up in 2028. If you'd like to explore this topic further, I'll put some really wonderful references in the show notes. And over the next coming months, we'll continue examining the evidence behind some of these most important and most fascinating conditions and treatments in the world of hair and scalp disorders. Until next time, keep asking good questions. Keep reading carefully and thinking critically. And I'll see you soon again on the evidence-based hair podcast. I'm Dr. Jeff Donovan. Thanks for listening. Take care. Goodbye.