YouTube2Text

Prolactin Is THE Root Cause of Hair Loss?! Absci’s New Drug May Prove It — Transcript

by Perfect Hair Health · 9,914 words · 1,504 segments · language en · Watch on YouTube

Full transcript

  1. 0:00Decades of research tells us that the
  2. 0:01hormone DHT is the primary cause of
  3. 0:04pattern hair loss. But what if that's
  4. 0:07wrong? This isn't some hypothetical
  5. 0:09anymore. This is the position of a
  6. 0:10growing number of researchers, all
  7. 0:13stemming from the craziest monkey hair
  8. 0:15regrowth photos I have ever seen.
  9. 0:18Results achieved with a drug that didn't
  10. 0:20even touch DHT. Now, a similar drug made
  11. 0:24by Absi, an AI powered pharmaceutical
  12. 0:26company, is in phase 2 clinical trials
  13. 0:29to put this thesis to the test. I just
  14. 0:32interviewed the company's founder about
  15. 0:34their novel drug, their current clinical
  16. 0:36trials, their forthcoming results, which
  17. 0:38we'll get a peek at at the end of this
  18. 0:40year and whether DHT is the root cause
  19. 0:42of pattern hair loss, or if perhaps
  20. 0:45another hormone is actually to blame.
  21. 0:47One that once we adequately target might
  22. 0:50unlock unprecedented levels of regrowth.
  23. 0:52Stick around, we're going to be covering
  24. 0:53it all. And for those new here, my name
  25. 0:55is Rob English. I research hair loss. I
  26. 0:57publish and review scientific papers
  27. 0:59about hair loss disorders. I serve on
  28. 1:01the editorial board of a dermatology
  29. 1:02journal and I make videos like this for
  30. 1:04anybody who is fighting hair loss and
  31. 1:06looking for a path forward based on the
  32. 1:08evidence.
  33. 1:13This video is about the company Absi,
  34. 1:16their novel hair loss drug ABS 2011, and
  35. 1:18the hormone that that drug targets. It's
  36. 1:21not DHT, it's prolactin. We'll get into
  37. 1:24what this hormone is, how this drug
  38. 1:26works, if the hype is real, and never
  39. 1:28before released data from ABSI's
  40. 1:30founder, Shawn Mlan, who we actually
  41. 1:32interviewed for this video. We talked
  42. 1:34for 2 hours. We asked him some very hard
  43. 1:37questions. For those short on time,
  44. 1:38please skip to this timestamp for the
  45. 1:41key takeaways on prolactin and hair
  46. 1:43loss, ABS 2011, and whether I think this
  47. 1:45drug will succeed. And for everybody
  48. 1:48else, stick around. I'm about to tell
  49. 1:50you a story that is going to sound like
  50. 1:52a Hollywood hero's journey about hair.
  51. 1:55And it actually begins nearly a decade
  52. 1:57ago, not with Absai, but with another
  53. 1:59pharmaceutical company, Bear, and an
  54. 2:02experiment that got some very unexpected
  55. 2:04results and changed the trajectory of
  56. 2:06their research project for many years.
  57. 2:12Back then, Andreas Bush, a researcher at
  58. 2:15Bear, wanted to see if lowering
  59. 2:17prolactin levels might help improve a
  60. 2:19condition called endometriosis. So, his
  61. 2:22team set up an experiment in mice
  62. 2:23[music] and injected them with a drug
  63. 2:25that Bear developed that would later be
  64. 2:27called HMI 115. We'll go deeper into
  65. 2:30this later, but at this stage, all you
  66. 2:32need to know is that this is a drug that
  67. 2:34once injected helps to lower levels of
  68. 2:37prolactin signaling in your cells. After
  69. 2:40injecting these mice, Andreas saw
  70. 2:42something unexpected.
  71. 2:44The mice who received HMI 115, the ones
  72. 2:47who had lowered prolactin activity, they
  73. 2:49actually saw faster hair regrowth
  74. 2:52compared to the mice who received a
  75. 2:53placebo injection. And so he took these
  76. 2:55findings to bear and said, "I think we
  77. 2:58might have something here. I think we
  78. 2:59might have a candidate for a hair loss
  79. 3:01treatment." Now, you might be thinking
  80. 3:02at this stage, big deal. And I actually
  81. 3:05agree because everything regrows hair on
  82. 3:08mice. peppermint oil, copper peptides,
  83. 3:11sugar molecules. And more often than
  84. 3:13not, those same interventions inevitably
  85. 3:16fail to regrow hair appreciably in
  86. 3:18humans because mice are not men. Mice
  87. 3:21don't naturally develop male pattern
  88. 3:23hair loss. And genetically spliced mice
  89. 3:26made to mimic that condition, they even
  90. 3:28still remain terrible proxies for
  91. 3:30treatment success because candidly, we
  92. 3:32don't even know all the genes involved
  93. 3:34in male pattern hair loss to begin with.
  94. 3:36For more information, watch our video on
  95. 3:38evidence quality. Suffice it to say that
  96. 3:40mouse models rarely translate to human
  97. 3:43success with respect to androgenic
  98. 3:44alipcia treatments. But there is another
  99. 3:47animal that isn't a human and it does
  100. 3:49naturally go bald. Stumptailed macak.
  101. 3:52And while monkeys also aren't men, they
  102. 3:55tend to fare a bit better, at least as
  103. 3:57surrogates for research translatability
  104. 4:00doing treatments on hair loss. And
  105. 4:01Andreas felt that his mouse model
  106. 4:03results were strong enough that he
  107. 4:05convinced Bear to commission a study on
  108. 4:08stumptailed [music]
  109. 4:08macaks, inject them with prolactin
  110. 4:11receptor antibodies, monitor to see what
  111. 4:14happens with their scalps, and see if
  112. 4:16there truly might be something here. And
  113. 4:19this is where things get a little wild
  114. 4:22because after 6 months of treatment, the
  115. 4:25monkeys, they appeared totally fine. So,
  116. 4:27no issues related to adverse events that
  117. 4:29we know about except they had regrown a
  118. 4:32ton of hair and all by suppressing
  119. 4:35prolactin signaling in cells, not by
  120. 4:38touching DHT. And not only that, but
  121. 4:40when the monkeys actually were withdrawn
  122. 4:42from treatment, they were monitored for
  123. 4:44an additional 4 years. And to
  124. 4:46everybody's shock during that time
  125. 4:48period, they actually continued to
  126. 4:50regrow hair. And not only that, but the
  127. 4:53older monkeykey's hairs that had turned
  128. 4:55gray had actually repigmented. They went
  129. 4:57from gray back to colored. And why do I
  130. 5:00find this crazy? Because in all other
  131. 5:03monkey and human hair loss research that
  132. 5:05I have ever seen. I've first never seen
  133. 5:08this level of regrowth for hair
  134. 5:09recovery. And I've never seen this level
  135. 5:11of hair repigmentation. And in all other
  136. 5:14monkey and human hair loss research,
  137. 5:16when a treatment is officially
  138. 5:18withdrawn, you inevitably lose all the
  139. 5:21hair that you've regained. But in this
  140. 5:23study, the opposite happened. Despite
  141. 5:26the monkeys having stopped treatment,
  142. 5:28they continued to regrow hair more and
  143. 5:30more. They continue to get better hair
  144. 5:32pigment better and better. That is
  145. 5:34crazy. And it's something to get really
  146. 5:36excited about. And it was these findings
  147. 5:38that prompted Bear to license their drug
  148. 5:41to Hope Medicine and expedite studies in
  149. 5:44humans. In 2023, a phase 1 and phase 2
  150. 5:47clinical study was launched. And for the
  151. 5:49next 12 months, everyone, including me,
  152. 5:52waited for the results, thinking this
  153. 5:55might be it. This could be the
  154. 5:57breakthrough everybody has been waiting
  155. 5:59for. This could actually be the thing
  156. 6:01that takes somebody from nearly fully
  157. 6:03bald to fully haired. And then
  158. 6:07nothing, nothing, nothing at all. In
  159. 6:092024, that research group announced that
  160. 6:12the phase 2 clinical studies had been
  161. 6:14completed. And then the company never
  162. 6:16even bothered submitting the results of
  163. 6:19the report. And then in 2025, a press
  164. 6:22release was issued about the phase 1
  165. 6:23human study on HMI 115 on a whopping 16
  166. 6:27people. The findings, it appeared safe,
  167. 6:30but it only regrrew about 14 hairs per
  168. 6:32square centimeter. That's a far cry from
  169. 6:34the 100 plus hairs per square centimeter
  170. 6:37on average seen in that monkey study.
  171. 6:40And all of a sudden, enthusiasm
  172. 6:41surrounding HMI5 and even prolactin's
  173. 6:44role in hair loss all but collapsed.
  174. 6:47Well, all except for maybe one person,
  175. 6:50and that was Andreas Bush. He contended
  176. 6:53that the hypothesis on prolactin and
  177. 6:55hair loss still could be right, but that
  178. 6:58the monkey study did not translate to
  179. 7:00the humans not because of differences in
  180. 7:02hair loss pathology, but instead because
  181. 7:05he thought the humans were underdosed,
  182. 7:09that their prolactin wasn't adequately
  183. 7:11reduced in the way that it was for the
  184. 7:13monkeys, and that the pharmacocinetic
  185. 7:15data later released by Bay
  186. 7:17Pharmaceuticals, according to him,
  187. 7:19actually proved it. And this is where
  188. 7:21Andreas Bush's journey moves from Bear
  189. 7:23to another company, ABSI, and from HMI
  190. 7:271115 to a partnership with Absi's
  191. 7:30founder, Shawn Mlan, and his research
  192. 7:32team to develop a better drug, one that
  193. 7:36fixed the problem in the original study
  194. 7:38on HMI 115, and one that would finally
  195. 7:41put to rest the thesis about prolactin
  196. 7:43and whether it was actually at the root
  197. 7:45cause of hair loss. And this is where
  198. 7:48our conversation with Shawn Mlan begins.
  199. 7:51I'll present this interview in two
  200. 7:53stages. First, I'm going to get out of
  201. 7:55Shawn's way. He has a compelling case to
  202. 7:58make about prolactin's role in hair
  203. 7:59loss, ABS 2011, their ongoing clinical
  204. 8:02trials, and that steelman argument
  205. 8:05deserves to be presented uninterrupted.
  206. 8:08And only after that will I then show you
  207. 8:10my questions to Shawn and how that
  208. 8:12altered aspects of our discussion.
  209. 8:14Prolactin
  210. 8:16has
  211. 8:17maybe had a a bad rap and I think it was
  212. 8:20just due to uh the naming you know
  213. 8:23prolactin is you know prolactation
  214. 8:25everyone thinks of it as a uh lactation
  215. 8:28hormone or or in women's health but what
  216. 8:31you see is that u prolactin is expressed
  217. 8:35all throughout the the body both in
  218. 8:37females but as well as males. You you
  219. 8:40actually see it in in the scalp. You see
  220. 8:42it in the the your joints. Uh you you
  221. 8:45see it in in in your heart. Uh in your
  222. 8:48bones it and it and it plays a really
  223. 8:51interesting uh role.
  224. 8:52>> Sean is describing our evolving
  225. 8:55understanding of the hormone prolactin.
  226. 8:57Originally only seen as a female
  227. 8:59hormone. Researchers have now found that
  228. 9:01prolactin is actually all over the body
  229. 9:03in men and in women. And its role in
  230. 9:06human health is not just relegated to
  231. 9:08lactation. It looks like it sits on this
  232. 9:11uh stress inflammatory uh axis, you
  233. 9:15know, in diseases like indometriosis um
  234. 9:18or even androgenic alopeesia. It's
  235. 9:19driving uh you know, in inflammation.
  236. 9:22Essentially, we have new work that
  237. 9:24hasn't been published yet, but it looks
  238. 9:25like, you know, it does have a role in
  239. 9:27in in metabolism. I think we're just
  240. 9:29starting to realize how important um
  241. 9:32prolactin's uh role is in in both male
  242. 9:36and females and you know in in the role
  243. 9:39of you know androgenic alipcia,
  244. 9:41autoimmune diseases uh and uh and and
  245. 9:45just kind of immunology in general. So
  246. 9:47Shawn sees prolactin as a hormone that's
  247. 9:49often elevated in conditions linked to
  248. 9:52inflammation, endometriosis, potentially
  249. 9:54some metabolic disorders, even
  250. 9:56androgenic alopecia. And at this stage,
  251. 9:59it's important to understand how
  252. 10:01prolactin is even arriving to these
  253. 10:03sites in the first place. Well, first,
  254. 10:05prolactin is made by our pituitary
  255. 10:07gland. That's the p-sized organ that
  256. 10:09sits at the base of our brain. And that
  257. 10:11prolactin will enter our bloodstream and
  258. 10:13then travel through a highway network to
  259. 10:16other places in the body where it can
  260. 10:18then attach to other organ sites and
  261. 10:20begin to influence the behavior of those
  262. 10:23cells. But according to Shawn, that's
  263. 10:25not the only way that our bodies make
  264. 10:27prolactin. There's also prolactin that
  265. 10:29our own organs can manufacture on site.
  266. 10:32And it's this production of prolactin
  267. 10:34that varies wildly in individuals. And
  268. 10:37it means that two people can have the
  269. 10:39same exact levels of blood prolactin,
  270. 10:41but then if you actually look at their
  271. 10:43tissue levels, they can be wildly
  272. 10:46different. So you have the classical
  273. 10:48pituitary
  274. 10:49um expression of of prolactin. Uh but
  275. 10:52then you also additionally have and and
  276. 10:55a lot of people don't un or or know this
  277. 10:57is that prolactin sits on a uh an
  278. 11:01additional promoter that controls the
  279. 11:04extra pituitary
  280. 11:06um uh expression. And so you could have
  281. 11:09normal systemic levels of prolactin that
  282. 11:13come from the pituitary, but an
  283. 11:15increased level of prolactin uh within
  284. 11:18the scalp and and it's being driven off
  285. 11:21of that extra pituitary promoter. And
  286. 11:23what's really interesting is that the
  287. 11:26transcriptional uh uh factors that that
  288. 11:30drive the extra pituitary u expression
  289. 11:33are completely different than what you
  290. 11:35see in um in the pituitary. So they're
  291. 11:39completely uh independent and I think
  292. 11:41that that's why for a long time
  293. 11:45this hasn't come up as a um potential
  294. 11:48you know prolactin being a potential
  295. 11:50cause of of hair loss cuz we've been
  296. 11:53looking at at you know systemic blood
  297. 11:55levels uh versus looking at what is the
  298. 11:58the actual expression within the scalp
  299. 12:01and you can have really high expression
  300. 12:02in the scalp but have completely normal
  301. 12:05um prolactin levels. One of the things
  302. 12:06though that we don't really know is that
  303. 12:09how does systemic prolactin
  304. 12:12uh in addition contribute to what's
  305. 12:15being locally um produced as as well.
  306. 12:18That's one of the um you know factors
  307. 12:21that we we still don't quite understand.
  308. 12:23>> Okay. So we see elevated levels of
  309. 12:25prolactin in organs affected by
  310. 12:27inflammation and autoimmunity. But just
  311. 12:29because something is higher in a
  312. 12:30diseased organ, that does not mean that
  313. 12:32the thing that's elevated is causing
  314. 12:34that disease. For a perfect example of
  315. 12:36this, look at blood levels of C reactive
  316. 12:39protein or CRP. CRP is often elevated in
  317. 12:42people with actively evolving
  318. 12:44atherosclerosis.
  319. 12:45And so researchers once wondered if CRP
  320. 12:48was actually causing that
  321. 12:49atherosclerosis. But later research was
  322. 12:51able to tease out that this wasn't the
  323. 12:54case. That CRP was elevated as a
  324. 12:57reaction to other inflammatory
  325. 12:58processes. In other words, it wasn't a
  326. 13:01cause of the inflammation. It was an
  327. 13:02effect. it was a surrogate of the
  328. 13:04inflammation happening. So in that same
  329. 13:06vein, what exactly is the evidence that
  330. 13:08causally links prolactin to androgenetic
  331. 13:11alipcia? Because I have to be honest
  332. 13:13with you, at first glance, the idea of
  333. 13:15prolactin causing androgenetic alopecia
  334. 13:18sounds crazy because the name of the
  335. 13:20condition describes its causes.
  336. 13:23Androgens, male hormones like DHT, and
  337. 13:26genetic, your genes. And it's not like
  338. 13:28somebody just gave this condition this
  339. 13:30name and called it a day. These causes
  340. 13:33were established over 100 years of
  341. 13:35research. Men who can't produce the
  342. 13:37hormone DHT, they never go bald. When
  343. 13:40you expose balding, sensitive scalp skin
  344. 13:42to DHT, it damages the hair follicles.
  345. 13:45And when you lower DHT levels in balding
  346. 13:48men with drugs like finasteride or
  347. 13:50dutastasteride, 80 to 90% of them will
  348. 13:53see a slowing, stopping, or partial
  349. 13:55reversal of their hair loss with results
  350. 13:57sustaining for decades. So you have
  351. 14:00decades of observational, mechanistic,
  352. 14:03and interventional data that all
  353. 14:05converge on the hormone DHT causing this
  354. 14:08condition. So where on earth does
  355. 14:10prolactin fit into this? And
  356. 14:12interestingly, Shawn doesn't refute
  357. 14:14DHT's role in androgenetic alopecia. He
  358. 14:17thinks it's absolutely causal. He just
  359. 14:19argues that DHT doesn't just magically
  360. 14:22appear in balding scalps. It increases
  361. 14:25in part because of prolactin. And not
  362. 14:28only that, but by targeting prolactin,
  363. 14:31he thinks we might even be able to
  364. 14:33overcome some of the biggest hurdles
  365. 14:35that prevent people from getting huge
  366. 14:37hair recoveries even after they lower
  367. 14:39their DHT.
  368. 14:42The name is actually very misleading,
  369. 14:44androgenic alipcia. It implies that it's
  370. 14:47androgen driven. Yes, androgens are a
  371. 14:49part of it, but from what we're seeing,
  372. 14:52prolactin actually sits potentially
  373. 14:55upstream of the androgen receptor. And
  374. 14:58we actually even have data that that
  375. 15:00does show that um prolactin drives uh
  376. 15:03increase in androgen uh transcript and
  377. 15:07and so it does look like it regulates uh
  378. 15:10the androgen receptor. There's a lot to
  379. 15:12unpack here, but basically Shawn is
  380. 15:14suggesting that when prolactin arrives
  381. 15:16to a balding scalp, it activates
  382. 15:18androgen receptors, which are what cells
  383. 15:21use to allow hormones like DHT, to
  384. 15:24attach to those cells and then exert
  385. 15:25their effects. And if I were to brutally
  386. 15:28simplify the old guard versus new guard
  387. 15:31or new hypothetical argument for
  388. 15:33androgenic alipcia, the old argument
  389. 15:36looks like this. after you hit a certain
  390. 15:38age, your genetics will cause DHT levels
  391. 15:41to rise in the scalp, and those DHT
  392. 15:43levels will then cause pattern hair
  393. 15:45loss. What Sean is saying, it adds a a
  394. 15:48step right here and a step right here.
  395. 15:50So, it doesn't nullify decades worth of
  396. 15:52data. It just adds steps to what's
  397. 15:54happening and makes the chain of events
  398. 15:56a little more specific. So, that is the
  399. 15:58supposition.
  400. 15:59But what is the evidence to support that
  401. 16:02supposition? In our interview, Shawn
  402. 16:04builds his case starting with
  403. 16:06observational data on genes, prolactin,
  404. 16:09and severities of balding. What we did
  405. 16:11here was we we took a look at the uh UK
  406. 16:15bio bank and uh evaluated um how changes
  407. 16:22in prolactin uh expression or prolactin
  408. 16:26receptor expression uh changed the the
  409. 16:31uh the outcome of of balding. And and
  410. 16:34what we saw was was really really uh
  411. 16:37pretty uh uh remarkable was that in
  412. 16:41those that were not balding or had very
  413. 16:44little balding, they had much lower
  414. 16:47prolactin receptor uh levels. And those
  415. 16:51individuals
  416. 16:53that had, you know, moderate to severe
  417. 16:55balding, you saw uh an increase in the
  418. 16:59overall um prolactin receptor uh
  419. 17:03expression.
  420. 17:04>> So to reiterate, Sean's team took
  421. 17:06genetic data from over 165,000 men in a
  422. 17:10database and then he focused an analysis
  423. 17:12on genes that control for prolactin
  424. 17:14receptor expression in cells. Now, some
  425. 17:17genes will relate to more prolactin
  426. 17:19receptor expression, others less. And
  427. 17:22then his team compared those genes to
  428. 17:24the men's self-reported severities of
  429. 17:26hair loss. Zero being no hair loss, four
  430. 17:29being severe hair loss. And they found
  431. 17:31that men with little to no hair loss
  432. 17:33also happened to have genes linked to
  433. 17:35lower prolactin receptor expression. And
  434. 17:38men with lots of hair loss had genes
  435. 17:40linked to higher prolactin receptor
  436. 17:42expression, which they also plotted in
  437. 17:44this chart. The thing that's pretty
  438. 17:46remarkable about this is these are like
  439. 17:48very very small changes in prolactin
  440. 17:51receptor expression. Uh and by very
  441. 17:56small changes, you're still able to
  442. 17:58actually see an effect in in male
  443. 18:01pattern baldness. Uh you know, these
  444. 18:03aren't massive changes in in overall
  445. 18:05expression, just minor minor changes.
  446. 18:08But we know that association doesn't
  447. 18:09equal causation. Ice cream sales are
  448. 18:12linked to shark attacks, but ice cream
  449. 18:13sales do not cause shark attacks. The
  450. 18:16confounder is the warm weather getting
  451. 18:18people to buy ice cream, but also to go
  452. 18:20to the beach and also into the ocean. So
  453. 18:23observation by itself is not enough. And
  454. 18:26then we have to ask what else is there
  455. 18:28that might causally link prolactin to
  456. 18:31androgenic alopecia. Well, even before
  457. 18:33Absai and Bear thought to target
  458. 18:35prolactin for androgenic alopecia
  459. 18:37research, Shawn actually told me about a
  460. 18:39research paper from 2006 by Ralph Paws
  461. 18:42and it showed in biopsies of human hair
  462. 18:45follicles that hair follicles could make
  463. 18:47prolactin on site and that prolactin
  464. 18:50could also trigger those hairs to shed.
  465. 18:53So that's a little data, but more
  466. 18:55interesting is some of the mechanistic
  467. 18:57work that Shawn and his team later did
  468. 18:59to start to tease out cause and effect
  469. 19:01between prolactin and balding. Some of
  470. 19:04which has not yet been made public until
  471. 19:06this interview. What we did here was we
  472. 19:08actually took uh human uh scalp biopsies
  473. 19:12uh of both male and female and what I'm
  474. 19:15showing here is is male and we then um
  475. 19:19cultured it in uh with ABS 2011 uh which
  476. 19:23is uh our antiproactin receptor um
  477. 19:26blocking antibbody and uh additionally
  478. 19:30uh um cultured it with prolactin and
  479. 19:32then had a rescue arm where we added
  480. 19:34both prolactin as as well as uh ABS
  481. 19:382011. And uh what you see here is uh how
  482. 19:43much hair uh was grown uh in a 3-day um
  483. 19:47period. And this is, you know, the the
  484. 19:49same um type of measurement uh that is
  485. 19:52actually going to be done uh in uh the
  486. 19:55clinical study uh looking at um the
  487. 19:58photography and seeing how much uh hair
  488. 20:00was actually grown. And you can actually
  489. 20:01see here um within a 3-day time period,
  490. 20:05ABS uh 2011 was able to uh increase hair
  491. 20:09growth uh and you you you saw more hair
  492. 20:13follicles go into the uh antigen state
  493. 20:16and um prolactin um when you added it uh
  494. 20:20essentially shunted the overall growth.
  495. 20:23And by adding both prolactin and uh the
  496. 20:26the receptor uh uh or and and the
  497. 20:29antibbody, you're able to rescue uh the
  498. 20:32the follicle.
  499. 20:33>> So here Sean's team took scalp biopsies
  500. 20:35from three balding men. And then they
  501. 20:37exposed their skin to either nothing at
  502. 20:39all or their prolactin receptor
  503. 20:41antibbody drug ABS 2011 or more
  504. 20:44prolactin or more prolactin plus ABS
  505. 20:482011. 3 days later, the samples exposed
  506. 20:51to ABS 2011, they grew longer hair and
  507. 20:54they still had more of their hairs in
  508. 20:55the growth phase of their hair cycle
  509. 20:57than the samples that were left
  510. 20:58untouched or given more prolactin. The
  511. 21:01implication from this Xvivbo research is
  512. 21:03that blocking prolactin might improve
  513. 21:05hair parameters in scalpkin samples of
  514. 21:07androgenic alopecia. Now, this is great
  515. 21:09and all, but growing longer hair or
  516. 21:11having hairs shed less frequently does
  517. 21:13not explain what happened in these
  518. 21:15stumptailed macaks. These monkeys saw
  519. 21:18near full hair regeneration and near
  520. 21:20full hair repigmentation.
  521. 21:23This is a level of recovery that simply
  522. 21:26does not happen in humans with
  523. 21:27androgenic alopecia. It also is very
  524. 21:29rare to see in other monkey studies as
  525. 21:31well. In fact, in humans in specific,
  526. 21:34studies show that if you take your DHT
  527. 21:36as low as you can or you stimulate hair
  528. 21:38growth with drugs like minoxidil, on
  529. 21:41average, the best that you can expect is
  530. 21:43that you'll be able to rewind the clock
  531. 21:45on how your hair looks by around 6
  532. 21:47months to 36 months. That's an average.
  533. 21:50Some people do a little bit better, some
  534. 21:51people do a little worse, but after that
  535. 21:54point, it's like people hit a wall. You
  536. 21:57just don't get more hair back. And the
  537. 22:00question is why? And at least so far,
  538. 22:02researchers have identified three
  539. 22:04possible reasons. First, when you have
  540. 22:07advanced stages of hair loss, hair
  541. 22:09follicles will develop scarring or
  542. 22:10paraphilicular fibrosis. And this might
  543. 22:12act like scaffolding that inevitably
  544. 22:14blocks hairs from shedding out and then
  545. 22:17resizing as thicker. So, in this case,
  546. 22:19you can reduce the DHT all you want. You
  547. 22:21can stimulate the hair to grow all you
  548. 22:22want. But if you don't get rid of that
  549. 22:24scaffolding, you're not going to be able
  550. 22:25to allow for the hairs to resize as
  551. 22:27larger in subsequent hair cycles.
  552. 22:29Second, after hairs become miniaturized
  553. 22:31beyond a certain point, research
  554. 22:34suggests that they become detached from
  555. 22:36their goosebump muscle. That's the
  556. 22:37erector pilli muscle. And research so
  557. 22:39far seems to show that while we can save
  558. 22:42hairs that haven't yet detached from
  559. 22:43that muscle, we've not yet reliably seen
  560. 22:47evidence that any treatment available
  561. 22:49today can reattach hairs that have
  562. 22:51detached from their erector pilli. So
  563. 22:53hairs that are miniaturized beyond a
  564. 22:55certain point just don't seem to be able
  565. 22:57to be saved. And again, reduce all the
  566. 22:59DHT you want, growth stimulate all you
  567. 23:02want. If you can't get that attachment
  568. 23:04back, you're not going to recover those
  569. 23:05hairs. And the third possibility is that
  570. 23:08at some point in the balding process,
  571. 23:11hair follicle stem cells which tend to
  572. 23:13be preserved even in advanced stages of
  573. 23:15hair loss, they start to lose their
  574. 23:17ability to convert from stem cells to
  575. 23:20other cell types like progenitor cells.
  576. 23:22And this failure step is why current
  577. 23:24treatments might not yet be able to
  578. 23:26fully revive those miniaturaturized
  579. 23:27hairs. And so for a prolactin receptor
  580. 23:30antibbody to do something like this in
  581. 23:33humans that's not coming from hair
  582. 23:36elongation. It's not coming from a
  583. 23:38longer growth stage of your hair cycle.
  584. 23:40That's quite literally hair
  585. 23:41regeneration. It's taking a fully
  586. 23:43vevelis hair and returning it to a fully
  587. 23:45terminal hair in many cases. It's
  588. 23:46solving somehow one of these rate
  589. 23:49limiting regrowth factors that we know
  590. 23:51about in androgenic alopecia. And so I
  591. 23:53asked Shawn about this and if his team
  592. 23:54had thought about this and it turns out
  593. 23:56they had. And not only had they thought
  594. 23:58about it, but they had run an experiment
  595. 24:01and they think they might actually have
  596. 24:02identified which of these rate limiting
  597. 24:04recovery factors their treatment could
  598. 24:06address. It's the stem cell failure.
  599. 24:09>> One of the big things that that you see
  600. 24:11uh with uh balding individuals is that
  601. 24:15they lose their progenitor uh uh cells
  602. 24:18over time. Um and as you slowly lose
  603. 24:21those over time, you get more and more
  604. 24:23uh miniaturization
  605. 24:25uh that that occurs. some of the data
  606. 24:27that that we generated um where we were
  607. 24:30able to see that uh ABS um 2011 uh is
  608. 24:35able to uh actually increase um the the
  609. 24:39proliferation of the K15 uh stem cell um
  610. 24:43population uh and it prevents the
  611. 24:46apoptosis
  612. 24:48uh which is is really great to to see.
  613. 24:51You see prolactin on on on the right
  614. 24:52hand side uh does drive um apoptosis of
  615. 24:56the uh K15 uh stem cell. And what's also
  616. 25:02uh really quite interesting
  617. 25:05is uh the progenitor uh cell line. What
  618. 25:10we see is the uh CD34 cells um are are
  619. 25:16decreased um when you add prolactin uh
  620. 25:20to these uh Xvivo u uh models. Again,
  621. 25:24these are human um um scalp biopsies.
  622. 25:27And so we do believe that blocking
  623. 25:31prolactin receptor uh is is able to at
  624. 25:34least um um uh keep the K15 stem cell
  625. 25:38population where it is and and be able
  626. 25:40to stop uh the uh CD34
  627. 25:44uh stem cell population from further uh
  628. 25:46uh depleting. So we we do think that
  629. 25:49there is this this ability to actually
  630. 25:51have a regenerative uh effect here.
  631. 25:54Currently there is no therapy that is
  632. 25:58able to reverse that that that stem cell
  633. 26:02niche and you know potentially be able
  634. 26:04to reverse uh the the
  635. 26:05miniaturaturization
  636. 26:07and this is where we think ABS 2011 uh
  637. 26:10could have a a big role here. Sean is
  638. 26:13saying that this mechanistic data
  639. 26:15suggests that if we block prolactin with
  640. 26:17ABS 2011, we might also be able to stop
  641. 26:20hair follicle stem cell degradation,
  642. 26:22improve their differentiation into other
  643. 26:24cell types, and that we might be able to
  644. 26:27overcome this rate limiting regrowth
  645. 26:28factor that other hair loss medications
  646. 26:30that exist today have failed to address.
  647. 26:33And that's how we might be able to
  648. 26:34explain these monkey results. But he
  649. 26:37also went a step further and he shared
  650. 26:38data not yet released in any public
  651. 26:40setting that he feels further
  652. 26:42corroborates this hypothesis.
  653. 26:44>> One of the phenotypes that has been
  654. 26:47shown is that uh collagen 17A uh
  655. 26:51decreases over time with the the stem
  656. 26:54cell uh population. And collagen 17A is
  657. 26:59really important uh in the sense that
  658. 27:01that is what anchors the stem cell
  659. 27:04population uh to the to the actual hair
  660. 27:07follicle itself. And what you see is
  661. 27:09that this gets degraded uh over time.
  662. 27:12And so the fact that you're actually
  663. 27:15modulating the stem cells in addition to
  664. 27:19increasing collagen 17A when blocking
  665. 27:22the the prolactin receptor uh is is
  666. 27:26really quite encouraging and and showing
  667. 27:28that indeed prolactin may sit up you
  668. 27:32know uh you know you know pretty far
  669. 27:36upstream where you know prolactin is is
  670. 27:38driving the the degragation of the stem
  671. 27:40cell niche. it, you know, it uh, you
  672. 27:43know, drives uh, it doesn't replenish
  673. 27:45the collagen 17A and then it's it's
  674. 27:48driving the the inflammation that that
  675. 27:50you see as as well. And so, uh, we we
  676. 27:52are encouraged to uh, see again this is
  677. 27:57a a 6 day uh, culture. So, you you have
  678. 28:01to kind of take it for what it is, but
  679. 28:03it is a human biopsy. And the the fact
  680. 28:05that we are seeing results in 6 days is
  681. 28:10is really incredible. The fact that
  682. 28:11you're actually regrowing the the hair,
  683. 28:13showing what prolactin does, you know,
  684. 28:16showing uh how prolactin uh is driving
  685. 28:19some of the same phenotypes that that we
  686. 28:21see uh in in prior research is is is
  687. 28:24really quite exciting. And so you're
  688. 28:26you're hitting on kind of a a whole
  689. 28:29different mechanism that both minoxidil
  690. 28:32as well as uh the you know androgen
  691. 28:35receptor uh um u you know mechanisms are
  692. 28:38are unable to uh to to to ultimately
  693. 28:42address.
  694. 28:43>> So there it is. That's the hypothesis
  695. 28:45around prolactin and androgenic
  696. 28:47alopecia. In balding hair follicle sites
  697. 28:49prolactin begins to increase. This
  698. 28:52causes a whole cascade of events. In one
  699. 28:55cascade, prolactin might increase
  700. 28:57androgen receptor expression, which then
  701. 28:59increases DHT, which then damages the
  702. 29:01hair follicles, causing them to shed
  703. 29:03prematurely and then come back
  704. 29:05miniaturaturized in the next hair cycle.
  705. 29:08And simultaneously, in another cascade,
  706. 29:10prolactin may degrade collagen 17A,
  707. 29:13which anchors the hair follicle stem
  708. 29:15cell bulge to the hair follicle itself.
  709. 29:18And with enough degradation, this then
  710. 29:20prevents the conversion from hair
  711. 29:21follicle stem cells into progenitor
  712. 29:23cells, which then limits our ability to
  713. 29:25regenerate hair that's already fully
  714. 29:27miniaturaturized, especially with DHT
  715. 29:29reducing treatments like finasteride or
  716. 29:31dutastasteride or even some treatments
  717. 29:33for females that are more powerful
  718. 29:34androgen receptor antagonists like
  719. 29:36spironolactone, flutamide, and
  720. 29:38bicolutamide because these things they
  721. 29:40only target part of the equation that
  722. 29:42Shawn proposes. But by blocking
  723. 29:44prolactin from reaching the hair
  724. 29:46follicle at all with a prolactin
  725. 29:48receptor antibbody like ABS201,
  726. 29:51Shawn argues that both of these
  727. 29:54disregulation cascades might get
  728. 29:56addressed and that because we're
  729. 29:58addressing both simultaneously rather
  730. 30:00than just one and we're targeting
  731. 30:02something further upstream than just
  732. 30:04DHT, these crazy monkey regrowth results
  733. 30:07all of a sudden might become a reality
  734. 30:10for people. And with this model, the
  735. 30:12monkeykey's continued hair regrowth even
  736. 30:13after quitting treatment becomes at
  737. 30:15least partly explainable because once
  738. 30:18those stem cell bulge connections are
  739. 30:20reestablished, hairs might just keep
  740. 30:22getting thicker and thicker in
  741. 30:24subsequent hair cycles regardless of
  742. 30:26whether you're continuing that treatment
  743. 30:28because it also takes time to build them
  744. 30:30and it takes time to degrade them as
  745. 30:32well. So, it's an enticing hypothesis,
  746. 30:35but then if the hypothesis is even true,
  747. 30:38why did HMI 115 fail to regrow hair
  748. 30:40appreciably in humans? I mean, it barely
  749. 30:43got any regrowth. Well, this is where
  750. 30:46Andreas Bush, the man who ran the
  751. 30:48original HMI 115 studies in monkeys,
  752. 30:51teams up with Shawn and proposes a
  753. 30:53datadriven conviction. The hypothesis on
  754. 30:56HMI 115 and prolactin was not wrong. the
  755. 31:00humans in the study were simply
  756. 31:02underdosed. And it's this very premise
  757. 31:05that led Andreas Bush, Sha, and Absai to
  758. 31:08develop a new drug and rerun this hair
  759. 31:11loss experiment in humans in what's
  760. 31:13currently a phase 2 clinical trial going
  761. 31:16on right now. Now, here's what Shawn had
  762. 31:18to say about the differences between HMI
  763. 31:21115 and ABS 2011. There are a lot of
  764. 31:24technical terms here. I will stop and
  765. 31:26explain them as we go, but this is the
  766. 31:29core thesis behind what Sean's company
  767. 31:31hopes will become a billiondoll drug.
  768. 31:34>> What you saw in the stumptails was 90%
  769. 31:38receptor occupancy and
  770. 31:42what you saw from the HMI115 data from
  771. 31:46the P. So they they publicly released
  772. 31:47the PK data and then we modeled the RO
  773. 31:51based off of of that. And it's it's
  774. 31:54really quite quite uh uh shocking and uh
  775. 31:57I think you'll you you'll see why we're
  776. 32:00quite confident in the profile that we
  777. 32:03have.
  778. 32:04>> Okay, so what is Shawn talking about?
  779. 32:0690% receptor occupancy PK data. What
  780. 32:09does all this mean? Well, first Shawn is
  781. 32:11referring to what he describes as
  782. 32:13publicly released pharmacocinetic data
  783. 32:15on HMI 115. PK stands for
  784. 32:18pharmacocinetics. That's basically just
  785. 32:20the study of how a drug behaves in your
  786. 32:23body, where it goes, how long it stays
  787. 32:25there, that kind of stuff. And what Sean
  788. 32:27is saying is that in this study on
  789. 32:29monkeys, HMI 115 was able to block 90%
  790. 32:34of those monkeys prolactin receptors. In
  791. 32:36other words, HMI15 achieved 90% receptor
  792. 32:40occupancy. And this is important because
  793. 32:42the more prolactin receptors occupied by
  794. 32:45HMI 1115, the less that prolactin can
  795. 32:48then attach to that cell and the less
  796. 32:50that prolactin can then affect it. So
  797. 32:53the more that you block, the less
  798. 32:55prolactin will have as an influencer on
  799. 32:58that cell's behavior. And at least in
  800. 32:59this monkey study, when you hit 90%
  801. 33:02receptor occupancy for prolactin, it
  802. 33:05seems to be doing some really impressive
  803. 33:07things. But this is the critical point
  804. 33:09that Sean is about to make.
  805. 33:11>> What's really interesting about the
  806. 33:14prolactin receptor is that you have to
  807. 33:18block it pretty substantially
  808. 33:21in order to completely block it. What do
  809. 33:26I mean by this? even if you're at let's
  810. 33:28say 60 uh 70 even you know 80% um
  811. 33:33receptor occupancy you could still have
  812. 33:36full activity of the the actual um
  813. 33:40pathway until uh you block it um you
  814. 33:43know substantially.
  815. 33:44>> Now what Sean is talking about here is a
  816. 33:46concept known as receptor reserves or
  817. 33:48spare receptors. Basically, this is a
  818. 33:51biological phenomenon where for certain
  819. 33:53hormones, your cells will express a ton
  820. 33:56of receptors to capture them such that a
  821. 33:59majority of those receptors that are
  822. 34:00active probably aren't even needed.
  823. 34:02They're considered spares. One of the
  824. 34:05most dramatic examples of this is with
  825. 34:07the hormone insulin. In muscle cells,
  826. 34:10you can have 100 insulin receptors
  827. 34:12available. And yet, insulin only needs
  828. 34:15to attach to one of them to exert its
  829. 34:17maximal metabolic effects on that muscle
  830. 34:20cell. In other words, it doesn't matter
  831. 34:22if insulin attaches to just one receptor
  832. 34:25or 99 of the receptors. The cell's
  833. 34:28behavior does not change after one. So,
  834. 34:30if you wanted to create an insulin
  835. 34:32receptor blocker that was effective,
  836. 34:35you'd literally need to find one that
  837. 34:37blocked as close to 100% of insulin
  838. 34:39receptors as possible. Because if the
  839. 34:41blocker only blocks 99% of them, that
  840. 34:44medication has zero effect. And it turns
  841. 34:47out this same receptor reserve
  842. 34:49relationship may also exist with
  843. 34:51prolactin. For instance, in one study on
  844. 34:54lymphoma, prolactin hit its maximal
  845. 34:57effect on cell growth at just 35%
  846. 35:00receptor occupancy. So you could be at
  847. 35:0235% or 100% and the same growth
  848. 35:06trajectory occurred. And what this means
  849. 35:07is that if you made a prolactin receptor
  850. 35:09blocker that blocked a whopping 65% of
  851. 35:12prolactin receptors, well, in this
  852. 35:14lymphoma study, your prolactin receptor
  853. 35:17blocker would still be 100% useless. In
  854. 35:20fact, in that study, even if you blocked
  855. 35:22prolactin receptors by 80%. You'd still
  856. 35:25see prolactin exerting more than 90% of
  857. 35:27its maximal growth effect on those
  858. 35:30cells. So, think of it like failing a
  859. 35:32test. It doesn't matter if you got a 59
  860. 35:34out of 100 or a one out of 100, you're
  861. 35:37still getting an F. And so now the
  862. 35:39question becomes with HMI 115 these
  863. 35:42monkeys were suspected to hit 90%
  864. 35:46receptor occupancy RO but did the same
  865. 35:50actually happen in the human clinical
  866. 35:52trials according to Sha and Andreas Bush
  867. 35:55the answer is no. What you see from HMI5
  868. 36:00data here again this was they publicly
  869. 36:04disclosed what the PK profile was in
  870. 36:06their phase one. We then modeled out
  871. 36:08what the RORO um would be based off of
  872. 36:10the the publicly available data. And you
  873. 36:14you see that uh it takes quite a bit of
  874. 36:17time uh for the HMI115 molecule uh to
  875. 36:22actually uh ramp up to um you know
  876. 36:26roughly 70% uh receptor occupancy is is
  877. 36:29the the CAX or 72 73 and it then
  878. 36:34oscillates. Now what's the significance
  879. 36:37of the oscillation? So when you block
  880. 36:41prolactin receptor, you actually get an
  881. 36:43increase in prolactin levels. And if you
  882. 36:47then dip below a certain receptor
  883. 36:49occupancy threshold, you could actually
  884. 36:52be activating the the prolactin receptor
  885. 36:55with the increased levels of prolactin
  886. 36:58that are in the system. So you're
  887. 37:00inadvertently going in between antigen
  888. 37:04and kadagen. antigen, kadagen, antigen,
  889. 37:06kadagen. And we know that it takes time
  890. 37:10to rebuild the stem cell niche. And you
  891. 37:13need a sustained
  892. 37:15uh level of receptor occupancy for a
  893. 37:18long period of time uh to be able to
  894. 37:21rebuild that stem cell niche. And we
  895. 37:25believe this is the reason why uh HMI115
  896. 37:29did not uh achieve the the full efficacy
  897. 37:32because they they weren't able to give
  898. 37:35the hair follicle kind of a long enough
  899. 37:38time to uh actually rebuild that stem
  900. 37:40cell niche um because it was kind of
  901. 37:42constantly oscillating in and out of
  902. 37:45antigen and and kadagen um phase versus
  903. 37:48you know what you see with with our
  904. 37:50molecule is that you get uh sustained
  905. 37:53the levels of greater than 90% receptor
  906. 37:56occupancy for for 6 months uh straight.
  907. 38:00Uh and and and we're above the 90%
  908. 38:03receptor occupancy which again is what
  909. 38:05you saw in in the stumptail macak and
  910. 38:09again
  911. 38:10we believe in the mechanism based on
  912. 38:12everything we've we've told you. It just
  913. 38:15came down to a PK and dosing issue with
  914. 38:19HMI115. So now you have the core thesis
  915. 38:21behind ABS 2011. Andreas, Shawn, the
  916. 38:24broader ABSI team, they developed this
  917. 38:27drug to solve for this receptor
  918. 38:28occupancy problem. And their phase 1 and
  919. 38:31pharmacocinetic studies according to Sha
  920. 38:33prove that ABS 2011 achieves a prolactin
  921. 38:37receptor occupancy level of greater than
  922. 38:4090%. That's a lot. And being above that
  923. 38:4290% receptor occupancy threshold
  924. 38:45according to Shawn through inferences in
  925. 38:46this monkey data is the level that you
  926. 38:48need to adequately suppress prolactin
  927. 38:50levels in hair follicles which with
  928. 38:53sustained suppression will hopefully
  929. 38:55give the stem cell bulge enough time to
  930. 38:57rebuild which is what we need to achieve
  931. 39:00truly regenerative effects in humans.
  932. 39:02And it's this very premise that guided
  933. 39:05Shaun's team through pre-clinical work,
  934. 39:08then phase one human studies for safety,
  935. 39:10and now a phase 2 human study on men and
  936. 39:14women with androgenic alopecia. And
  937. 39:16this, Shawn believes is the test that we
  938. 39:19need to truly ascertain if this
  939. 39:21hypothesis surrounding prolactin and
  940. 39:23hair holds validity and if this
  941. 39:26treatment target might be the
  942. 39:28breakthrough the entire hair loss
  943. 39:30community has been waiting for. This
  944. 39:32phase 2 study, it is already underway
  945. 39:34and in December of this year, Shawn and
  946. 39:36his team plan to do a preliminary
  947. 39:3913-week read and an announcement of the
  948. 39:42data.
  949. 39:42>> The 13we data readout that's going to
  950. 39:45come in December. We have not given firm
  951. 39:48metrics on what we would like to achieve
  952. 39:51there. We have talked about it being
  953. 39:53directional. We want to see that the
  954. 39:55mechanism is is is working and that
  955. 39:57there is kind of a path to uh 30 plus
  956. 40:01hairs per square centimeter at at 26
  957. 40:03week. And so I think that we have set
  958. 40:06ourselves up in a way that again this is
  959. 40:09a more an exploratory look at how the
  960. 40:13mechanism is is performing uh to to
  961. 40:16date. Uh, additionally,
  962. 40:19we do feel good about the readout at at
  963. 40:2326 weeks, but
  964. 40:25let's say we achieve 30 plus at at 26
  965. 40:29weeks. My guess is that 9 12 months, you
  966. 40:34continue to see an increase uh from from
  967. 40:37there. Um, but you know, we are
  968. 40:39confident that
  969. 40:42you'll see, I think, strong or good
  970. 40:46results at at at 26 weeks, but that
  971. 40:48doesn't take away from the fact that you
  972. 40:50could continue to see uh hair regrowth
  973. 40:53past uh 6 months. And and the great part
  974. 40:56about that is we are actually going to
  975. 40:58be tracking these patients for a year
  976. 41:01post uh treatment. So we are going to
  977. 41:03understand like how
  978. 41:05the you know uh how how the changes are
  979. 41:08are progressing over over that year um
  980. 41:10post treatment. So I think that that'll
  981. 41:13also be a really exciting uh insight
  982. 41:15that we're going to see as well.
  983. 41:16>> And here's the part that I really
  984. 41:18appreciated about my conversation with
  985. 41:19Shawn. So I voiced to him about how
  986. 41:22frustrated I've been this year and last
  987. 41:24year with all of the press releases from
  988. 41:25other companies that are pursuing future
  989. 41:27hair loss treatments. The through line
  990. 41:30here is that there is a massive amount
  991. 41:32of data withholding, manipulation of
  992. 41:34statistics, disingenuous comparisons
  993. 41:35across hair count groups, subgroup of
  994. 41:38subgroup analyses, the absence of
  995. 41:40reporting and critical metrics like
  996. 41:41absolute terminal hair count changes.
  997. 41:43And so I just asked him directly that
  998. 41:44regardless of these 13week results if it
  999. 41:47would be possible if he could commit to
  1000. 41:49just not pulling any of these weird
  1001. 41:51statistical stunts or obscurities in the
  1002. 41:54way that data is reported. And here is
  1003. 41:55what he said. I can promise you that we
  1004. 41:58will have a non-controversial
  1005. 42:01data cut. We are not going to do a sub
  1006. 42:03population of sub population like it.
  1007. 42:05It's going to be clear-cut
  1008. 42:08uh data that will be uh easy to to
  1009. 42:12interpret.
  1010. 42:12>> And that to me is amazing news. So kudos
  1011. 42:15to Shawn for this. And in December or
  1012. 42:17January, you can bet that I'll be making
  1013. 42:19a follow-up video about that press
  1014. 42:21release and the data. I hope that it is
  1015. 42:23fantastic.
  1016. 42:25So this about wraps the Steelman case
  1017. 42:27for prolactin, its role or potential
  1018. 42:30role in androgenic alopecia, HMI 115,
  1019. 42:33why it supposedly failed, the evolution
  1020. 42:35to ABS 2011, and how lowering prolactin
  1021. 42:38might unlock near full levels of hair
  1022. 42:40regrowth. And I really want to present
  1023. 42:42Shawn and the broader data here in the
  1024. 42:44strongest way possible. I wanted to slow
  1025. 42:47down the storytelling to reiterate
  1026. 42:49information that was key to add support
  1027. 42:50where I could because he reached out to
  1028. 42:53us. He gave me a ton of time. We had two
  1029. 42:55calls. The first one was an hour. The
  1030. 42:57second one was 2 hours. He's committed
  1031. 42:59to giving public and uncontroversial
  1032. 43:01reads of ABS 2011's 13week data. And
  1033. 43:04that part of the discussion here
  1034. 43:06deserves to stand alone without my
  1035. 43:08interruptions, without my nitpicking,
  1036. 43:10because it truly is a service that Shawn
  1037. 43:12took the time to do this. But I
  1038. 43:14obviously asked Shawn a ton of questions
  1039. 43:15during our conversations because I
  1040. 43:17wanted to improve my clarity on the
  1041. 43:19strengths, but also the limitations of
  1042. 43:21the evidence presented. And in this
  1043. 43:23respect, I always think that we should
  1044. 43:24remain hopeful about future treatments
  1045. 43:26and the work going into them. But in
  1046. 43:28general, it usually helps to have a
  1047. 43:31precaution to temper expectations about
  1048. 43:34what a clinical trial is going to do.
  1049. 43:36Because as it stands, I think the data
  1050. 43:39around prolactin and hair loss is
  1051. 43:41compelling in some respects and limited
  1052. 43:44and potentially conflicting in others.
  1053. 43:46And I'll get into that level by level
  1054. 43:48right now.
  1055. 43:50First, let's start off with HMI15's
  1056. 43:53monkey data. The results are absolutely
  1057. 43:56incredible. I mean it when I say that
  1058. 43:58I've never seen monkey results to this
  1059. 44:00degree and the continued hair regrowth
  1060. 44:02post treatment withdrawal, I have never
  1061. 44:04seen that in any androgenic alopecia
  1062. 44:06study, monkey or human before. But is
  1063. 44:08this magnitude of regrowth truly
  1064. 44:10unrivaled in monkey studies? Not
  1065. 44:12necessarily. While it's not the same
  1066. 44:14recovery, take a look at this study on
  1067. 44:16stumptailed mac and minoxidil. In
  1068. 44:19humans, topical minoxidil tends to
  1069. 44:20produce relatively modest hair gains.
  1070. 44:23The cosmetic results tend to peak around
  1071. 44:24six to eight months and then you plateau
  1072. 44:27and even slowly decline thereafter. But
  1073. 44:29in this monkey study, the monkeys using
  1074. 44:32minoxidil, they just kept regrowing
  1075. 44:34hair. They regrrew hair at month three,
  1076. 44:36at month six, and at month nine. In
  1077. 44:38fact, they kept regrowing more and more
  1078. 44:40hair all the way up until they withdrew
  1079. 44:41from treatment. So these monkey models
  1080. 44:44can directionally translate to humans
  1081. 44:47with androgenic alopecia, but they're
  1082. 44:49not perfect surrogates by any means. In
  1083. 44:51fact, in my opinion, I think that
  1084. 44:53reconverting vellis hairs into terminal
  1085. 44:56hairs in monkeys is just a ton easier
  1086. 44:58than it is in humans because the
  1087. 44:59treatments that seem to do it in monkeys
  1088. 45:01like minoxidil, when you apply those to
  1089. 45:03humans, you don't get the same results.
  1090. 45:07You get the same direction in results.
  1091. 45:09You don't get the same magnitude of
  1092. 45:11results. And it's not just minoxidil.
  1093. 45:14We've also seen the magnitude of results
  1094. 45:16really decrease with other treatments
  1095. 45:18when Latanoprost or Bataprost moved from
  1096. 45:21monkey models to humans. They went from
  1097. 45:24pretty impressive results to very tiny,
  1098. 45:27ambiguous, and cosmetically
  1099. 45:29insignificant results. So again, we have
  1100. 45:32to temper our expectations. What we see
  1101. 45:34in these monkeys directionally might fit
  1102. 45:36with humans, but the magnitude of
  1103. 45:38impact, it might be very tiny. We don't
  1104. 45:40know yet.
  1105. 45:42Next, let's talk about this genetic data
  1106. 45:44that Shawn presented. Now, Shawn showed
  1107. 45:45us some charts suggesting that genes
  1108. 45:47linked to higher prolactin receptor
  1109. 45:49expression were also linked to higher
  1110. 45:51severities of balding. And Shawn noted
  1111. 45:53in the interview that all it took were
  1112. 45:55just tiny differences in prolactin
  1113. 45:57receptor expression to show this
  1114. 46:00relationship in the severities of
  1115. 46:01balding. However, this observational
  1116. 46:04relationship to me appears somewhat at
  1117. 46:06odds with the idea that prolactin is a
  1118. 46:09hormone with high spare receptors or
  1119. 46:11receptor reserves. So, for instance, if
  1120. 46:15this gene for prolactin receptor
  1121. 46:17expression is linked to little to no
  1122. 46:18hair loss, but this gene is linked to a
  1123. 46:21ton of hair loss, and yet the difference
  1124. 46:23in prolactin receptor expression across
  1125. 46:25the genes is maybe 10, 20, 30, 50, 70%,
  1126. 46:29even if it's 80%.
  1127. 46:31Then what difference does that make if
  1128. 46:33all prolactin needs to do to trigger
  1129. 46:35hair loss is occupy 10% of those
  1130. 46:39available prolactin receptors? With such
  1131. 46:41a low threshold for prolactin to
  1132. 46:42potentially have these adverse hair
  1133. 46:44effects, wouldn't that imply that micro
  1134. 46:47differences in prolacter receptor
  1135. 46:48expression shouldn't matter at all that
  1136. 46:50instead we would want to see 10 or
  1137. 46:5220fold differences in expression
  1138. 46:55activity before we start to see these
  1139. 46:56problems? And yet, we can't really
  1140. 46:59answer these questions right now because
  1141. 47:01the data is so preliminary, but it's
  1142. 47:03also really limited.
  1143. 47:05Next, let's talk about some of the Xvivo
  1144. 47:07data that Shawn presented. These studies
  1145. 47:09on balding hair follicles are great
  1146. 47:12mechanistic tools for research. But the
  1147. 47:15one relationship that they don't
  1148. 47:16establish is if all these same
  1149. 47:19directional relationships with hair also
  1150. 47:22exist in nonbalding scalp hair. Because
  1151. 47:25if prolactin also does all the same
  1152. 47:27exact things at the same exact
  1153. 47:29magnitude, it stops hair elongation. It
  1154. 47:32turns hairs from growing into resting.
  1155. 47:35If it does that the same in balding hair
  1156. 47:37samples as it does in non-balding hair
  1157. 47:40samples, then this would actually weaken
  1158. 47:42the case for prolactin causing
  1159. 47:44androgenic alopecia. So I asked Sean
  1160. 47:46about this in the interview and if they
  1161. 47:47had done experiments with non-balding
  1162. 47:49human scalp hair as well. And here was
  1163. 47:51his response really quickly. So we can
  1164. 47:54see that there's some degree of
  1165. 47:56significance with those effect sizes on
  1166. 47:58the slides you just showed. Those are
  1167. 48:00from three male scalp biopsies with
  1168. 48:03androgenic alipcia. Do you have control
  1169. 48:06data on a scalp unaffected by androgenic
  1170. 48:08alipcia as a comparator of any type?
  1171. 48:11>> Uh we do not. We took it from an
  1172. 48:13androgen sensitive uh area uh which is
  1173. 48:17the frontal temporal uh uh region. Um
  1174. 48:21but we did not have a uh healthy or a um
  1175. 48:28non uh um you know balding sensitive uh
  1176. 48:33uh area. Um so so we currently do not
  1177. 48:36have that at the moment.
  1178. 48:38>> Okay. Yeah. I'd be really curious to see
  1179. 48:40if you see the same exact drops in an
  1180. 48:43unaffected AGA scalp skin sample as well
  1181. 48:46because then that would help to tease
  1182. 48:47out whether or not those drops are truly
  1183. 48:49relevant to the mechanisms being
  1184. 48:51explored.
  1185. 48:52>> Yeah, absolutely. And uh one of the
  1186. 48:55things that I think we'll we'll we'll
  1187. 48:57see uh actually before we even get data
  1188. 48:59like that is actually how this performs
  1189. 49:01in in in humans later this year.
  1190. 49:03>> Now, for what it's worth, Sean's reply
  1191. 49:05here is fair. You could spend your whole
  1192. 49:08life running XVivo studies trying to
  1193. 49:10guess what happens in a human.
  1194. 49:11Eventually, you just need to prove
  1195. 49:13safety as best as you possibly can and
  1196. 49:15move on to human testing. But without
  1197. 49:17those non-balding control samples, I
  1198. 49:20personally don't find this mechanistic
  1199. 49:22data nearly as compelling. And the same
  1200. 49:24criticisms apply to research presented
  1201. 49:26about collagen 17A. It's a fascinating
  1202. 49:29hypothesis. New research indicates this
  1203. 49:31could be a critical piece to the balding
  1204. 49:33process and a rate limiting recovery
  1205. 49:35factor with respect to androgenic
  1206. 49:37alipcia treatment success. But without
  1207. 49:39knowing how non-balding hairs also react
  1208. 49:42and also express collagen 17a, we can't
  1209. 49:46actually distinguish if any increased
  1210. 49:48expression or decreased expression of
  1211. 49:50anything is meaningful and if it's truly
  1212. 49:54unique just to balding affected skin.
  1213. 49:56And this matters even more when we
  1214. 49:58actually revisit that original 2006
  1215. 50:00paper from Ralph Poss. And we actually
  1216. 50:02see that in that study, the effects of
  1217. 50:05prolactin on human hair follicles, they
  1218. 50:08were coming from nonbalding hair
  1219. 50:10follicles. And yet the researchers found
  1220. 50:12the same relationships.
  1221. 50:14This study used occipital regions of the
  1222. 50:17scalp during facelifts and hair
  1223. 50:19transplant surgeries. And in both cases,
  1224. 50:21those should be hairs that are
  1225. 50:22relatively protected from the balding
  1226. 50:24process that aren't as prone to DHT's
  1227. 50:27mediating effects on androgenic
  1228. 50:28alopecia. And yet, in that 2006 study,
  1229. 50:31the same relationship seemed to exist
  1230. 50:33with a non-balding human hair follicle
  1231. 50:35as we're now seeing in the balding human
  1232. 50:37hair follicle research 20 years later.
  1233. 50:41Next, there's the discussion about
  1234. 50:43safety, about lowering prolactin levels
  1235. 50:46in cells and whether or not that's
  1236. 50:48something that's safe long term. In
  1237. 50:50women, hypoprololactinemmia is
  1238. 50:52associated with an inability to lactate
  1239. 50:54and it's also associated with
  1240. 50:55infertility. Now, on their own website,
  1241. 50:57ABSI sites papers to suggest that
  1242. 51:01reducing prolactin is potentially very
  1243. 51:04safe for many adverse events, but they
  1244. 51:06mainly focus on safety metrics that are
  1245. 51:08not fertility and not lactation. And
  1246. 51:11because of this, I wanted to understand
  1247. 51:12not only which adverse events of
  1248. 51:14interest Shaun's team will be tracking
  1249. 51:16in their clinical study. I also wanted
  1250. 51:19to know how they were tracking the
  1251. 51:21adverse events themselves. And there's a
  1252. 51:23very specific reason why I wanted to ask
  1253. 51:25this. I will show you the clip here. So,
  1254. 51:27you mentioned Ferodmics just a second
  1255. 51:29ago. I've been tracking their data as
  1256. 51:31well. I think that anybody innovating
  1257. 51:33and attempting to run well-designed
  1258. 51:36studies in the hair loss space for
  1259. 51:38interventions is incredible. So, I
  1260. 51:41really appreciate what they're trying to
  1261. 51:42do. When I looked at their clinical data
  1262. 51:46and specifically their adverse events, I
  1263. 51:48saw some things that didn't make total
  1264. 51:50sense to me. So, we know that their
  1265. 51:52formulation is uh a larger dose of
  1266. 51:56minoxidil that is supposed to slow leak
  1267. 51:58into your system over time. [snorts] And
  1268. 52:00when we look at all of the clinical
  1269. 52:02studies on oral minoxidil ranging from
  1270. 52:040.25 25 milligrams and higher 60 to 90%
  1271. 52:09of participants both male and female
  1272. 52:11across literally all of the studies
  1273. 52:13report hypertriosis.
  1274. 52:15And when I looked at veroddermic safety
  1275. 52:17data I saw the hypertriosis numbers at
  1276. 52:19something like 3 to 6% which was
  1277. 52:22comparable to the placebo group. So this
  1278. 52:25means that either they have
  1279. 52:26inadvertently developed a hypertargeted
  1280. 52:29oral minoxidil that then attaches only
  1281. 52:32to the scalp and does not affect hair
  1282. 52:33anywhere else on the body which is not
  1283. 52:35listed in any of their patents or there
  1284. 52:39is something with their data collection
  1285. 52:40with respect to safety that is not
  1286. 52:43adequately capturing the real risk of
  1287. 52:45hypertriosis that we would expect for
  1288. 52:46this medication. Which then leads me to
  1289. 52:49the question about how your team in your
  1290. 52:51phase 2 clinical studies is capturing
  1291. 52:53clinical adverse events reporting. Are
  1292. 52:56these done through interviews with the
  1293. 52:59investigators? Are these done through
  1294. 53:01survey systems? Are they done through
  1295. 53:04open-ended questionnaires? Each of these
  1296. 53:06has their pros and their cons, and
  1297. 53:08there's plenty of debate as to which is
  1298. 53:10the best in clinical research and
  1299. 53:12reporting. But I'm curious just to get
  1300. 53:13your perspective on how you are tracking
  1301. 53:15these things because earlier we were
  1302. 53:17talking about libido and the potential
  1303. 53:20for no effect, maybe even benefit from
  1304. 53:23this type of therapy, but then libido is
  1305. 53:27separate from semen parameters and you
  1306. 53:29can have a strong libido and be
  1307. 53:30completely infertile. And so I'm just
  1308. 53:32kind of curious at this point for how
  1309. 53:34you guys in your phase 2 clinical
  1310. 53:36research have prespecified how you'll
  1311. 53:38collect adverse events and the specific
  1312. 53:40adverse events that you are exploring.
  1313. 53:43>> Yeah, it's a great question. So first
  1314. 53:45off, we have a fantastic chief medical
  1315. 53:50officer. Uh he was the the SVP and head
  1316. 53:54of clinical development at Vertex. uh
  1317. 53:57prior to that he was at uh uh Amgen and
  1318. 54:00you know has you know 20 plus years of
  1319. 54:03uh very hardcore clinical development uh
  1320. 54:06experience uh you know well outside of
  1321. 54:09even you know dermatology. I mean he's
  1322. 54:11he's run some of the most complex and
  1323. 54:12and hard clinical trials. He actually
  1324. 54:14ran a you know the Draavx the the NAV uh
  1325. 54:181.8 8 uh clinical study which is a pain
  1326. 54:21study and pain studies are notoriously
  1327. 54:24very very challenging and and was uh
  1328. 54:26successful in in in that study and so he
  1329. 54:29has brought a lot of rigor into uh ABSI
  1330. 54:34on the clinical development uh side and
  1331. 54:37we are wanting to make sure that uh we
  1332. 54:40have a uh robust uh study uh and you
  1333. 54:44know we believe with how we have set up
  1334. 54:47this study uh If successful, we see this
  1335. 54:49as a um potentially a a dose range
  1336. 54:52finding study where we can go on to a
  1337. 54:56phase three clinical study. and and in
  1338. 54:58terms of like the overall clinical
  1339. 55:00operations, the the protocols that that
  1340. 55:02we have in in place, uh I can say that
  1341. 55:05they're of the the highest uh standard
  1342. 55:07and uh you know, these um you know, the
  1343. 55:11questionnaires and and how they're
  1344. 55:12asked, they're they're all done uh in a
  1345. 55:15way to uh really help uh uh do it in an
  1346. 55:18unbiased uh uh way. And and we have a
  1347. 55:21really incredible clinical development
  1348. 55:23team on all the specific, you know,
  1349. 55:25details. uh we'd have to bring him on to
  1350. 55:28to kind of go through all of those. Um
  1351. 55:29but uh I I I do know though that uh it
  1352. 55:33is a is a very rigorous uh study and you
  1353. 55:36know you can even look at how we
  1354. 55:38designed the the phase 2 um clinical
  1355. 55:40study. It's a you know it's a 3:1 um um
  1356. 55:44um um placebo um uh uh controlled um
  1357. 55:48double blinded uh study. Uh and yeah,
  1358. 55:52really excited about the the overall um
  1359. 55:55uh kind of results both on the efficacy
  1360. 55:58as well as um being able to ensure that
  1361. 56:00the uh you know adverse um uh events are
  1362. 56:03are are um recorded and and uh uh done
  1363. 56:06in an appropriate manner.
  1364. 56:07>> By the way, I'll link our video on
  1365. 56:09vermics below if you caught that
  1366. 56:11comment. But in any case, Shawn
  1367. 56:13preferred in this case to defer
  1368. 56:14answering this to his head of clinical
  1369. 56:16development. Also, it is totally fine to
  1370. 56:19do that. He's the CEO of ABSI. He's not
  1371. 56:22in the weeds leading every aspect of the
  1372. 56:25clinical research projects, but adverse
  1373. 56:27event tracking and how it's reported is
  1374. 56:29one piece of a clinical trial protocol
  1375. 56:31that basically almost never gets
  1376. 56:33published publicly in a repository, but
  1377. 56:36that is super important to understand
  1378. 56:38because the way that you collect results
  1379. 56:40can drastically change the outcomes. And
  1380. 56:42those differences in results don't
  1381. 56:44actually reflect product safety
  1382. 56:45differences. they just reflect
  1383. 56:47differences in the way that data was
  1384. 56:48collected and that's the issue. So
  1385. 56:50perhaps in the future we can get a
  1386. 56:51little bit better clarity on this and
  1387. 56:53it'll be one of the things that I'd love
  1388. 56:55to focus on if and when it's time to see
  1389. 56:57ABS size clinical data or if we're able
  1390. 56:59to connect with their head of clinical
  1391. 57:01research.
  1392. 57:03Next, for veterans of the hair loss
  1393. 57:05community, you will know that we have
  1394. 57:07reached this milestone in research for
  1395. 57:10hair loss treatments many times before,
  1396. 57:12where observational, mechanistic, and
  1397. 57:15even early interventional data all point
  1398. 57:17toward a new novel breakthrough just
  1399. 57:20around the corner for androgenic alipcia
  1400. 57:22only for the human clinical trial to
  1401. 57:24fail miserably. One recent example,
  1402. 57:26prostaglandon D2 antagonism,
  1403. 57:30specifically the drug setup.
  1404. 57:32Observational data found that
  1405. 57:34prostaglandin D2 was elevated in balding
  1406. 57:36scalps. Mechanistic data suggested PGD2
  1407. 57:39stopped hair lengthening. Monkey data on
  1408. 57:41prostaglandin analoges like
  1409. 57:43latannoprostatrost
  1410. 57:45even to a certain extent minoxidil all
  1411. 57:48showed impressive hair gains. And yet
  1412. 57:50when a prostaglandon D2 inhibitor
  1413. 57:52setipin was finally tested in men with
  1414. 57:55androgenic alopecia
  1415. 57:57there was literally no effect at all. So
  1416. 57:59remain hopeful, but don't forget about
  1417. 58:02the graveyard of prior prospective
  1418. 58:05treatments that made it all the way to
  1419. 58:07this point only to still fail a clinical
  1420. 58:10trial. Statistically, it is the phase 2
  1421. 58:13human study that tends to make or break
  1422. 58:16most real contenders in the hair loss
  1423. 58:18treatment space. And lastly, this one I
  1424. 58:22really love. Even if ABS 2011 fails, it
  1425. 58:26still actually does not disprove the
  1426. 58:28hypothesis that prolactin might be a
  1427. 58:30root cause of androgenic alopecia. It
  1428. 58:32might turn out that it's still just not
  1429. 58:34powerful enough of a prolactin receptor
  1430. 58:37antibbody. Keep in mind that that
  1431. 58:39lymphoma study I talked about earlier,
  1432. 58:41even when prolactin occupied just 7% of
  1433. 58:44the available receptors, cell growth was
  1434. 58:47still at 50% its maximum speed. In this
  1435. 58:50respect, we might actually find that a
  1436. 58:52failed clinical trial on ABS 2011 simply
  1437. 58:54means that we have to build something
  1438. 58:55that suppresses prolactin even harder.
  1439. 58:58It's wild to think about that fact that
  1440. 59:00even with ABS 2011, a drug that was
  1441. 59:02designed to overcome this one liability,
  1442. 59:05we still might not be able to confirm or
  1443. 59:07rule out the legitimacy of the
  1444. 59:08hypothesis. And that is just the way
  1445. 59:11that hair loss research goes.
  1446. 59:14Okay, that is everything. I know this
  1447. 59:16video was a long one. I hope that you
  1448. 59:18enjoyed it. I hope that you learned
  1449. 59:19something. We put a ton of time into
  1450. 59:21creating it and I really wanted to thank
  1451. 59:23Sean for taking the time to reach out
  1452. 59:26and do this interview with us. I loved
  1453. 59:28our discussion. I really appreciated his
  1454. 59:30time and most importantly, I want his
  1455. 59:32product to succeed. I think that if it
  1456. 59:34does, it's going to be an incredible
  1457. 59:36compliment or even a replacement to
  1458. 59:39every other treatment out there. We can
  1459. 59:41only hope and time will only tell. We'll
  1460. 59:44know more at that 13week data read.
  1461. 59:46However, even if the results are very
  1462. 59:48tiny at 13 weeks, recall that the
  1463. 59:51monkeys took four years for significant
  1464. 59:54hair regeneration. You can't look at the
  1465. 59:56data at 13 weeks for efficacy. I think
  1466. 59:59that would be a huge mistake. What we
  1467. 1:00:01should be looking at here is whether or
  1468. 1:00:03not the prolactin receptor antibodies
  1469. 1:00:06are safe. What are people reporting in
  1470. 1:00:08terms of fertility markers, libido
  1471. 1:00:11markers, anything else that's an adverse
  1472. 1:00:14event of interest? That is what's most
  1473. 1:00:16important at this read. Even if hair
  1474. 1:00:19counting data looks flat or slightly
  1475. 1:00:21negative, I am not discouraged by that.
  1476. 1:00:23It is 13 weeks, so give it more time.
  1477. 1:00:26Speaking of time, if you are fighting
  1478. 1:00:28hair loss, you have to make do with the
  1479. 1:00:30treatments that are available today. And
  1480. 1:00:33if you're interested in pursuing
  1481. 1:00:34treatments that are hyperpersonalized
  1482. 1:00:36and evidence-based, I helped co-found
  1483. 1:00:38the brand ULO, that's ul.co,
  1484. 1:00:42to facilitate access to those
  1485. 1:00:44treatments. We're a US-based teleahalth
  1486. 1:00:46brand focused exclusively on hair loss.
  1487. 1:00:49And we deliver what I would consider an
  1488. 1:00:51unrivaled level of personalization to
  1489. 1:00:53make sure you're getting the best
  1490. 1:00:54possible outcomes that respect your
  1491. 1:00:56needs, but also your preferences. We
  1492. 1:00:58have lowd dose formulations of topical
  1493. 1:01:00finasteride and topical dutastride.
  1494. 1:01:02their full strength counterparts. We
  1495. 1:01:04have topical minoxidils paired with
  1496. 1:01:06retinoic acid or tininoan. We offer high
  1497. 1:01:09strength oral dutastasteride, standard
  1498. 1:01:11dose oral dutastasteride, standard dose
  1499. 1:01:13finasteride, and a lot of other options
  1500. 1:01:15to help facilitate your journey. I hope
  1501. 1:01:18you check out the brand and in the
  1502. 1:01:19meantime, we'll keep making content like
  1503. 1:01:20this. I hope you enjoy it. I look
  1504. 1:01:22forward to next

About this transcript

This page contains the full transcript of Prolactin Is THE Root Cause of Hair Loss?! Absci’s New Drug May Prove It by Perfect Hair Health, generated from the public captions YouTube serves with the video. The transcript has 9,914 words across 1,504 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.