The connection between thyroid hormone and hair is old, well-established clinical knowledge, not a novel discovery: both an underactive and an overactive thyroid gland are known to cause noticeable hair thinning or shedding, which is one of the reasons doctors check thyroid function when investigating unexplained hair loss. Thyroid hormone receptor beta-1 was shown to be expressed in the human hair follicle over two decades ago, and independent laboratory work has since gone further, treating isolated human hair follicles directly with thyroid hormone outside the body.

The obvious problem with simply giving someone thyroid hormone for hair loss is that systemic thyroid hormone affects the heart, bone density, and metabolism throughout the body, making it an unacceptable trade-off for a cosmetic condition. TDM-105795's proposed advantage is selectivity on two fronts: it is designed to activate TRβ specifically (a receptor subtype expressed in the follicle) rather than TRα (more associated with cardiac and bone effects), and the company reports it has poor systemic absorption and a short elimination half-life when applied topically, so that whatever reaches the follicle stays largely local and clears quickly rather than circulating. A related academic mouse study, independent of Technoderma, found that a separate experimental thyromimetic compound accelerated entry into the anagen phase via activation of Wnt/β-catenin and Sonic Hedgehog signalling in hair follicle stem cells, which is broadly consistent with, though not proof of, how TDM-105795 is proposed to work.

This is a genuinely different mechanistic category from anything else in AmpleLab's pipeline coverage: not a hormone blocker, not a Wnt-pathway activator acting through a receptor tyrosine kinase, but a receptor-selective hormone mimetic applied topically and designed to stay out of the bloodstream.

TDM-105795 completed a Phase 1 dose-escalation study (NCT04913519) in late 2021, reported as safe and well tolerated with no material issues, before moving into a Phase 2a proof-of-concept trial (NCT05802173) in men with androgenetic alopecia.

Read at face value, both doses showed a numerically larger increase than vehicle, roughly 10 hairs/cm² for the high strength and 6 hairs/cm² for the low strength, and the safety and systemic-absorption data are genuinely reassuring for a hormone-pathway drug. The absence of a disclosed p-value, and the unusually strong vehicle response discussed below, are the two things that keep this from being a clean result.

A 14.0 hairs/cm² increase in the vehicle arm, over just four months, is a large number for a group that received no active drug. Large, well-established minoxidil trials running for a comparable or longer period have found statistically significant separation between active treatment and vehicle. TDM-105795's own vehicle arm, by contrast, sits unusually close to the active doses in absolute terms. Without a published p-value for this trial, it is not possible to say from the public data whether the gap between TDM-105795 and vehicle here would hold up as statistically significant were the full analysis released.

There are ordinary, non-sinister explanations a strong vehicle response like this can have. TAHC is measured in a small fixed area (1 cm² in this trial), and small-area photographic hair counts are known to carry meaningful measurement variability between visits, from lighting, hair-clipping technique, and manual or software-assisted counting differences, which can inflate apparent change in any arm, including vehicle. Regression to the mean is also a real factor: participants are often enrolled during a period of visible thinning, and hair density in androgenetic alopecia can fluctuate somewhat over time even without treatment, so a group selected at a low point may show some natural recovery regardless of what they're applying.

None of these explanations is a criticism unique to Technoderma; they are known limitations of TAHC as an outcome measure generally, and a reason major trials favour tools like statistical significance testing and larger sample sizes precisely because a single topline vehicle number can look unusual for reasons that have nothing to do with the active drug. But it does mean the responsible reading of TDM-105795's Phase 2a data is "encouraging, incompletely reported, and worth a properly powered Phase 2b with disclosed statistics," not "proven to beat placebo."

Technoderma Medicines is a privately held, China-based biopharmaceutical company, originally located in Jiaxing and now headquartered at Chengdu Tianfu BioPark, though the TDM-105795 trials themselves were run entirely at US clinical sites under an FDA Investigational New Drug application. Beyond TDM-105795, the company's pipeline includes TDM-180935, a JAK1/TYK2 inhibitor for atopic dermatitis, which completed its own Phase 1 testing and a separate Phase 2a proof-of-concept trial in January 2025, a program that has continued to generate public updates on a fairly regular cadence.

TDM-105795 itself has not had a comparably specific update since the February 2024 Phase 2a topline release. A January 2025 company statement, made in the context of announcing the atopic dermatitis result, described TDM-105795 as having "successfully completed Phase 2a clinical testing" and being "further advancing in development," without naming a next trial, a timeline, or a funding status. Independent pipeline-tracking communities following the compound have separately noted the same gap: a completed, encouraging Phase 2a with no confirmed Phase 2b or Phase 3 update since. That is not evidence the programme has been shelved, but it is a genuine gap in the public record, and one this article can only report as unresolved rather than explain.

This article is provided for educational purposes and does not constitute medical advice. TDM-105795 is an investigational compound not approved for any use by the FDA, MHRA, or any other regulatory body. AmpleLab products are unrelated to TDM-105795 and are not intended to diagnose, treat, cure, or prevent any condition.

AmpleLab.