Hair regrowth in male and female pattern hair loss does not involve the conversion of vellus hair to terminal hair

Clinical changes in female pattern hair loss (FPHL) and male pattern hair loss (MPHL) are characterised by a progressive alteration in the terminal hair population, typified by an increase in vellus hair and a reduction in total hair density [THD, hair per square centimetre] 1. Hair regrowth in FPHL and MPHL is attributed to the reversal of these processes 2, 3, where there is a significant increase in THD together with a significant decrease in the percentage of vellus hairs 4. However, a change in the vellus hair percentage does not reflect the absolute vellus hair change. If a baseline sample of 100 hairs per square centimetre contained 50 vellus hairs (50%) and the same site had 200 hairs per square centimetre with 50 vellus hairs after one year on treatment, the vellus hair frequency would be 25%. Does hair regrowth following effective medical treatment of FPHL and MPHL involve the conversion of vellus hair to terminal hairs? Retrospective analysis of previously published data, employing the same hair evaluation methodology, the unit area trichogram (UAT) in all studies 1. The experimental methods are detailed in the Supporting information while the vellus hair definition involved both hair diameter and length, that is a hair <40 μm in diameter <30 mm in length. Student's t-test for paired samples. Table 1 details MPHL treatments, involving 2% topical minoxidil, oral finasteride (1.25 mg), and oral finasteride together with topical 3% minoxidil plus anti-androgens. Propecia (finasteride 1 mg) was unavailable in these studies as was the higher minoxidil strength. In the 2% minoxidil study, there was no significant change in THD, vellus hair percentage or the absolute vellus hair per square centimetre. In the oral finasteride study, a mean increase in THD(+13) was found, while there was a decrease(−2) in the absolute vellus count per square centimetre. When finasteride was combined with topical 3% minoxidil plus anti-androgens, a significant (P < 0.0001) mean increase in THD(+37) and a significant (P < 0.05) mean decrease in the vellus hair percentage after 12 months was found. However, there was no change in the absolute mean vellus hair count per square centimetre. Table 2 details the changes in women exhibiting FPHL treated for 12 months with oral cyproterone acetate/ethinyl oestradiol and having a serum ferritin ≥40 μg/l; the lowest concentration in women without hair loss 5. A significant (P < 0.01) mean increase in THD(+28) was found after 12 months. However, there was no change in the absolute vellus hair count per square centimetre. In women treated with topical 5% minoxidil, there was a significant (P < 0.001) mean increase in THD(+38) after 12 months, while the absolute mean vellus hair count per square centimetre increased by only five hairs. Originally, we detailed the percentage of vellus hairs in treated and untreated subjects with FPHL and MPHL, but the significance of the absolute vellus hair count was not fully appreciated (Supporting information). Preventing further expression of the underlying genetic predisposition in MPHL occurs with oral medroxyprogesterone acetate (Supporting information) and in oral finasteride therapy where the mean THD(+13) increased while the absolute mean vellus hairs count decreased (−2). When oral finasteride was combined with topical anti-androgens and minoxidil, a significant (P < 0.0001) mean increase in THD(+37) was associated with a significant (P < 0.05) mean decrease (−3%) in the vellus hair percentage, implying that the observed increase in THD might have arisen from the conversion of vellus hairs to terminal hairs. However, despite an increase in THD of 37 hairs per square centimetre, there was no change in the absolute mean vellus hairs count per square centimetre; findings do not support the hypothesis that vellus hairs are converted to terminal hairs following effective MPHL treatment. In some women with FPHL, oral anti-androgen treatment significantly (P < 0.01) increased the mean THD(+28) but there was no change in the absolute mean vellus hair count per square centimetre. In Syrian hamsters, an androgen-sensitive animal model, minoxidil had no anti-androgen effect 6 and we would not expect minoxidil treatment to influence the underlying genetic expression in FPHL or MPHL. The recent paper of Mirmirani et al. 3 does nothing to change this position nor supports the reversal of vellus to terminal hairs following minoxidil treatment. Minoxidil does shorten the delay in the initiation of the new hair cycle 7. In addition, there is some evidence of minoxidil extending the anagen duration, shortening the extended interval between the previous telogen phase and the initiation of the new anagen phase 8, 9. These effects potentially reduce the dormancy between the exogen release of the old telogen hair and the initiation of the new anagen phase, resulting in more hair per square centimetre and influencing the significant amount of hair regrowth observed within 16 weeks of starting minoxidil treatment. Findings supported in our women with FPHL treated with topical 5% minoxidil for 12 months, where a significant (P < 0.001) mean increase in THD(+38) occurred, with no change in the absolute mean vellus hair count per square centimetre, confirming previous concerns raised by Messenger & Rundegren 7. In a prospective MPHL study, Van Neste 9 found dormant follicles but did not observe any vellus to terminal hair conversions during 24 months of oral finasteride treatment, employing a vellus hair definition of <40 μm (Fig. S2). Terminal follicular hair growth is undeniable following effective medical treatment of FPHL and MPHL, and our re-analysis clearly demonstrates this does not involve vellus to terminal hair conversions. The data support a hypothesis that in both female and male pattern hair loss, non-vellus (terminal) hair follicles enter a period of dormancy between subsequent hair cycles, which are re-activated with effective medical therapy. There is an intriguing possibility of follicular neogenesis as an alternative hypothesis, which to our knowledge has yet to be fully explored; however, the Van Neste 9 data suggest this is unlikely. We would propose there is a population of growth restricted (dormant/kenogen) non-vellus hair follicles, which are re-activated by effective medical treatments as an explanation for the increased hair growth observed in FPHL and MPHL. Our findings have a fundamental impact on the pathophysiology of hair changes occurring in patterned hair loss. We would like to thank Professor Andrew Messenger, for his helpful comments. All were involved in varying degrees with the original publications. Data analysis Rushton & Van Neste. All wrote and revised submission. The authors have declared no conflicting interests. Figure S1. Distribution of hair diameter and corresponding length for uncut hairs from four normal females. Figure S2. Eleven hairs followed for 48 months in an individual exhibiting male pattern hair. Table S1. Mean ± se values for hair length and hair diameter <40 µm obtained from normal male and female subjects and male and female patients with patterned hair (sample size 7603 hairs). Table S2. Publications re-analysed. Table S3. Changes in total hair density (THD), per cent of vellus hair and absolute vellus hair count in untreated men (mean age 26yrs) with male pattern hair loss (MPHL) after 12 or 24 months compared to baseline value. Table S4. Changes in total hair density (THD), per cent of vellus hair, and absolute vellus hair count in eight men aged between 22-34yrs and treated with oral medroxyprogesterone acetate (MPA) employing a dose between 20-60mg per day in order to maintain the serum testosterone concentration within the normal male range and untreated controls after 12 months. Table S5. Changes in total hair density (THD), per cent of vellus hair and absolute vellus hair count in untreated premenopausal women (aged 27 - 42yrs) with female pattern hair loss (FPHL) after 12 months compared to baseline. Table S6. Changes in total hair density (THD), per cent of vellus hair and absolute vellus hair count in premenopausal women (aged 30 - 41yrs) with female pattern hair loss (FPHL) treated with oral spironolactone (75-100mg daily) and untreated controls after 12 months compared to baseline Table S7. Comparison between controls (mean age 29yrs) and women with female pattern hair loss (FPHL, mean age 32yrs) for total hair density (THD), per cent of vellus hair, absolute vellus hair count, per cent of telogen hair ≤30mm in length, and the absolute telogen hair ≤30mm in length. Data S1. Trichological changes in female and male pattern hair loss. Data S2. Supplementary References. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.