ARTICLE
Auteur(s) : Dominique Van Neste
Skin Study Center, Skinterface sprl, 9 rue du Sondart, B-7500
Tournai, Belgium
For those involved in medical practice it may seem trivial to
state that a doctor’s daily work consists in the subtle balance
between art and science, deep belief and hard evidence. In this
editorial I wish to share some thoughts that we in the scientific
hair evaluation community might consider as technological
shortcomings or uncertainties around “hair measurement”. We think
that these are actually considered as part of the placebo effect as
opposed to a real “placebo effect”. Without entering into the
details and restricting our thoughts to the field of scalp hair, we
feel the placebo effect reflects an ill defined and complex
component of well-being usually associated with perception of
improvement of a condition. Besides becoming conscious of “a
change”, the triggering factor appears to be the very fact of
seeking some form of assistance and/or having a nice encounter with
someone with an empathic attitude and who is acknowledgeable in the
field of hair-care. Someone suffering from hair loss, a condition
that is subjectively perceived to adversely affect an individual’s
quality of life, and who seeks medical support and experiences
empathy will probably improve his “skin-deep” relationship with the
hair disorder, which makes him/her feel better. In order to
evaluate this perception, questionnaires and clinical categorical
or scoring systems have been developed and in this issue it is
shown by Kaufman’s team [1, 2] that these methods are performing
rather poorly in comparison with objective methods. What are these
methods and can they be improved?
Representation and classification of hair patterns as a
measurement tool?
Global views of scalp hair were already engraved on the walls of
caves (figure
1A). Some images fit the typical severity grades listed in
various classification systems developed during the second half of
the 20th century for androgenetic alopecia (AGA) (figure 1B).
Although clinicians feel comfortable using these cartoons for
diagnostic purposes, most of them consider the classifications as
being too rigid and useless for evaluating worsening or improvement
of AGA in the context of a clinical trial. To the best of our
knowledge, no one has provided evidence to support or to reject
this option. At Skinterface, we ran a test-retest experiment and
this taught us that rating of severity from a series of
standardised views of the top of the head (37 males with AGA; of
which 30 images are shown in figure 1C) can vary by 1
grade or more in 60% of cases. This means that reproducible scores
were found only in 40% of cases (Skinterface unpublished data).
Because prepubertal males have a full head of hair,
post-pubertal hair loss is undoubtedly the consequence of a dynamic
process, i.e. a time-related modification. We presume that nobody
amongst our illustrious ancestors would have thought that it would
take close to 30,000 years to conceive methods to quantify these
changes. We acknowledge that all the classical textbooks mention
that hair productivity diminishes as the follicles enter into a
chronic degenerative process associated with hair follicle
miniaturisation, as shown by scalp microscopy or histology. What
exactly do we know about the resolution of methods proposed for
establishing the kinetics of this process? Well, let us try to
answer these questions by looking to the methods reported by
Kaufman et al. in this issue in order to know where we are in 2008
[1, 2].
Clinical evaluations, global view and close-up photography:
other “validated” tools with a range of resolution power?
Measuring the progression of patterned hair loss is best performed
under placebo controlled conditions. As one can read in Kaufman’s
reports, the progression of hair loss can be appreciated by various
“validated” tools. What exactly “validated” means is not known.
Those interested in improving their understanding and/or those who
would like to claim validation for an existing method or even quote
a new “validated” technological approach, are warmly invited to
read the paper on methodological development for skin
bioengineering methods [3].
For the sake of demonstration, let us describe the resolution
power of a technique as “the time required for a given method to
detect 50% of the subjects who show a change”. In the case of an
inefficacious treatment like a placebo pill (data from Kaufman et
al. [1]) this change would mean “getting worse”. The simplified
diagram displayed in figure 1D speaks for itself
as hair counts (blue dots) cross the 50% detection threshold at
about 9 months (blue arrow) while global photography (magenta dots)
reaches this parameter at about 3 years (magenta arrow). For the
clinical observer, a period of 5 years is not enough to reach the
50% threshold. Subjectivity of the clinical observer together with
self-administered questionnaires appears to be the least effective
of methods to detect worsening of AGA in males.
Is the best method found under such conditions the best
available or can we do better?
For many clinical trials, it appears that the techniques were
selected because they had already been accepted in the past by
authorities (this term encompasses industrial, academics, experts
and health authority administrations) or because they were
published in peer-reviewed journals in order to communicate results
of said trials to the medical and scientific communities.
In the Merck sponsored trials presented here, the most sensitive
method for the detection of worsening in male AGA subjects was hair
counts on close-up photography. The method was selected in order to
evaluate a one-inch diameter circle, which displays around
5 cm2 of scalp, captured on a 24 × 36 mm film
at high resolution. The high number of hairs evaluated would reduce
the relative value of counting errors. In comparative assays
(paired measurements on 15 subjects) we found that about 10-20% of
hairs remained undetected by this photographic method (average 13%)
and also that 12 of 15 subjects had more hair with microscopy than
counts on close-up photographs, which was also sensitive to the
hair length i.e. the longer the clipped hair the better its
detection [4]. However, this close up photography does not tell us
about what type of hair was visible and, as a consequence, what is
really meant by higher counts at the end of finasteride treatment.
There is also no indication as to where these hairs came from.
While we do not have a definitive answer to this question, some
hypotheses have been proposed over the past decade and repeatedly
alluded to during lectures about finasteride. Examples being that
the hair would grow faster, thicker, longer, there might even be a
sudden reversal of miniaturized follicles which turn back into the
production of terminal hair… and we have even heard that patients
might grow “more juicy hair”! One may like these comparisons but
unfortunately there is no direct evidence to support such
statements.
In recent experiments, 11 follicles were followed at monthly
intervals for a period of 2 years to characterise the natural
changes occurring in AGA affected hair follicles, to measure
productivity and eventually characterise regression while
untreated. The same follicles were followed for a further 2 years,
using the same sampling methodology, while taking
Propecia® (finasteride 1 mg/day). The purpose of
this non-invasive investigation was to observe drug responses in a
series of well characterised end-organs. During this 4 year study
of individual hair follicle production, we were unable to observe
either the reversal of vellus hair to terminal hair or the
production of thicker hair under finasteride therapy. However, we
did observe a reduction in the duration of the lag phase between
hair cycles. This rescuing of hair follicles only occurred in hair
thicker than (50 μm) before treatment; a thickness far too
large to be classified as a vellus hair follicle [5]. The paper
finally points to the fact that interpretation from statistics of
pooled data might lead to erroneous conclusions about what a drug
does to scalp hair follicles. Our position today is to remain very
cautious about stating that vellus hair follicles can be
transformed into thick terminal hair until proper documentation
becomes available.
Because these techniques were available well before the
launching of the finasteride trials, the problems of the optical
resolution of hairs and the usefulness of hair dye application were
known to the authorities involved in the design of scalp hair
trials [6] before their commencement. It later became clear that
the academic community does not always welcome these novelties [7].
Improved hair detection after hair dyeing is described in the
following terms: “dyeing of these grey hairs may now also make
visible all non-pigmented vellus hairs that previously were
appropriately not captured by the photographic technique”. Later in
the same paper, it is mentioned that the contrast-enhanced
phototrichogram technique is more costly and there might be a
problem with subject retention because of the increased number of
visits. There is no evidence to support this opinion!
The review does not comment in terms of scalp image resolution
and hair growth kinetics. It is even suggested to make left-right
scalp biopsies at different times for evaluating therapeutic
responses. Anyone who has ever looked at the top of the head of
males with AGA will acknowledge that a human scalp as not perfectly
symmetrical.The use of “enantiomer” or mirror images for left-right
sites versus midline has to be very carefully considered, if not to
be thought of as abusive. It remains that scalp biopsy, usually
considered as the best technique [7], has two major drawbacks,
first its very invasive nature hampers repeat sampling of the same
target, and second is the fact that it is not always easy to
identify hair follicles, especially during the dormant stage
[8].
As scalp surface image processing, including automation,
evolves, it does not always lead to improvements in the resolution
of individual hairs. This was referred to in Kaufman’s papers.
Further, advocating technological problems about other methods is
unacceptable if there is no direct comparison to hand. Clearly the
point has been made (using paired comparison of the same images)
that a particular automated computerised method underestimates hair
counts for various reasons [9] and again we refer the interested
reader to the appropriate literature when validation of new
technologies is desirable [3].
Besides these reflections on photographic or optical
non-invasive methods, and in continuity with our comments as to the
best techniques to detect worsening in untreated subjects,
Kaufman’s paper alluded to other techniques such as the unit area
trichogram. The unit area trichogram described important
structural-functional characteristics and translated those
analytical parameters in clinically meaningful terms [10]. This
validated objective method appears to detect rapid worsening in
over 50% of the subjects in one year [11]. Using this method it was
shown that slightly over 50% of finasteride treated subjects had a
statistically significant increase of clinically meaningful hair
after one year of treatment, while those on minoxidil did not [12].
Interestingly, subjective impressions did not match the results of
objective measurements, in fact subjective appreciation pointed
exactly in the opposite direction in terms of drug efficacy:
minoxidil did work and finasteride did not help [12]!
Conclusion
Can we run better studies in the future using even better
techniques? The answer is definitely: yes. But are the authorities
willing to support this endpoint?
One suggestion would be that the authorities in the broadest
sense (industry, health authorities academic or private research
institutes) and without any geographical restriction (racial
variation of hair colour) use and allocate significant funds for
standardization and quality screening of methods, according to a
canvas proposed by bioengineering technology. Once a methodology is
validated and reaches the objective criteria set forth in advance,
one may define how to report changes in terms of clinically
significant benefits. Previous proposals [13] can be completed with
more recently discovered aspects of hair cycling, such as exogen
[14], with methods specifically developed for use in humans
[15].
While waiting for this ideal time, the least one can expect from
those who propose and sell tools for hair measurement, is that they
clearly indicate the limitations of their methods. As an additional
and last very personal wish: no more advertising claims on “30%
more hair in 3 months” after application of a cosmetic preparation
or claiming that a drug grows “fatter and more juicy hair”! This is
not helpful, particularly when such claims seem poorly documented
and borders on the field of trichoquackery. A field that we, as
scientists and medically qualified persons involved in hair care,
wish to stay away from or eventually leave as soon as possible…
References
1 Kaufman K, Girman CJ, Round EM,
Johnson-Levonas AO, Shah AK, Rotonda J. Progression
of hair loss in men with androgenetic alopecia (male pattern hair
loss): long-term (5-year) controlled observational data in
placebo-treated patients. Eur J Dermatol 2008; 18(4): 400-6.
2 Kaufman K, Rotonda J, Shah AK, Meehan AG.
Long-term treatment with finasteride 1 mg decreases the
likelihood of developing further visible hair loss in men with
androgenetic alopecia (male pattern hair loss). Eur J Dermatol
2008; 18(4): 407-11.
3 Bircher A, de Boer EM, Agner T,
Wahlberg JE, Serup J. Guidelines for measurement of
cutaneous blood flow by laser Doppler flowmetry. Contact Dermatitis
1994; 30: 65-72.
4 Van Neste D. Folliculogram demonstrates more anagen hair
roots in male androgenetic alopecia after one year treatment with
finasteride 1mg/d. In: Van Neste D, Blume-Peytavi U,
Grimalt R, Messenger A, eds. Hair Science and Technology.
Tournai-Belgium: Skinterface sprl, 2003: 311-6.
5 Van Neste D. Natural scalp hair regression in preclinical
stages of male androgenetic alopecia and its reversal by
finasteride. Skin Pharmacol Physiol 2006; 19: 168-76.
6 Van Neste D, De Brouwer B, De Coster W. The
phototrichogram: analysis of some technical factors of variation.
Skin Pharmacol 1994; 7: 67-72.
7 Olsen EA. Current and novel methods for assessing
efficacy of hair growth promoters in pattern hair loss. J Am Acad
Dermatol 2003; 48: 253-62.
8 Messenger AG, Sinclair R. Follicular miniaturization
in female pattern hair loss: clinicopathological correlations. Br J
Dermatol 2006; 155: 926-30.
9 Van Neste D, Trüeb R. Critical study of hair growth
analysis with computer-assisted methods. J Eur Acad Dermatol
Venerol 2006; 20: 578-85.
10 Rushton H, James KC, Mortimer CH. The unit
area trichogram in the assessment of androgen-dependent alopecia.
Br J Dermatol 1983; 109: 429-37.
11 Rushton DH, Ramsay ID, Norris MJ,
Gilkes JJ. Natural progression of male pattern baldness in
young men. Clin Exp Dermatol 1991; 16: 188-92.
12 Rushton H. Comparative efficacy of oral finasteride and
topical 2% minoxidil in male pattern baldness. In: Van
Neste D, Blume-Peytavi U, Messenger A,
Grimalt R, eds. Hair Science and Technology. Tournai-Belgium:
Skinterface sprl, 2003: 317-29.
13 Van Neste DJ. Hair growth evaluation in clinical
dermatology. Dermatology 1993; 187: 233-4.
14 Milner Y, Sudnik J, Filippi M,
Kizoulis M, Kashgarian M, Stenn K. Exogen, shedding
phase of the hair growth cycle: characterization of a mouse model.
J Invest Dermatol 2002; 119: 639-44.
15 Van Neste D, Leroy T, Conil S. Exogen hair
characterization in human scalp. Skin Res Technol 2007; 13:
436-43.
16 Schmitt Daniel, Noly Valerie. L’homme et sa peau.
Lyon (France): ARPPAM - Edition, Museum de Lyon, 1996; 1-111.
17 Leroy T, Van Neste D. Contrast enhanced
phototrichogram pinpoints scalp hair changes in androgen sensitive
areas of male androgenetic alopecia. Skin Research and Technology
2002; 8: 106-11.
|
Printable version
Version PDF