All content credit due to our forum member hellouser for putting all the painstaking time and effort into attending the conference, and gathering all the audio and video content presented here.
Dr. Annika Vogt’s presentation is below (and above). This presentation covers the following topics:
An analysis of the chemical, structural, and functional differences between scalp that has already endured hair loss, versus unaffected scalp. Also the establishment of new processes for doing this type of research was covered.
The need to understand the causes of the “disease process” in hair loss is of primary importance. Current methods for examining various aspects of the hair follicle functioning and how it is affected by Androgenetic and other Alopecia’s are inefficient. Likewise, being able to test whether a treatment works efficiently is difficult due to the need for multiple facets of the testing process, including hair counts, biopsies, and other inconvenient, manual steps.
Novel methods were discussed for examining the effectiveness of topical treatments for hair loss, including Optical Coherence Tomography, experimental methods, readouts, and molecular processes. An initial trial of 6 and then 12 individuals was conducted to run these analysis methods on the effectiveness of Minoxidil. The approach was to use RNA markers which they hoped to find consistent across multiple patients, which could be monitored as a new standard in future clinical trials. The ability to collect hair samples from patients in a non-invasive way that could be used in future trials was also examined.
We encourage you to follow along with the video, and discuss this presentation with others in our forums:
Dr. Annika Vogt – Differences Between Affected and Clinically Non-Affected Scalp
Speaker: We’ll change a little bit from the order that was presented in the agenda, and we’ll have our next speaker, and then we’ll take some questions. Then we’ll have our remaining speakers who are going to take some more questions. So if we can have the next presentation. Our next speaker is Dr. Annika Vogt. She’s a senior physician at the Hair Competence Center and Pediatric Dermatology unit, as well as the Scientific Director of the Experimental & Translational Research on the Department of Dermatology at Charité-University of Medicine in Germany. She is also a faculty member in Paris at the University of Pierre et Marie Curie. So Dr. Vogt.
Dr. Vogt: We are very pleased to see that our session attracted so many participants. After this very impressive and exciting talk, we are now shifting more towards clinical focus and trying to understand a little bit of the disease process. And I think we already illustrated very nicely that it’s complex to understand this organ, and see how we can play with the organ to really reach a level where we don’t focus on only one facet, and are really able to regenerate, in this case, hair follicles or, in our case, to really understand the disease and also the drug action.
I will now talk a little bit about protocols we developed in our lab to improve the way we conduct our clinical trials and also speed up our readouts. And I have to say that a lot of the work I’m presenting now is part of an investigative initiated research project, where we received funding from Johnson & Johnson.
But the starting point, I mean, those who have been following my work over the past years know that I’m mostly interested in targeted drug delivery and packaging target active molecules to create in other ways photonic dermatotherapy. So we found ourselves in a position where we felt that it’s really hard to have fast screenings of novel compounds and also novel formulations. And also like when it comes to studying the molecular processes and identifying novel targets, we can only always look from the very specific questions.
What we want to see is the whole disease process at the follicle, but there we really rely on the patient biopsies. So this limits the way or the things we can do with our patients because we are not free to take biopsies in as many numbers as we want, and it’s not that cosmetically accepted. Like we have scarring alopecia, for example. People are not very happy to have [inaudible] scars after a clinical trial. So we do a lot of work on organ cultures. And with the cell culture, we can prove certain signaling pathways and certain questions. But it would be really nice if we could add molecular markers to our current understanding of what we see in the clinical trials.
Because the typical situation is that what we have – and over here will focus on androgenetic alopecia – as a typical situation for a drug efficacy trials. You will see here a male patient. And what we usually want to look at is those who participated in our Ulrike’s session this morning, we are now able to do mini-zone trials. We have certain areas we want to focus on, but then we still have the situation that we have to wait for many weeks to sit there and have global expert panels and negotiate numbers of hair follicles and micrometers of thickness. And so we felt that we wanted to look at different ways of improving the way we do these trials.
So we designed a set-up where we wanted to get ready to have efficient androgenetic or testing of topical treatments for androgenetic alopecia. And we felt that we wanted to look at frontal, vertex, and occiput using the clinical methods: adding to our panel things like optical coherence tomography (I will show in a minute) and adding experimental methods and readouts that will tell us a little bit about the scalp/skin milieu, which is really hard to mimic in vitro, and also molecular processes. And I will now guide you a little bit through those complex studies set-up that we created and tested in six individuals, until we then repeated the whole procedure in 12 individuals and monitored the effectiveness of minoxidil treatments, as Ulrike presented this morning.
So we were trying to generate protein material that helps us to look at scalp milieu, and we decided to look at RNA material from plucked hair follicles. Because if you pluck the hair follicles and there’s [inaudible] – and we also had discussions – of course, you don’t remove these entire compartments, but it would be so nice to be able to pluck hair follicles, get a molecular readout, and then, over time, go back to the same area or, at the same time, look at different sites to get a more comprehensive view of what’s going on with this patient. So the idea was to look if we can generate robust RNA microarray data that allows us to set up a panel of selected molecular markers that we could then use in multiple upcoming trials
So what we did here was, we said, if you want to – and we weren’t really sure how the inter- and intraindividual variability would be if we really decided to do microanalysis. So we picked six individuals, and said they had to have AGA, Norwood-Hamilton IIIv-IV. So very narrow, well-picked study population.
And our goal was to test the different approaches of noninvasive sample collections, mostly to integrate them in a nicely package for a clinical trial, which is doable on a patient that acceptable. But we also used what we found to screen a little bit towards more molecular markers.
So one challenge to start with was to integrate the different methods we were using into our study protocol, which allowed handling on three investigational sites, in a way that you don’t interfere with the different approaches we want to do. So we had the regular clinical readouts: the phototrichogram, where you shave this 1.8 diameter circle. And then we integrated optical imaging – optical coherent tomography – and also ultrasound. And a new method or/and the hair plucking next to it, as well as the method I will present, which allows us to collect protein from the skin surface to get to look at the different portions. Also because the hair plucking, of course, only yields a limited material. And we were thinking that if we – end up with array data – might want to have some protein material, at least, to validate or get some idea.
So the first thing was that we had the regular clinical results with the phototrichogram from these three affected sites, which I will show on the table on the next slide, confirmed that we had exactly that IIIv stage we were looking for. We found that the 20 MHz sonography, which we were doing to see if we could learn something from angles or extension of the hair follicle, was not very well reproducible. We did find similar to studies we had done before on hair regrowth that for some many patients, the OTC measurements of regrowing hair shafts right at the skin surface level might be helpful. Here, we had very significant changes in the diameter, so that could be a very helpful addition to the regular phototrichogram Trichoscan readouts.
And then because we frequently talk about microinflammation or involvement of subclinical seborrheic dermatitis sebum compositions, we wanted to add the earlier readouts. And we’ve seen in a lot of other talks that people were looking at interleukin-A, for example, as the classic indicator. But usually, we use Sebutapes. I’m not sure if those who have worked with those tapes. These are adhesive tapes and they require quite extensive shaving.
So what we did was something – people who followed our work know that we like to play with cyanoacrylate-based resin. Basically superglue, and we have a lot of experience on other body sites because we use it for targeted delivery of molecules into hair follicle for penetration enhancement. And here, we used the superglue method, which allows us to remove 30% of the stratum corneum. It’s well-tolerated and only gives a short array, basically interval, but doesn’t really destroy the skin integrity.
And we use this very special feature of the cyanoacrylate to flow specifically into hair follicles. The past use, we had used that in a very targeted drug [inaudible] approach to get casts from hair follicle openings. And together with all our partners, we were able to collect material drugs that we had topically applied from the hair follicle opening.
What we did in this study, we combined these approaches. This was the study where we recovered material. But now in this new study, we did it on the shaved circular area that we had use personally for the phototrichograms, so we shaved that down to the skin surface level and did one of these strippings on an area of 1.5 cm diameter. And doing this allowed us to remove the upper stratum corneum layers, but more representative than just skin surface material – oops, sorry – plus the follicular casts. So we felt that we’d get a pretty – a more follicular-specific readout of what we find here in the opening of the infundibulum parts.
We were – it is not easy to allude it. So we did a lot of protein quantification testings because usually we have colorimetric assays that interfere with the glue components. But here we were able to allude proteins for analysis and normalize them to total protein content.
I had shown you the quantification for the interleukin-1α and receptor antagonist, which is standard, but then we felt it could be also suitable to learn a little bit more about possible other inflammatory molecules. Because for the disease process, other things might be interesting, too.
So we did a wide range of chemokines because the hair follicular immunology is of part of my interest, and we found them present, although we did not find major changes among the different investigational sites. Interestingly however, if you compared the ones that we detected with the paper of Margo et al, who described the differential expression of chemokines all along the hair follicle. A lot of what they reported was in the infundibulum was found in here, except that in the AGA patients, we did not see substantial differences.
In contrast, and here in this array, I found it pretty reassuring or interesting that in these markers, we did not have large variability. In the expression of different proteases, we had substantial differences. And we are currently looking in the meaning. And obviously, there is a lot of ways for interference and for different sources of these proteases, but we found it very striking that while on that chemokine profile levels, we had to really have significant changes. Here, we had a very diverse picture. I’ll mention that again when we come to the RNA.
So what did we do with the RNA? Here, we used the plucking. And those who regular do the plucking of hair follicles for the phototrichogram know that it’s a rather reproducible way of gathering and collecting hair follicles. We decided to look at the un-staged hair follicates. We did not go specifically for anagen because we felt that the AGA process is really reflected by the mix, and that we would miss things if we would just go for one hair follicle type.
So we extracted RNA from 13 hair follicles, and together with [inaudible] did whole genome oligomeric microarrays. We found a large – of course, you’re always going to get this huge bulk of data, but we found over 700 differentially expressed genes. And the vast majority was expressed both in frontal and in the vertex, which is very much in line also with Dr. [inaudible], for example, had published on biopsies that they’ve taken. And only a smaller proportion of genes was differentially expressed among frontal and vertex.
When we did the array analysis, we first of all were very pleased to see that we did grouping. Of course, we have listed the areas of molecules of interest. So we were very pleased to see that even though we had the plucked hair follicles and not the whole biopsies, which an advantage is, of course, highly specific for the hair follicle material, we found a large number of hair differentiation and cycling genes and keratins as well as keratin associated proteins in line which what we have seen with the clinical data, which is in line what we think what’s happening. Of course, we have less active hair growth. So this was reassuring in a way that we could see what we think.
We were also interested in seeing a lot of ion channel solute carrier molecules, which became even more interesting than minoxidil study, which Ulrike mentioned, where we indeed found differential regulation also on the minoxidil treatments.
And then we felt that, I mean, just knowing that you have more hair growth is interesting, but really trying to find new areas for activity – that is really the challenging part, we felt, and it took a lot of manual thinking and sorting of genes. We were surprised to find a lot of genes associated with polyubiquitination as well as methylation. So now, we are gradually working ourselves a little bit through these areas of interest, and we always have the problem when you do these micro-analysis, especially in this one in ways that we’ve obtained material, that, of course, you cannot go to the lab and provide extensive validation studies on the same material. So we try to look for things we could do to further substantiate findings.
Picking the RNA methylation. We for example, did one – did a check for the methylation status of the RNA among the different investigational sites and found higher methylation frontal/vertex compared to occiput. That’s the N6-Methyladenosine, and it’s interesting because the methylation that has been associated with circadian clock regulation. And people have shown in other contexts that you can shorten or extend circadian both clock regulation by blocking this N6 methylation. So it could be very interesting the more we hear about such clock regulations to further look at such areas. This is now the way we tried to work a little bit through these stages.
The other thing is that we were trying to find matching movements among the genes or the expression of the protein findings. Consistent with the idea that we did not have substantial differences in the chemokines on the protein level, we found very few purely typically inflammation molecules differentially expressed, which doesn’t mean that we didn’t have genes in innate immunity and antimicrobial peptides, but it’s now that we have huge interleukin differences.
In contrast, we did test significantly differentially expressed proteases, including serine proteases, metalloproteinases, and also [inaudible] types. Even though not all the RNA equivalents jumped the level for significance, when we looked at the full changes – this is not on these slides – we found a similar – like we never had the situation where a protein was very high but the RNA equivalent would not match at all. So we found a trend towards matching expressions.
What we did here, of course, is a very complex assembly of different puzzle pieces, and this was establishment of a lot of standard operating procedures. But overall, what I learned during this whole exercise was that we have several ways now of collecting these things in a reproducible way. We can extract, quantify, and screen. And a lot of these things, we will then use them for the minoxidil trial, where we will still do final analysis, but it worked really nicely to really monitor in short times: 0, 4, and 8 weeks (Ulrike presented it this morning).
And although they are the clinical data, we know these patients, for example, have shedding. But even though the phototrichograms only indicated this trend. We didn’t have significant clinical readouts during this time, but we do have very significant RNA levels. And if we were able to set up a limited number of molecules of interest to have something like a miniature readout for certain drugs that includes the targets of interest, we think that we could tremendously increase and speed up the way we look at efficacies of topical treatments.
And obviously, there’s always advantages and disadvantages to models and especially the hair follicle with its complex, different aspects. Hair follicle isn’t [inaudible] but always need all these models for understanding. But what I think what’s been missing these days, and this is partly to the restriction of human material, is that we do not have much initiative that come directly from the patients and has been brought to the labs.
A lot of the drugs which are currently, the drugs which we mostly use, are side effects where other disciplines reported side effects to us, and then we went to the lab and tried to understand why a drug which was not designed for that purpose increased hair growth, for example. Now, I think it’s the time to look more specifically in the disease pathology and then try to look for these very little new niches where we can increase our understanding.
All this work was done in our center at Charité. Most of you probably know when you look at the [inaudible]. We try to combine. We have our clinical trial in the experimental lab. I understand that a lot of this presentation was very complex, we really have to work with this stepwise, but I hope that you got that glimpse of the same enthusiasm that we have with these kind of projects. Thank you.
Please end this video at 22:05 and save as its own unique video. Please begin a new video for the content below this point on this document. If possible, please change the text on the video itself from Dr. Annka Vogt to Question and Answer. Beneath that, please modify the text from “Differences between…” to “Audience questions for Dr. Tsuji and Dr. Vogt” Then of course the same date. Don’t need to change that.
Kevin Rands is the Founder of HairLossTalk.com and President of Online Health Networks, Inc. a Miami based corporation providing consumer health education on the web. He is also the Founder and Principal Writer for DisrupterDaily.com, an online publication on disruption of health and tech sectors.