Auxilium Pharmaceuticals AUXL
July 11, 2012 - 3:24pm EST by
2012 2013
Price: 25.50 EPS -$0.09 $0.60
Shares Out. (in M): 48 P/E NM 41.9x
Market Cap (in $M): 1,232 P/FCF 136.0x 26.0x
Net Debt (in $M): 0 EBIT -5 35
TEV (in $M): 1,069 TEV/EBIT NM 31.0x

Sign up for free guest access to view investment idea with a 45 days delay.

  • Biotech
  • Operating Leverage
  • FDA approval



I have a lot of graphs and tables in my write up, which I was not able to paste correctly below.  

Here is a link to the write up with all the tables and graphs.

Here is a link to a slide deck for a quick summary of the thesis.



AUXL is a biotech company on the brink of profitability due to the substantial operating leverage in their business model.  AUXL’s orthopedic and urology sales forces are currently subscale with only one product to sell in each.  However, every incremental product/indication they are able to sell through these sales forces is hugely accretive due to them already having the infrastructure in place.  They are in a pivotal time in their development - having just read out positive Phase III (the final stage of testing before a drug gets approved by the FDA) data in Peyronie’s Disease that they can sell with their urology sales force and that will drive profitability.  AUXL also has two more indications they are studying for Xiaflex that can be marketed with their existing sales force, and offer substantial upside if successfully developed (+100-200%).


Capital Structure




Overview of the company


AUXL is a biotechnology company with two products on the market.  Testim ($205M in 2011 sales) - a testosterone gel, which is the number two product in the $1.2B and rapidly growing testosterone replacement market.  The other product is Xiaflex ($44M in 2011 sales), a collagenase (an enzyme which breaks down collagen) approved to treat Dupuytren's Contracture (DC).  DC is a disease where a collagen cord grows in the hand, and prevents one from straightening the finger.  For more information on Testim and DC please see the appendix. 


Xiaflex is being studied in a number of other indications.  AUXL in early June reported positive P3 data for Xiaflex in Peyronie's Disease (PD).  PD affects men, and causes curvature of the penis which can affect the ability to have intercourse. 


AUXL has partnered Xiaflex for DC and PD with PFE in Europe and Actelion in Canada, Brazil, Mexico and Mexico. 


Xiaflex has earlier stage studies ongoing for cellulite and frozen shoulder.  Xiaflex has already shown that it works in these indications.  Cellulite data reports out in 4Q 2012, and frozen shoulder in 1H 2013.  As investors turn their attention to these indications they will recognize that there is a high probability of success in these indications and the enormous market potentials. 


Key catalysts driving upside

  • Estimates and valuations will rise as the investors recognize the high probability of approval in Peyronie's Disease, coupled with the large commercial potential and substantial sales force leverage in the next 12-18 months.
  • Data on cellulite and frozen shoulder will read out over the next 12 months.  Despite having solid proof of concept data and offering 100%+ upside to the stock if successfully developed, they are not in street models. 


The stock is worth $37 as a standalone with just the existing indications and very conservative Peyronie's Disease sales.  If cellulite or frozen shoulder is successfully developed they would be worth an additional $59/share and $35/share respectively. 


Peyronie's Disease (PD)


PD is the development of collagen plaque, or scar tissue, on the shaft of the penis that may harden and reduce flexibility, thus occasionally causing pain and causing the penis to deform in a bend or arc during erection.  In addition to difficulty with sexual intercourse, PD is commonly associated with emotional distress, loss of self-esteem, disease bother and depression. PD is a heterogeneous disease with an initial inflammatory component.  This inflammatory phase is poorly understood with a somewhat variable disease course and occasional spontaneous resolutions of not greater than 13%.   After 12-18 months of disease, the disease is reported to often develop into a more chronic, stable phase.  The estimated prevalence in adult men of PD has been reported to be approximately 5%; however the disease is thought to be underdiagnosed and undertreated.  Based on U.S. historical medical claims data, it is estimated that between 65,000 and 120,000 PD patients are diagnosed every year, but only 5,000 to 6,500 PD patients are treated with injectables or surgery annually.


AUXL ran two P3 clinical studies for Xiaflex in PD called IMPRESS I and IMPRESS II.  These studies had co-primary endpoints, change in penis curvature, and the Peyronie’s Disease Questionnaire Bother Domain (PDQ).   The PDQ asks patients to rate how much their Peyronie’s Disease bothers them.  To fill out the PDQ, you had to have had sex in the past three months.  Many men with PD are unable to have sex at all, thus not everyone in the study was able to fill out the PDQ.  For more details on the PDQ bother domain please see the appendix.


In order for the trials to be successful both the curvature and PDQ had to be statistically significant (have a p-value < .05).  Below are the results of the trials. 




Approval Risks


The street is focused on three main approval risks. 

  1. That the methodology of the mITT analysis which is the population used to calculate the primary endpoints is not appropriate. 
  2. That the p-value of the PDQ is not statistically robust. 
  3. The clinical significance of the change in curvature or PDQ


Below I address these approval risks in detail.


Methodology of mITT analysis

The ITT (intent to treat, ie the all patients who received drug) population in the P3 studies was 832 patients.  However the mITT (modified intent to treat, the group that was used to test the primary endpoint) population was 612 patients.  So the mITT population was only 612/832 = 73.5% of the ITT population, causing confusion among investors over the difference.




The mITT analysis was the pre-specified statistical analysis.  It mandated that patients have both a baseline PDQ and curvature measurement, as well as a post-baseline measurement in order to be included in the mITT population (which was used to calculate the primary endpoint analysis). 


A substantial number of patients did not have a baseline PDQ score.  Why did they not have a baseline PDQ score?  Many patients were not able to fill out the PDQ because they had not had sex in the past three months and that was a requirement to fill out the PDQ.  In fact only 677 patients had a baseline PDQ assessment (from conversations with the CMO).  Since the mITT population is 612 patients, that means that only 65 patients dropped out or were not able to have a post-baseline PDQ (or curvature) assessment. 


So 65/677 or 9.6% of the starting population with a baseline PDQ dropped out over the course of the study.  The overall completion rate for the study was 86-90%, so this discontinuation rate for patients with a PDQ at baseline is not alarming at all.  It is very much in line with the overall completion rate for the study, and so alleviates concerns that the patients who were evaluable on the PDQ or mITT analysis were not representative of the overall population.  Furthermore, the overall completion rates of ~90% in a year long study is very impressive in a non-life threatening condition.  This speaks to the high quality conduct of the study. 



The other reason to be optimistic about the methodology of the mITT (excluding patients who did not have a baseline and post-baseline measurement) is that the exact same methodology was used in the primary endpoints of the patient report outcome of the Viagra studies for approval in Erectile Dysfunction.  This is an appropriate comparison for Xiaflex in Peyronie's disease because it is the same Division of the FDA, and was also in an indication with no previously approved drugs, and also used a questionnaire that asked patients about their disease (similar to the PDQ here).  The caveat is that the Viagra results were highly statistically significant, so the robustness was never called into question - but the important thing is the methodology of their study endpoint used an mITT analysis which only evaluated patients with a baseline and post-baseline assessment.  This confirms that the mITT analysis used by AUXL is statistically and scientifically appropriate.  The CMO of AUXL was not aware that Viagra had used a similar mITT population for their analysis, thus it is highly likely that other investors do not know of this either. 


Are the PDQ results statistically robust?

We know that the methodology of the mITT analysis is sound, and has precedent with the Urology Division of the FDA.  However, how statistically robust are the results?  The p-value in IMPRESS-1 was .0451 and the p-value in IMPRESS-2 was .0496.  Investors are worried that one or two patients could change the p-value enough to make it >.05. 


The FDA standard is two trials with a p-value of less than .05.  The standard is not two trials with a p-value of less than .01. 


Let us look at Provenge as an example.  Dendreon conducted one well conducted P3 study with OS as the primary endpoint.  It was with 512 patients and the p-value was .032.  All the other supportive evidence was not pre-specified for OS.  What does the approval document say for the Provenge approval (I know it was Biologics approval, and that division is not the most rigorous).  This is what they said: 


Regulatory standard for evidence of effectiveness:

As stated in the FDA guidance Providing Clinical Evidence of Effectiveness for Human Drug and Biological Products, the usual standard for marketing approval is two or more adequate and well-controlled studies. In the sipuleucel-T BLA, the recommendation for approval is based on only one adequate and well-controlled investigation, D9902B. As stated in the guidance on effectiveness, FDA will generally rely on a single trial only when a second trial is not ethical and/or feasible. Because D9902B provides substantial evidence of improved survival, a second study would be neither ethical nor feasible in the United States.


In addition, as stated in the guidance on providing evidence of effectiveness, the assessment of the adequacy of a single trial will consider the characteristics of that single trial and the availability of supportive evidence of efficacy. Characteristics of Study D9902B that support its use as the only primary evidence of effectiveness include that the study was relatively large (N = 512), multicenter, with results that were consistent in multiple sensitivity analyses and across numerous subgroup analyses. In the sipuleucel-T BLA, supportive evidence of efficacy comes from two Phase 3 studies, D9901 and D9902A.


In summary, D9902B, supported by the results of D9901 and D9902A, meets the regulatory standard for a single trial that provides the substantial evidence of effectiveness necessary to support a marketing approval.


Despite only have a p-value of .032 in one study, the FDA did not question the robustness of the result.  Before Provenge got approval investors were worried if it was a sufficiently low p-value to allow for approval.  I think that having two pre-specified P3 trials with statistically significant results (even if only marginally significant) provides substantial evidence of the effectiveness of this drug on the PDQ.  Furthermore, we also saw evidence of the efficacy in the P2 results, though it was not pre-specified.  This gives us additional confidence in the "truth" of the results we see in P3.  This is not one P3 study with a p-value of .05 (similar to what happened with ISPH) where there are real doubts on the reliability of this one estimate. 


Are the results clinically meaningful?


I have included the efficacy results below again.






Though we are looking at a ~8 degree net improvement on curvature relative to placebo, I think that is fairly reasonable given that these patients have ~50 degree curvature at baseline.  So we are talking about a relative 16% improvement in curvature on a placebo adjusted basis.  Not great, but I would expect that the secondary endpoints on efficacy that looks at different cuts of degree improvement will give further evidence to support the clinical meaningfulness of the efficacy.      This would be a similar analysis to how ACOR and UTHR presented their data to buttress the argument that their drug was clinically significant.  For instance look at the UTHR TRIUMPH study of Tyvaso.  The baseline 6MW was 348m. 




However, the 6MW median change was only 20m at week 12.  20/348 = 5.75% treatment effect at the median.  




However, this was approved by the FDA without an advisory committee, and when you talked to doctors they all pointed to this slide showing the percentage of patients that had different levels of 6MW improvement. 




I believe that presenting similar data for Xiaflex in Peyronie's will go a long way in convincing doctors, investors and the FDA of the clinical relevance of the data. 


On the PDQ we see a similar 12-16% relative improvement from baseline PDQ score on a placebo adjusted basis.  Seeing these sorts of rates of response over a placebo effect reminds me somewhat of pain trials with high placebo response rates.  These seem like they would be sufficient.  The absolute effect size seems reasonable given the baseline PDQ scores.  People question what is the significance of one point on the PDQ.  That is hard to tell, but I think people will say that a 3 point improvement is "clinically meaningful" and so just cut the data to see the % of patients on drug and placebo that a 3 point improvement on PDQ to show the clinical meaningfulness of the data. 


Market potential

AUXL has said the initial opportunity is only ~5,000 to 6,000 patients.  They are essentially limiting themselves to the surgical and verapamil injection population.  (In the chart below, it goes from most invasive treatment on the left to least invasive on the right.) 




However if you look at their actual entry criteria - they are not limited in such a way.  The only real entry criteria was not having previous surgery and having a curve of between 30 to 90 degrees.  Patients were excluded if they had verapamil injection in the past 3 months, but this argument that it will only be used in the surgical and verapamil injection population is an artificial one that is not supported by the clinical trial. 


So why would AUXL limit their treatment population to 5,000 patients?  In short the answer is because at $3,250 per injection and with 90% of patients getting at least 6 injections, a 3250*6 = $19,500 treatment cost per patient would be a very tough sell to payers if it was for 50,000 patients.  But for 5,000 patients, essentially an orphan or ultra-orphan indication, payers will be much less likely to object. 


The FDA label will of course not limit the population to patients who were supposed to get surgery or verapamil injections as that is not how the trial was done.  But from a payer perspective you want to compare it to surgery which costs ~$15,000 per treatment. 


And the treatment population that uses Xiaflex will be substantially larger than the surgery population.  Dr. Hellstrom, Professor of Urology and Chief of Andrology at Tulane University said:

"A lot of the patient in the [Xiaflex] study only have oral treatments.  Most of them haven't been treated.  They are naïve to treatment." 


This supports the idea that the patient population treated once the drug is commercially available will be much wider than the 5,000 patients the company is currently telling analysts. 


It makes sense to avoid surgery even if patients have such bad PD that they are unable to have sex.  This is because patients can lose up to an inch or two of length with surgery, in addition to numerous complications like erectile dysfunction.  With this type of result from surgery, it is no wonder that very few patients choose to get surgery every year. 


Dr. Hellstrom commented that these patients are incredibly motivated to do something about the disease and they spend more time learning and worried about Peyronie's on average than men do with other sexual dysfunctions such as erectile dysfunction.  As a sign of how active this group he remarked that the website, just one of the patient websites, gets 500,000 hits per month. 


Dr. Levine, Professor of Urology at Rush does 200 surgeries a year, but would use this before surgery in his patients because "there is no risk in using Xiaflex first."    And he was willing to use it in a lot of patients that he doesn't treat now with verapamil as well since it would be FDA approved and have strong clinical evidence - as opposed to all the other treatments they use now.


Dr. Jed Kaminetsky, Clinical Assistant Professor of Urology at NYU, remarked that it would be first line therapy because it would be the only treatment approved.  He is confident that a lot of Peyronie's disease patients coming out of the woodwork.  He personally would use this before surgery because the injections are not that difficult to do.  However it will probably still be treated mostly by the specialists as opposed to community urologists. 


A Leerink survey of 32 urologists who treated 2,000 PD patients conducted in June of this year (after the release of the top-line data) found high willingness to use Xiaflex.   More than a quarter of doctors found the data to be very compelling.  And 69% were willing to try it and experience it for themselves.  Note that this impression is just based on the top line data, before AUXL has had a chance to present other analysis (at conferences or in journals) that doctors may find more compelling. 





Doctors were also more than willing to treat a wider group of patients than what AUXL has suggested would be the initial population. 






The survey indicated very high unmet medical need in Peyronie's and projects very rapid uptake. 




Sales force leverage

The existing Testim sales force is about 200 people (including sales reps and support).  AUXL has said historically they would need to hire very few if any additional people in order to add Xiaflex for Peyronie's to the bag.  This is consistent with what Dr. Kaminetsky said above about PD being mostly treated by specialists. 


Let us compare the sales force size relative to MDVN since they will also be marketing to urologists (in addition to oncologists).  MDVN does not disclose the exact number of sales reps, but they have 69 full time employees at March 31, 2012.  They are fielding 50% of the sales force for the MDV3100 launch.  They said on their 1Q call that they had finished hiring their entire sales management team and MSL and are in the process of hiring their sales reps.  In their 10-Q they said they had 69 SG&A headcount at March 31, 2012 vs. 32 at March 31, 2011 and they expected that number to increase to 164 by year end.  So MDVN SG&A will be about 130 people incremental for the MDV3100 launch, or 260 total between Astellas + MDVN to cover both oncologists and urologists.  Thus the  existing 200 person field force for Testim should have very good coverage for the Peyronie's indication.   


Typically, Peyronie's patients are funneled to specific urologists, especially those that deal with sexual medicine.  Thus the target universe of Peyronie's is likely to be a subset of the overall number of urologists.  This is likely due to a number of factors, including lack of approved treatments, as well as specialization necessary for the surgery.  This means that the current market for treatments now may understate the potential for Xiaflex.  Historically in many markets as easier to use and effective treatments have become available the number of treating doctors has increased and further increased the number of patients treated.  The introduction of oral Tracleer for Pulmonary Arterial Hypertension (PAH) is a good example of this.  At the time of its introduction as the first oral medication for PAH when previously only IV prostacyclins were available (which require overnight hospitalization to begin treatment) led to community doctors being willing to treat and greatly increased the overall PAH market.  Another example of this phenomenon is Zytiga for post-chemo prostate cancer. 




Despite JNJ being new the prostate cancer market with Zytiga, and Jevtana being marketed by Sanofi which had previous experience with docetaxel in the market, it is clear that Zytiga has done far better than Jevtana even though both are approved for the post-chemo therapy indication.  There is a difference in the dosing and the duration of treatment, but the immediate uptake in Zytiga shows that there are a lot more prescriptions being written, and probably more doctors prescribing as well. 




Sell side estimates are very conservative

Street models are very conservative on revenue expectations for Peyronie's disease offering easy comps for the company to outperform expectations. 


Though still early here are current Peyronie's estimates




These estimates are way too low!  They are actually not that different than the Dupuytren's launch for year 1 and 2. 


Below I compare the actual DC revenues vs. the projected PD revenues.  The numbers are different than the mean I show in the consensus table above, because I adjusted estimates that assume a launch in 2014 and used that as year 1. 




The sell side is overly pessimistic on the PD commercial potential. 


Here is the market opportunity as laid out by AUXL:





They believe that there are only 5,000 to 6,500 patients in the surgical and injection category being treated now.  I actually think it is higher than that.  Here is the data from the BESS database (Medicare database of procedures).  In just the Medicare population there are ~4,300 procedures done in 2010.  With the average age of onset for Peyronie's disease being 53 years old, the Medicare database should capture a relatively small number of the actual treated population. 




Here is Leerink's model:



Their peak penetration is 8% of incidence patients, which is less than the injection and surgery populations.  So they give no credit for usage in patients who are not willing to accept surgery right now. 


The Leerink model only includes the newly diagnosed incidence each year.  It does not include prevalence for those patients already treated or who had previously sought treatment.  In fact a 2008 paper on a survey of Peyronie's treatments by members of the practicing members of the American Urologic Association, found that 32% favored observation as the initial management for Peyronie's, and 24% of doctors favored it as the second-line treatment after failure of first line.  Remember that surgery comes with loss of length and a risk of erectile dysfunction, so even though patients are highly motivated, they may still not be willing to risk surgery.  This is not a disease that goes away with time.  The fact that many doctors simply observe patients is related to the lack of effective treatments.  These patients are highly unlikely to be "cured" unless they have surgery.  There is a fairly large pool of prevalent patients out there who will be quick to demand Xiaflex therapy – as it is an effective therapy without many of the side effects of surgery. 


Supporting the unmet medical need here and the pent-up demand for the therapy is the fact that the trials were overenrolled by more than 33% and finished enrollment in ahead of timelines.  This despite the fact that enrollment began in October, and thus was ramping up during the typically slow holiday season at year end.  Another factor is the consideration that the completion rate of the study was 89.7% for the two P3 studies – amazing when you consider that this is an invasive trial with 52 weeks of follow up.


Comparison with Dupuytren's Contracture


Below is a comparison of Dupuytren’s Contracture with Peyronie’s Disease.




Investors are worried that the Peyronie's launch will be slow like the Dupuytren's Contracture launch.  However patients are much more concerned about Peyronie's disease than they are about Dupuytren's contracture.  The doctors have nothing else to offer these PD patients.  And surgery here is used as a last resort because of the risk of erectile dysfunction as well as the reduction in length.  With an estimated prevalence of 5% of adult men, at a much higher price point, the consensus launch expectations should be easily beatable.


Here is the US Peyronie’s Model:





Cellulite, also known medically as edematous fibrosclerotic panniculopathy, describes a pathologic inflammatory condition, in which lobules of subcutaneous adipose tissue extend into the dermal layer.  These changes can visibly affect the shape of the epidermis and resemble an orange peel-like dimpling of the skin.


In the normal subcutaneous fat layer directly under the skin, there are both perpendicular columnar and net-like fibrous connective tissue called septae. These fibrous septae, made of types I and III collagen, connect the epidermis to the dermis and create a network of compartmentalized adipose deposits. Women tend to have a higher proportion of columnar septae that are perpendicular to the epidermis, while men tend to have more of the net-like system.  In cellulite, the subcutaneous fat cells swell and push upwards.   As a result, the skin between the septae is pushed up and the perpendicular septae act as an anchor to pull the epidermis downwards and form the classic cellulite dimple. The surrounding adipose tissue forms small bulges under the epidermis around the dimple that can give skin a "cottage cheese" texture.




Cellulite occurs mainly on the pelvic region, lower limbs, and abdomen and has been reported to occur in 85-98% of post-pubertal females and rarely in men.  The condition is prevalent in women of all races.  Cellulite is different from generalized obesity. The fat cells found in generalized obesity are not limited to the pelvis, thighs, and abdomen. Further, the fat cells found in cellulite have different physiologic and biochemical property than fat tissue located elsewhere. There is no definitive medical explanation for the presentation and prevalence of cellulite and, despite multiple types of therapeutic approaches for the attempted treatment of cellulite, there are no approved treatments and little scientific evidence that any current treatments are beneficial.


Why Xiaflex should show efficacy in cellulite


There are three reasons why Xiaflex should work in cellulite:

  1. We already have data showing the efficacy of Xiaflex in cellulite 
  2. Cellulite is caused by collagen septae and we know that Xiaflex dissolves collagen. 
  3. Other treatments with similar mechanisms of action such as Cellulaze and subcision have also shown efficacy in cellulite.


There was an early pilot study done in 2006 in 10 patients that showed the efficacy and safety of Xiaflex in cellulite.  For more details of the study, please see the appendix. 


Conceptually Xiaflex should work in cellulite because the septae are made of collagen.  We also have a couple examples of cutting the septae by other means that have shown evidence of efficacy. 


There is a technique called subcision, where a needle is inserted under the skin to cut the septae.  Though most of the evidence of its efficacy in the medical literature is of lower quality, it does seem to have some effect on cellulite. 


The FDA has also recently approved Cynosure's Cellulaze for treatment earlier this year.  Please see the appendix for the press release.  Cellulaze uses a laser to cut through the septae to reduce cellulite.  It also claims to melt the fat and promote elasticity of the skin.  But fundamentally it is another form of subcision that uses a laser rather than a knife to cut the septae.  Just as importantly as the Cellulaze being another example where cutting the septae shows efficacy in the reduction of cellulite, is the fact that it was FDA approved.  Though not approved by CDER, it still shows that the FDA is conceptually comfortable with the idea of cutting the septae underneath the skin. 


So we have three forms of evidence that Xiaflex should work in cellulite - the actual Xiaflex 2006 study, and two indirect forms of evidence in needle subcision and Cynosure's Cellulaze which attack the same target - the septae. 


Why Xiaflex should be safe in cellulite

Collagenase has thus far shown safety in two indications in P3 studies - Dupuytren's contracture and Peyronie's disease.  There have not been immunogenicity issues, and short of injecting into a tendon there have been few SAEs.  With cellulite they are examining smaller doses than with the other diseases. 




The above doses are for a targeted area of 8cm x 10cm.  So women may want to treat a larger area than this to cover all their cellulite.  But overall the dose being examined is quite low.  Even if they end up moving forward with the highest dose studied, the .116mg dose, you could still treat 400 sq cm of area with just one vial's worth of dose, which is only two injections worth of doses for DC.  And AUXL has already done safety studies looking at multiple injections at once and there have been no safety concerns. 


Furthermore, in terms of Xiaflex getting into the circulation - there is a topical collagenase called Santyl which is used for wound healing for decades.  It is applied directly over the wound.  There was the original concern of Xiaflex getting into the circulation with Dupuytren's contracture, but since Santyl is applied directly over open wounds and therefore must get into the circulation, it is a good proxy for showing that Xiaflex should not have systemic side effect being injected into the skin. 


What is the market potential for Xiaflex in cellulite?


The market potential for Xiaflex in cellulite is very large as the vast majority of women older than 20 years have it. 


CynoSure's Cellulaze offers some insight into how large the demand is for an effective treatment for cellulite.  Leerink conducted a survey in January 2012 on 43 dermatologists and 40 plastic surgeons.  When asked on cellulite devices/lasers (which can cost ~$100-200,000 to purchase) this is what the doctors said:





Amazingly ~18% of doctors said they were interested in purchasing these laser devices for cellulite despite the high cost of the machines.  If that many are willing to spend $100,000 for a machine, then virtually all of them would be willing to treat cellulite with a capital light option such as Xiaflex. 


And how much are patients willing to pay for Cellulaze treatments after the doctor has spent $100,000 on the machine?   Let us look at the Cellulaze website for some answers:


What does the procedure feel like?

Because the area is numbed with local anesthesia, there is minimal discomfort during the procedure. You may feel a light pressure, but it is a mild sensation. Following a Cellulaze cellulite laser treatment, you may experience some aches, much like after a physical workout.

How long does the Cellulaze procedure take?

On average, this cellulite laser procedure will take one to two hours. This is dependent on the size of the areas to be treated.

Do I have to stay in the hospital?

No. Cellulaze can be performed under local anesthesia either in-office or in a surgery center.

How many treatments are necessary?

Results are achieved with just one Cellulaze cellulite laser treatment. Your results will continue to improve over the next three to twelve months.

How long do the results last?

Clinical studies have shown results can last for one year or more1. However, it is important to maintain a healthy diet and exercise regimen to continue to enjoy Cellulaze cellulite laser results.

Will there be scarring?

The incision sites will be very small. With proper care, any scars will be barely noticeable. Your physician can evaluate your skin type and provide the best information on scar prevention.

What's the recovery time?

The Cellulaze cellulite laser treatment is a minimally invasive procedure which usually requires only local anesthesia. Some light bruising can occur, but you'll be able to return to normal activities within a day or two. Your physician may ask you to wear a compression garment for a short period after treatment.

How much does Cellulaze cost?

The average cost is $5,000-$7,000 depending on the size of the area to be treated and the city or town where you choose to have your cellulite laser treatment.


Pasted from <>


Patients are willing to spend $5,000-7,000 out of pocket (this is not covered by insurance) for a treatment that lasts a year.  This is a procedure that can also leave a scar, and takes 1-2 hours to perform.  Xiaflex would be a simple outpatient procedure that can be done in minutes and will not leave a scar and will very likely cost a fraction of the $5,000-$7,000.  Patient demand would be overwhelming because the profile of the product is clearly superior.


Here is the US cellulite model. 




Botox is a decent but imperfect proxy for how one should think of this market.  Botox was approved in 2002 for cosmetic.  In its fourth year on the market Botox sold $357M in cosmetic indications.  However Botox only costs ~$400.  The price of Xiaflex in Cellulite will likely be 2.5 -3.0x that of Botox.  The model has Xiaflex in cellulite doing $717M in its fourth years on the market, so actually less procedures than Botox.  This is conservative because Botox’s target population of women with wrinkles is substantially smaller than the cellulite population. 


The incremental increase in sales and marketing for cellulite will be minimal.  Xiaflex is already sold to various orthopedic or plastic surgeons - the two major specialties that treat Dupuytren’s Contracture.  Plastic surgeons and dermatologists are the ones most likely to treat cellulite.  So AUXL is just need to expand coverage to the dermatologists.  AUXL’s current 90 person field force for Dupuytren's Contracture covers ~7,000 doctors.  The increase in sales force for cellulite would likely be to expand coverage to Dermatologists of which there are an estimated 8,000  ( is ~100.   At ~$250,000 average cost for a sales representative, that would be ~$25M in incremental S&M costs.  Thus cellulite would be highly accretive even at very low penetration numbers.  In fact every 10% increase in probability of success for cellulite equals ~$7 increase in NPV. 


What is the probability of success for cellulite and what are the risks for its successful development?

Dose ranging

The ongoing dose finding trial that reports out in 4Q 2011 will tell us if they can find a commercially feasible dose so that patients can afford to pay out of pocket for treatment.  The company chose the doses they are currently studying based on some animal models they had tested out (in pigs actually).  Conceptually a small percentage of the dose used for DC should be enough to treat cellulite.  DC is a cord that has developed over years that is so thick and strong that it pulls the finger and prevents it from extending – it is like rope.  The septae that cause cellulite are very thin fibers that simply pull down the skin a bit.  With less collagen to be dissolved, less Xiaflex should be needed.


Regulatory risk

It is unclear what endpoints the FDA will want to approve Xiaflex for cellulite.  They may have to develop some sort of imaging technique to accurately measure the decrease in dimpling.  On the plus side, the device side of the FDA has already approved Cellulaze for cellulite and Botox for wrinkles so it can be done.    


Commercial risk

This drug is not without its side effects.  There can be some bruising at the site of injection, but the lower the dose the less the bruising.  Another big question is how much Xiaflex can be injected at once for Cellulite.  Right now Xiaflex is only given once a month at .58mg for Dupuytren's Contracture.  If that limit remains in place for cellulite that could limit the attractiveness of this procedure for patients who need to receive large doses because they have large areas of cellulite.  However, this one dosing limitation was really a holdover from very cautious early development of the drug.  AUXL has data showing that patients can receive much more than .58mg of Xiaflex at once in DC.  We also have the Peyronie's P3 studies which had two injections within 24-72 hours of each other that also gives confidence in the ability to dose a lot at once.  Thus it should be safe to do more than .58mg doses at once, especially in cellulite where the dose is spread over much larger areas than Peyronie's or Dupuytren's and reduce the amount of bruising. 


Frozen Shoulder


Frozen Shoulder syndrome, also known as idiopathic adhesive capsulitis, is a common, prolonged, and painful disorder of diminished shoulder motion. The full duration of the frozen stage of disease can last from 1 to 3.5 years with a mean of 30 months. Increased capsular collagen thickening and subsequent capsular contraction in the glenohumeral joint is thought to cause restriction in both active and passive range of motion in the shoulder. As the condition progresses, pain can become more severe and is accompanied by stiffness and decreased range of motion. The stiffening may increase to the point where the natural arm swing that accompanies normal gait is lost. The most common treatments for Frozen Shoulder syndrome are long-term intensive physical therapy, corticosteroids, manipulation under anesthesia and/or arthroscopic release, but the long-term outcomes literature uses varying endpoints and each treatment has differing associated risks. Drugs are often used to manage associated pain, but none have been demonstrated to have an impact on Frozen Shoulder syndrome.




Frozen Shoulder syndrome is estimated to affect approximately two percent of the general adult population and has an increased incidence among patients with diabetes, Dupuytren's contracture, and thyroid disease. It tends to occur in a patient's fourth to sixth decade of life. The condition occurs slightly more often in women than in men, often presents bilaterally and commonly affects the contralateral side years after onset of symptoms in the first shoulder, but it does not typically affect the same shoulder twice.


Why should Xiaflex work in Frozen Shoulder (FS)? 

FS has been described as similar to DC. 

"The pathological appearances have been likened to those of Dupuytren's disease, with a fibrous contracture of the coracohumeral ligament and the surrounding tissues of the rotator interval, which is composed of a dense matrix of mature type-III collagen containing fibroblasts and myofibroblasts.  The compact arrangement of capsular collagen fibres causes the limitations of movement."

Robinson, Frozen Shoulder, J Bone Joint Surg Br, 2012


Conceptually, the shoulder's movement is limited by a buildup of collagen.  Therefore it makes sense that injecting collagenase to break down the collagen should improve the range of motion. 


If arthrographic distension works for frozen shoulder than so should Xiaflex

There is a procedure called arthrographic distention that is sometimes done for frozen shoulder.  The idea is that you inject saline into the shoulder socket so that it "breaks" the collagen and frees up the range of motion.  Here is the description from the Cochrane Review:


Another name for adhesive capsulitis is “frozen shoulder” or “painful stiff shoulder”. In fact, it is so painful and stiff, it becomes difficult to move your shoulder in a normal way. Sometimes the loss of movement to your shoulder makes it feel like it is completely frozen. It is thought to be caused by scar-like tissue (adhesions) forming in the shoulder joint. Arthrographic distension is a procedure where fluid is injected into the shoulder joint to break up the adhesions that might be restricting the shoulder’s movement and causing disability. Depending on the treatment, the fluid might contain a saline solution or steroids.

Buchbinder, Arthrographic distension for adhesive capsulitis (frozen shoulder) (Review),  Cochrane Review, 2009


If by injecting saline into the shoulder socket, so that it exerts so much pressure on the collagen that it breaks it and helps restore range of motion, then it seems something as simple as injecting Xiaflex to break up the collagen should work just as well if not better, and potentially be less risky. 



Xiaflex's Efficacy in Frozen Shoulder


In 2006 a small pilot study of Xiaflex in Frozen Shoulder was presented.


The results show some evidence of efficacy after the double blind first injection. 


Here are the results of additional injections:







These graphs definitely seem to show an improvement in degree of movement after injections. 


And here is a picture of patient before and after injections of Xiaflex. 





The side effects listed in the presentation seem not too concerning. 





If doctors are forcing saline into the socket the collagen pops, then injecting a little bit of Xiaflex shouldn't definitely be safer.  Injections into other parts of the body have shown it to be relatively well tolerated, and the topical Santyl (a collagenase ointment) also shows that Xiaflex is relatively safe even if it gets into circulation. 


Ongoing P2a Study

On December 1, 2011 AUXL announced that it had dosed the first patient in the P2a study of Xiaflex for Frozen Shoulder. 


The phase IIa study is an open-label, controlled dose-ranging study designed to assess the safety and efficacy of XIAFLEX for the treatment of Frozen Shoulder syndrome in comparison to an exercise-only control group. The study will involve approximately 50 adult men and women at approximately nine sites throughout the U.S. Topline study results are expected in the first half of 2013.


To qualify for the Frozen Shoulder syndrome study, patients must have unilateral idiopathic adhesive capsulitis of the shoulder with restricted range of motion in the affected shoulder for at least three months, but not more than 12 months. Following screening and determination of study eligibility, subjects will be assigned to four groups that vary in dosing (injection volume and concentration) or a fifth group receiving shoulder exercises only. Patients may receive up to three ultrasound-guided injections of varying doses of XIAFLEX (ranging from 0.29 mg to 0.58 mg in varying concentrations) separated by a minimum of 21 days and all patients will be instructed to perform home shoulder exercises. The study's primary endpoint is the change (degrees) from baseline to the day 92 follow-up in forward flexion (active) in the affected shoulder. Safety assessments, including immunogenicity testing, will be made during all study visits. More information on the study, including study sites participating in this trial, can be found at


The results of this study should allow them to go into a P3 study according to a call with the CMO.  The endpoint of this trial is at 92 days, so only 3 months.  This means they should be able to start a P3 trial in 2H 2013, and could potentially report out in 2014 and get approval in the 2015 timeframe.


Market opportunity


The market opportunity for FS is large.  With a 2% incidence of the general US population that means there are 6M new patients a year. 


The "standard of care" is steroids and/or physical therapy.  There is very little evidence to support either steroids or physical therapy.  It is just what is done in order to do something.  Here is an excerpt from a review on frozen shoulder in the American Family Physician (2011):


There is no high-level evidence to support or refute many of the commonly used treatments for adhesive capsulitis. Because the condition is often self-limited, observation and reassurance alone can be considered for patients in whom concomitant conditions have been excluded. However, patients should be counseled that it may take many months (or even years) to resolve. The painful and debilitating nature of adhesive capsulitis makes this option unacceptable for many patients.


Patients than then escalate to distension, the procedure described earlier in the Cochrane review.  And finally patients can move onto surgical options including manipulation under anesthesia or capsular release.  Here is what the American Family Physician  (2011) says:


Surgical options for adhesive capsulitis include joint manipulation under anesthesia and capsular release. Manipulation involves placing the patient under general anesthesia and manipulating the humerus to disrupt adhesions. There is moderate evidence that this alleviates pain and facilitates recovery of motion when it is followed by early physical therapy.Manipulation under anesthesia is associated with a risk of iatrogenic proximal humeral fracture, glenohumeral dislocation, and rotator cuff tearing. It generally should be avoided in patients with osteoporosis or significant osteopenia, with a history of glenohumeral instability, or who have previously undergone manipulation with subsequent recurrence.


Surgical release of the capsule has proved beneficial in patients with persistent or severe adhesive capsulitis. The benefit has been demonstrated with pain relief and functional recovery. Complications are minimal.


Clearly surgery is not the standard of care for Frozen Shoulder the way it is for Dupuytren's Contracture.   Thus Xiaflex will be viewed less as replacing surgery, but rather as something to be used earlier.


Recovery from surgery on the shoulder (as in FS) is more bothersome to patients surgery on the finger (such as in DC).  In a review on frozen shoulder in the Journal of the American Academy of Orthopaedic Surgeons (2011), this is what the authors suggest for the patients post surgically (either arthroscopically or through manipulation):


Following surgery, the patient is kept immobilized in 90° abduction and external rotation, with the head of the bed elevated 30° (Figure 4). Exercises are begun on postoperative day 1. For the next 2 days, the patient’s arm is kept in the abducted position. The patient is encouraged to maintain the arm in 90° of abduction and reach across the top of the head to touch the opposite ear as well as to internally and externally rotate the arm. Patients perform all actions (eg, walking, sitting, using the toilet, eating) with their hand on top of their head. Patients are discharged when they are comfortable, usually 2 days after surgery. Sleeping in abduction is continued for 2 weeks following surgery using a humeral cuff. Outpatient therapy is continued until full motion is restored and maintained. The intensity of this therapy is similar to that used preoperatively, and it should not be excessively painful or vigorous. A home stretching program is taught; stretching should be done three times per day. Patients undergo supervised therapy three times per week with no restrictions on motion. Strengthening is not emphasized.



If approved for Frozen Shoulder Xiaflex would be used after steroids or physical therapy and before distension or manipulation or surgery.  In fact this is the population that they are studying in their ongoing P2a trial. 


CPT codes stand for current procedural terminology, and are unique codes given for medical procedures.  Below are the CPT Codes for Frozen Shoulder:



MUA of Shoulder


Arthoscopic - Shoulder with lysis of adhesion/MUA

I could not find a dedicated CPT code for distension, which is also called hydrodilation.  This makes sense in some ways - it is probably viewed as an experimental treatment and thus does not have a CPT code.


If we go through the BESS Database (a database for procedures done through Medicare), we can get an estimate for the # of frozen shoulder surgeries done every year. 


All together there are about ~10,000 surgeries listed in the BESS database. 




The BESS database is limited to Medicare, for which most people are eligible at 65.  There are two considerations as we consider what the total number of surgeries for FS is:

  1. What % of patients with FS are >65 years of age
  2. What is the likelihood of getting surgery if you do have FS when you are < 65 years olds vs. > 65 years old


The presentation of FS is mostly in the 40-60 age range, and most people seem to get it in the 50s.  So likely less than <25% of patients are >65 when they get FS. 


I would also think that you are much more likely to get surgery to treat FS when you are younger than when you are older.  For two reasons, if you are younger you are likely to be more active and thus it bothers you more and you are inpatient about waiting.  If you are older, you are going to take longer to recover from surgery and it is going to be riskier. 


I would say that a reasonable multiplier for the BESS database number of FS surgeries to the total US population is in the 5-15x range.  So there are probably between 50k to 150k FS surgeries done each year in the US.


Frozen Shoulder Model

AUXL notes that FS is probably under diagnosed and under treated.  This makes sense given the diagnosis for FS is almost that of exclusion.  I suppose doctors may not treat FS since it can resolve with time.  So I put in discounts for diagnosis and treatments rates. 


I also found some evidence that only 5-10% of patients are treated with surgery.


So clearly surgery is not the standard of care here for FS, but is rather a last resort.  I think that bodes extremely well for Xiaflex uptake since we will not have the Xiaflex vs. surgery dynamic that we had in DC. 


I now have ~130k patients treated with Xiaflex in 2020.  I believe this seems like a reasonable number given the 50-150k patients that I estimate are treated with surgery every year. 





A key question determining how big Xiaflex is in this indication will be how soon is it used after steroid injections.  Do you wait 3 months and if there is no response from injections move to Xiaflex?  Do you use it straight from the onset once the FS is "frozen"?  Or do you just use it as a last resort before surgery? 


 It is important to note that frozen shoulder can resolve over time.  However, it can take many years to fully resolve.  Unlike Dupuytren's where patients are able to function without the full range of motion of one of their fingers, constraints in the movement of the shoulder is extremely bothersome to patients.  FS patients already undergo treatment with steroids and physical therapy now, so fitting Xiaflex in with or after these treatments will be convenient and natural, leading to a substantially quicker uptake than in DC.  Even though it is the same surgeon's using it, you would be launching with established billing codes and many surgeons already familiar with the drug from the DC experience.   


Another potentially informative benchmark for potential Xiaflex uptake would be the number of hydrodilations that are done each year.  It seems it is probably not done that often since it doesn't have its own CPT code.  Or maybe it just doesn't have enough published data yet to get its own CPT code.  It might only be used by a subgroup of surgeons in the space.  But this seems like easy pickings to switch to Xiaflex since the procedures are so similar. 


Also of note is that the BESS payment rates for MUA and surgery are only in the ~$200 range.  CPT payments are very hard to accurately tease out, but these seem significantly less than the rates for DC.  Judging from this, the economic incentives to perform surgery are not very high in FS, which could potentially further benefit Xiaflex uptake. 


The current orthopedics sales force will easily be able to sell Xiaflex to surgeons.  This is what AUXL said in September of 2011:


The frozen shoulder one would fit very nicely with our current focus in orthopedics. I think that as you know, many of these hand surgeons practice in offices with other sub-specialists, shoulder guys, knee guys and hand guys, and so we're in those offices today, so I think that's one that our current organization would be very capable of handling.





On a DCF valuation AUXL is worth $37/share on just Peyronie’s Disease and the current approved products.  This assumes no value for Cellulite or Frozen Shoulder.  DCF is an appropriate valuation for a stock like this where the cash flows are easy to model given the patent life. 




Looking at comps in the mid cap US biotech group we see a wide dispersion on P/E.  This is because many companies are in different stages of their development.  Overall AUXL is trading at a discount to the group for 2015 P/E multiples.  It is difficult to compare 2013 and 2014 because AUXL is just turning profitable then, and a number of the stocks in the comp group are much more mature in their life cycles.  If AUXL traded at the group median for 2015 PE it would be worth $53/share.   The most comparable growth stocks in the universe are ONXX, BMRN and ALXN to some extent.  ALXN trades at a premium to the entire group.  But ONXX and BMRN both will turn profitable for the first time in the next 2-3 years.  AUXL trades at a discount to both these stock on 2015 numbers.




EV/REV is often used as a valuation metric for mid cap biotechs rather than P/E.  On an EV/REV basis AUXL is also trading at a discount to the group.  In fact if you look at the higher quality names with revenue growth and/or long patent lives such as BMRN, ONXX, VRTX, ALXN you see they have 2014 EV/REV of 5-10x or 5-6x excluding ALXN (which is trading at a premium to the group).  A 5x revenue multiple on 2014 revenue would equate to $46/share valuation for AUXL.




M&A is also a very real possibility for AUXL given the shortage of biotech firms with strong revenue growth prospects and big pharma desperate need to bridge their patent cliffs.  Assuming 50% SG&A savings for an acquirer would equate to a $52/share DCF keeping the 10% discount rate and assuming zero revenue synergy and zero probability of success for cellulite or frozen shoulder.  Testim and Xiaflex for PD would be strong additions to a number of pharmaceutical companies looking to fully leverage their existing urology sales force.  If cellulite or frozen shoulder work, every pharmaceutical company with a PCP or dermatology sales force would be interested in AUXL.  Adrian Adams, the CEO also has a long history of maximizing shareholder value by selling to larger pharmaceutical companies.  See below for more on Adrian Adams. 


Adrian Adams

Adrian Adams became CEO of AUXL on December 8, 2011.  He has a long history of having his companies be acquired.  Immediately prior to becoming the AUXL CEO Adrian was CEO and chairman of Neurologix, a small pink sheet public biotech company.  He had only joined Neurologix in September 2011.  Prior to that he was CEO of Inspire Pharmaceuticals which he sold to MRK for $430M.  This was a 26% premium to the stock price the day prior to the deal announcement of April 5, 2011.


Adrian was CEO of Sepracor from May 15, 2007 until February 19, 2010.  Sepracor was acquired by Dainippon Sumitomo Pharma Co., for $2.6B or a 27.6% premium to the stock price before the deal was leaked. 


Mr. Adams was CEO for Kos Pharmaceuticals from 2002 until 2006 when it was acquired by ABT for $3.7B, a 56% premium to the stock prior immediately prior to the deal. 


During his 30 years of experience, Mr. Adams also held general management and senior marketing positions at ICI (now part of AstraZeneca), SmithKline Beecham and Novartis. He has extensive international and national experience and has been instrumental in launching major global brands in addition to driving successful corporate development activities encapsulating financing, product and Company acquisitions, in-licensing and company M&A activities. Mr. Adams graduated from the Royal Institute of Chemistry at Salford University in the U.K. Mr. Adams serves as a director of Amylin Pharmaceuticals.


Adrian clearly has an impressive history dressing up a company to be acquired.  His last three companies were acquired (excluding Neurologix).  Adrian has already shown the power of his rolodex by signing the GSK and Actelion deals.  Both deals have been impressive in creating value for the company.  To receive substantial milestones for countries that normally have very little value associated with them (Canada, Australia, Brazil and Mexico), when the prior management team had been unable to execute speaks to Adrian's ability to drive shareholder friendly deals.  Similarly, the GSK deal not only brings additional marketing muscle to the Testim franchise, but it does so without minimum payments and it also brings another potential acquirer to the table if an auction process is run.  By signing the co-promotion deal with GSK, AUXL now has partnerships with PFE, GSK and Actelion - any of which might be interested in acquiring the company at a later stage. 



Testim could continue to lose market share despite the GSK co-promotion agreement.  Testim could also come under pressure if WPI or PRX is undisciplined with their pricing for Androgel generics. 


The Peyronie's indication could come under substantial regulatory scrutiny since it is a new disease area lacking established precedents.  AUXL has worked with the endpoints team at the FDA, and so the PDQ and penile curvature endpoints should be well vetted with the FDA.  Furthermore, PFE also has experience getting drugs approved for new disease areas in urology and has provided guidance to AUXL. 


The sales of Xiaflex in Dupuytren’s Contracture could stagnate and not grow as projected.  This would lead to serious lack of scale for the orthopedic sales force if cellulite and FS do not work.  In this case, AUXL would probably try to scale back the sales force and marketing costs to bring the cost structure more in line with the opportunity.  It would also put additional pressure on Adrian Adams to sell the company.

































Income Statement







Testim is a testosterone gel for men with low testosterone levels that did $205M in US sales in 2011.  As men get older, their testosterone levels tend to drop.  By restoring their testosterone levels to normal, men improve their sex drive, increase their muscle mass, decrease body fat and increase bone density.  Testosterone is very difficult to deliver through pills, thus it needs to be delivered into the body through gels applied on the skin, or patches or injections.  Gels are the most popular form of testosterone delivery since they do not cause skin irritation (as patches do) nor do they require shots (as injections do). 


Testosterone gels did $312M in sales in the 1Q of 2012, so annualizing at a $1.2B rate.  Underlying script growth was 9.6% quarter over quarter, and an amazing 33% for year over year growth due to the introduction of new products.  However, the growth of the category has been strong for a number of years. 





Testim is the number two product in the category with ~18% of scripts in the testosterone gel category.  Though they have been losing share they recently signed a deal with GSK that will bring them additional marketing support.  Though they will have to share some of the upside economics from Testim, they will now be able to reach significantly more prescriptions than they were able to previously.  This increases their number of scripts targeted from 2M to 3M a substantial increase, with very limited impact in fixed cost SG&A.  AUXL as a relatively small company had a limited sales force that only targeted the top prescribers - the co-promotion agreement with GSK is a smart way to increase their reach without a large increase in their cost structure. 





With GSK's help Testim should be able to hold or regain its market share in this rapidly growing market. 


This market is highly promotion sensitive.  The testosterone gels themselves are not very differentiated. The differentiation in this category is not around a differentiated clinical profile - after all they all do the same thing, increase your testosterone levels.  Patients and doctors often choose based on small things like the Androgel pump dispenser or the Testim gel backs.  So this is a very promotionally sensitive market.  Until recently there were no major pharmaceutical companies in this market.  Androgel, the market leader, was originally owned by Solvay.  It was not until 2010 that ABT completed its acquisition of Solvay and thus became the owner of Androgel.  In 2011 we had two new market entrants, LLY with Axiron and ENDP with Fortesta.  Thus it went from a two player market with only AUXL and Solvay promoting to a four player market with two major powerhouses (LLY and ABT) marketing.  All of a sudden the airwaves (especially during golf broadcasts) were inundated with commercials about "Low T."  With the additional marketing muscle of the new entrants, the number of sales details to doctors exploded, and with it sales of the category.





Testim has been growing well, despite the entrance of a number of new competitors.  They recently signed a deal with GSK, to help co-promote the drug to a wider audience of doctors.  This should be a growing product, despite them losing market share because the market is growing so rapidly.  TRx grew 33% year over year for the category in the 1Q 2012. 


The big question with Testim is what happens in 2015 when Androgel 1% authorized generics come onto the market.  Androgel 1% is the market leader in the testosterone gel market. 




The testosterone gels are in a unique position with generics.  Testosterone is the active ingredient and there are penetration enhancers (PE) that get the testosterone through the skin an into the circulation.  So normally it would be very easy to show that PK/PD is the same for a generic testosterone gel as for the branded.  However, there were cases of testosterone transfer to children which resulted in enlarged genitalia and so the FDA called an advisory panel for the testosterone gel and implemented a REMS.  Because of this risk, the FDA determined that not all PE are the same, and so if you file for a generic testosterone gel, you need to conduct certain transfer studies to show that the risk of transfer with your PE is acceptable.  So now all generic testosterone gels with different PE have to go through the 505b2 pathway, rather than through the ANDA process.  This creates a huge barrier to entry.  The patents for all the testosterone gels are around the PE, so if you use the same PE you will violate the patent.  But if you don't use the same PE you will have to file through a 505b2 pathway rather than ANDA and run studies and probably not get an AB rated drug (it is very rare for the FDA to grant AB rating for a 505b2 drug.)  


However, because Solvay (which was bought by ABT) had settled with WPI and PRX on Androgel 1% before this had happened they should be able to enter the market in August 30, 2015 and Feb 16, 2016 respectively.   But since this settlement was done before the FDA had concerns around transfer it is unclear if this will in fact be AB rated.  If you go to the FDA website on Testim 1%, it says "There are no Therapeutic Equivalents" below.  It is unclear if this is simply because WPI and PRX are not on the market or if something else is at play. 




There are also filings by TEVA and PRGO for which ABT sued.  Since these were done post the advisory panel, they were filed under 505b2 pathway and are unlikely to be AB rated. 


Furthermore the Androgel 1.62% switch is going extremely well.  In just one year it has already converted over 50% of the Androgel 1% market.  ABT has 3 more years to convert the market to 1.62% before WPI enters the market. 





Furthermore, with limited competition coming, WPI and PRX have little incentive to compete on price.  Rather they will price at a slight discount and try to maintain the market since there will be limited competition for a while.  If we look at Adderall XR in the ADHD market for example, we see how disciplined the players are.  Shire, the branded player in this market, saw a drop with Adderrall XR generics (2 generics were introduced), but the market share low quickly flattened, and their other brands (Vyvance and Intuniv) have continued to grow in this market despite the presence of generics.





We also see below that pricing has been stable in the market.  Generic companies recognize a oligopoly when they see it and will act in their best interest - to keep price as high as possible and maximize their own economics. 





I believe this is a good proxy for the Testosterone gel market.  Par and Watson have every incentive to be disciplined when they enter the market.  However, I still conservatively model both price and script loss for Testim post 2015. 




Xiaflex in Dupuytren's Contracture

In DC, the tough connective tissue within one's hand becomes abnormally thick which can cause the fingers to curl and can result in impaired function of the fingers, especially the small and ring fingers.


It usually has a gradual onset, often beginning as a tender lump in the palm. Over time, pain associated with the condition tends to go away, but tough bands of tissue may develop.  These bands, which are the source of the reduced mobility commonly associated with the condition, are visible on the surface of the palm and may appear similar to a small callus.


The cord in DC is made of collagen.  Xiaflex is collagenase, an enzyme that breaks down collagen. 


There are a couple of treatments now for DC.  One is surgery, where you cut the finger open and carefully remove the cord.  The other is called needle aponeurotomy, where you try to break up the cord by repeatedly stabbing it with a needle.  The cause of DC is unknown and neither of these treatments treats the underlying condition that cause the cord to form in the first place, so it makes sense that over time the cord would come back.  Xiaflex is no different, it is like a chemical injection that dissolves the cord, you then come back the next day to "pop" the cord once it has been weakened from the collagenase. 


Here are pictures of injection with Xiaflex at the top row, and with surgery in the bottom. 




Xiaflex's launch in Dupuytren's contract has been a disappointment.  Though there are many surgeons who have completely switched over to Xiaflex from surgery.  However, there are also many surgeons who have refused to switch over from surgery to Xiaflex.  Partly that may be due to the lower reimbursement for performing the Xiaflex procedure than surgery (especially since many of them own surgery suites.)  However, on a dollar per hour basis, it is actually not that different since Xiaflex only takes 15 minutes to perform vs. hours for the surgery.  But surgeons are not used to buy-and-bill drugs, they are not used to filling out J-codes, and their practices are not amenable to seeing a patient two days in a row (the Xiaflex label now asks for patients to come in the following day after an injection to get manipulated).  Another issue is that surgeons can only do one injection a month.  So if they have multiple cords to treat, you may have to get treated over 2-3 months.  These are all things that can be corrected with time. 


In particular, there is going to be data on the multi-cord study which should report soon.  And they will soon initiate a trial to get this data onto the label.  ~40% of patients have more than one cord at presentation, so once this data is available it will give doctors comfort in using it in multiple cords at once - as opposed to having to wait a month between injections.  This will lead to an increase in usage in the second half of 2012 and for the next few years.  Price increases and usage in the non-surgical population will also drive continued increases in adoption. 






PDQ Bother Domain Items


Thinking about the last time you had an erection, how bothered were you by any pain or discomfort you may have felt in your erect penis?

Thinking about the last time you ejaculated, how bothered were you by any problem you may have had ejaculating?

Thinking about the last time you looked at your erect penis, how bothered were you by the way your penis looked?

Thinking of the last time you had or tried to have vaginal intercourse, how bothered were you by any physical problem you may have had?

How bothered are you with having vaginal intercourse less often?


1—Not at all bothered

2—A little bit bothered

3—Moderately bothered

4—Extremely bothered

Total potential score 0 to 20


Viagra Trials

From the NEJM publication (1998)

Intention-to-treat analyses were performed on all variables and included all the men who were randomly assigned to treatment (and received treatment) and who had any assessments after base line, regardless of protocol deviations or whether the men completed the study. All statistical tests were two-sided.


Baseline in the dose response had 216 for placebo and 316 for sildenafil.  For question 3, the placebo only has n=199.  Drug has 96+105+101 = 302.  So 17 pt difference between baseline and primary endpoint analysis for placebo, and 15 pts for sildenafil. 


Baseline for the dose-escalation study had 166 for placebo and 163 for sildenafil


For question 3, there are 138 pts vs. the 166 that started off for placebo.  For the drug patients the 163 at baseline has become 138 as well. 


166-138 = 28 pt difference between baseline and primary endpoint analysis for this 12 week dose escalation study.  28/166 = 16.8% difference.


163-138 = 25 or 25/163 = 15.3% pt difference between baseline and primary endpoint analysis for this 12 week part of the study. 








From the JAMA publication (1999)

Intention-to-treat analyses were performed on all efficacy variables and included all subjects who had a baseline measurement and at least 1 measurement after the start of treatment.  All analyses of significance were 2-sided and tested at the 5% level.  All patients who received the study drug were included in the safety analysis. 














2006 Study of Collagenase in Cellulite


Collagenase Injection in the Treatment of Cellulite


Alexander B. Dagum, M.D., and Marie A. Badalamente, Ph.D.


INTRODUCTION:  There remains no effective treatment for dimpling of the thighs and buttock skin known as cellulite (1, 2). We postulated that if the anchoring connective tissue septae and/or hypertrophied subcutaneous adipose tissue could be lysed by the enzyme, collagenase that the contour irregularity would be improved.  The purpose of this pilot, Phase 2, FDA regulated study was to test the safety and efficacy of open label collagenase injection as a treatment for cellulite.


METHOD:  Ten patients, all female, mean age 41 + 10 years entered the study. The mean body mass index (BMI) was 28.  A 10x10 cm area was outlined on the posterolateral thigh and 0.58mg collagenase (Biospecifics Technologies Corp, Lynnbrook, NY) injected.  Patients were followed at days 1, 7, 30, 90 and 180 days post injection for reduction/elimination of the cellulite appearance in the injected area.  The area of cellulite was quantified in cm with photo documentation.


RESULTS:  There was a significant reduction in cellulite appearance of the injected area (Figures 1). Cellulite area was reduced by 77% by day 1 in comparison to baseline. This result was sustained in the longer term.  In comparison to baseline, cellulite area was reduced by 74% at 1 week, by 89% at 1 month, by 86% at 3 months and by 76% at 6 months.  Patient satisfaction score was 1.75 at 6 months (1= completely satisfied, 4= not satisfied) BMI and thigh circumference showed no significant differences from baseline values.  Side effects included injection area soreness, ecchymosis and mild edema which resolved well in a mean of 10, 18 and 4 days, respectively.


CONCLUSIONS:  This study has shown that collagenase injection for cellulite is safe and has merit.  The collagenous connective tissue septae and adipose tissue of cellulite appear to be a good substrate for lysis by injectable collagenase. This therapy warrants continued testing within the FDA regulatory process.








Figure 1. Patient before Collagenase injection, 1 month post-injection and 6 months post injection of collagenase






1. Avram MM. Cellulite: a review of its physiology and treatment. J Cosmet Laser Ther 6(4):181-5, 2004


2. Draelos ZD, Marenus KD. Cellulite. Etiology and purported treatment. Dermatol Surg 23(12):1177-81, 1997 


Cellulaze Approval Press Release


News Release

Cynosure Receives FDA 510(k) Clearance for Cellulaze Cellulite Laser Workstation

World's First Minimally Invasive Aesthetic Device for Cellulite Reduction

WESTFORD, Mass., Jan. 30, 2012 /PRNewswire/ -- Cynosure, Inc. (NASDAQ: CYNO) today announced that the U.S. Food and Drug Administration has cleared the Company's Cellulaze™ Workstation for commercial distribution, giving millions of American women the first minimally invasive solution in their battle against cellulite.


"Cellulaze is the world's first and only minimally invasive medical device designed to treat women who have struggled to eliminate cellulite through diet and exercise, or have tried the myriad of lotions and creams currently on the market," said Cynosure President and Chief Executive Officer Michael Davin. "Unlike those products, Cellulaze is the only aesthetic device that directly treats the physiological structure of cellulite, providing clinically proven, results."  

An estimated 85% of women over the age of 20[1,2]  have some form of cellulite – pockets of fat deposited just beneath the surface of the skin, around the hips, thighs, and buttocks, which produces an "orange-peel" or a "cottage cheese" appearance.[3] In the United States alone, the treatment of cellulite is a multibillion-dollar industry, according to market estimates.

Cellulaze reduces cellulite – in just one treatment – by restoring the normal structure of the skin and underlying connective tissue. Cellulaze diminishes the lumpy pockets of fat, releases the areas of skin depression typifying cellulite and increases the elasticity and thickness of the skin. 

Nearly four years of clinical research puts Cellulaze in a class of its own: the only single-treatment aesthetic procedure clinically shown to reduce cellulite.  Clinical data has demonstrated that Cellulaze treatments can increase the thickness of the skin by 25% and increase the elasticity of the skin by 29% at one year.  No other anti-cellulite treatment has been proven to deliver these results.

"Based on the scientific data and the patient results we have observed over the past several years of evaluating the device, Cellulaze represents a new and highly effective standard of treatment for cellulite," said Barry E. DiBernardo, MD, a board certified plastic surgeon and Clinical Associate Professor, Department of Surgery, Division of Plastic Surgery, University of Medicine and Dentistry of New Jersey.  Dr. DiBernardo, one of five clinical investigators in a multi-center study of Cellulaze, has published one year results on a subset of the treated patients.[4]

"The study was designed with both traditional, subjective measurements of clinical improvement and new objective, quantitative measures as well.  All of the analysis led to the conclusion that patients treated with Cellulaze maintained results at least one year," said Dr. DiBernardo.

Cellulaze uses a proprietary SideLight 3D™ side-firing technology to target the causes of cellulite beneath the skin: herniated pockets of fat, stiffened septae and thin skin. During the procedure, a tiny laser fiber is inserted under the skin, which gently heats the skin and disrupts or melts the fat. Next the laser releases the fibrous bands that create the dimpling effect from pulling down on the skin. This release creates a smoother look.  Finally the laser's energy stimulates collagen production to increase skin's thickness and elasticity for a more even, healthier appearance.

Patients usually see improvements within a few weeks, with the most significant results continuing to develop for a few months following the procedure.  Cellulaze also complements other body shaping technologies, such as Cynosure's Smartlipo™ family of workstations.

Cellulaze is also currently being marketed in Canada, the European Union, Australia and South Korea.  Cynosure is pursuing regulatory approvals in additional international markets.  

About Cynosure, Inc.

Cynosure, Inc. develops and markets aesthetic treatment systems that are used by physicians and other practitioners to perform non-invasive and minimally invasive procedures to remove hair, treat vascular and pigmented lesions, rejuvenate the skin, liquefy and remove unwanted fat through laser lipolysis, reduce the appearance of cellulite and treat Onychomycosis. Cynosure's products include a broad range of laser and other light-based energy sources, including Alexandrite, pulse dye, Q-switched, Nd:YAG and diode lasers, as well as intense pulsed light. Cynosure was founded in 1991. For corporate or product information, contact Cynosure at 800-886-2966, or visit

Forward-looking Statements

Any statements in this press release about future expectations, plans and prospects for Cynosure, Inc., including those related to the market potential of Cellulaze, as well as other statements containing the words "believes," "anticipates," "plans," "expects," "will" and similar expressions, constitute forward-looking statements within the meaning of The Private Securities Litigation Reform Act of 1995. Actual results may differ materially from those indicated by such forward-looking statements as a result of various important factors, including the market's acceptance of Cellulaze, the global economy and lending environment and their effects on the aesthetic laser industry, Cynosure's reliance on sole source suppliers, the inability to accurately predict the timing or outcome of regulatory decisions, changes in consumer preferences, competition in the aesthetic laser industry and economic, market, technological and other factors discussed in Cynosure's most recent Annual Report on Form 10-K and Quarterly Report on Form 10-Q, which are filed with the Securities and Exchange Commission. In addition, the forward-looking statements included in this press release represent Cynosure's views as of the date of this press release. Cynosure anticipates that subsequent events and developments will cause its views to change. However, while Cynosure may elect to update these forward-looking statements at some point in the future, it specifically disclaims any obligation to do so. These forward-looking statements should not be relied upon as representing Cynosure's views as of any date subsequent to the date of this press release.

[1] Harvard Women's Health Watch Cellulite meltdown. Harv. Health Pub. Group 5, 7 (1998).

[2] Sainio, E.L., Rantanen, T. and Kanerva, L. Ingredients and safety of cellulite creams. Eur. J. Dermatol. 10, 596– 603 (2000).

[3] Drealos, Z.D. and Marenus, K.D. Cellulite: etiology and purported treatment. Dermatol. Surg. 23, 1177–1181 (1997).

[4] DiBernardo, BE, "Treatment of Cellulite with One Year Follow-up using a 1440nm pulse laser" Aesthetic Surgery Journal, March 2011

Links to the Frozen Shoulder Study Press Release and site


  • Estimates and valuations will rise as the investors recognize the high probability of approval in Peyronie's Disease, coupled with the large commercial potential and substantial sales force leverage in the next 12-18 months.
  • Data on cellulite and frozen shoulder will read out over the next 12 months.  Despite having solid proof of concept data and offering 100%+ upside to the stock if successfully developed, they are not in street models. 
  • The stock is worth $37 as a standalone with just the existing indications and very conservative Peyronie's Disease sales.  If cellulite or frozen shoulder is successfully developed they would be worth an additional $59/share and $35/share respectively. 
    show   sort by    
      Back to top