Join the BioNap Email List!

Thursday, August 6, 2015

Does Celgene's Success With Otezla Strengthen The Investment Case For Can-Fite BioPharma?

By Jason Napodano, CFA

Can-Fite Biopharma Ltd. (NYSE: CANF) is developing CF101 (piclidenoson) for the treatment of psoriasis. In my analysis below, I provide an overview of the psoriasis market and why I believe that Can-Fite's drug may have a place, albeit niche, in this massive market given certain advantages over leading blockbuster therapies. I think the most direct comparison for CF101 is Celgene's Otezla® (apremilast), only just approved in March 2014 and currently annualizing at nearly $400 million in sales.

Quick Background On The Psoriasis Market

Psoriasis is one of the most common chronic inflammatory disease of human skin. The disease most commonly presents on the elbows, knees, scalp, back, and thighs. It is an immune-mediated disease characterized by T cell-mediated hyperproliferation of epidermal keratinocytes. Scientists believe there is a strong genetic component of the disease (1), and there is high comorbidity (10-20%) with other autoimmune disease such as arthritis (2). Though common, the disease is not contagious.

Psoriasis can be triggered by a number of external stimuli, including skin trauma, stress, infection, allergens, illness, and drugs. There are multiple types of psoriasis, with plaque psoriasis being the most common and affecting 80-90% of all individuals. Plaque psoriasis involves the hyperproliferation of epidermal keratinocytes that results in red or white, scaly, and typically itchy skin lesions. There is no known cure for the disease, thus depending upon the severity of the condition and how responsive it is to treatment, some patients are on therapy for life.

Psoriasis is a multifactorial disease that is influenced by both genetics and immune-mediated components. That being said, the pathogenesis of the disease is poorly understood. In the simplest terms, psoriasis is the result of new skin cells forming every 3-5 days as opposed to every 28-30 days in a healthy individual (3). The initiating factor(s) for this phenomenon are still being understood; however, what is known is that there is substantial leukocyte recruitment to the dermis and epidermis that results in the formation of the characteristic plaques. The epidermis is infiltrated by a number of activated T cells that release pro-inflammatory cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-22 (4). The secreted inflammatory signals are what prompt the proliferation of keratinocytes.

According to a research study published in the Journal of the American Academy of Dermatology (March 2014), approximately 7.4 million U.S. adults (3.2±0.5% of the adult population) had psoriasis in 2013 (5). No significant differences are found between males and females. 

Quantification of psoriasis treatment efficacy is done through the use of the Psoriasis Area and Severity Index (PASI). The PASI is a composite score that takes into account the average redness, thickness, and scaliness of the psoriatic lesions (each graded on a 0–4 scale), weighted by the area of involvement (6). The body is divided into four areas: head (10%), arms (20%), trunk to groin (30%), and legs to top of buttocks (40%). Each of the area’s severity is separately scored on a scale of 0 (none) to 4 (maximum) for three clinical signs: redness, thickness, and scaling. The percent of skin involved in each area is then estimated and assigned a score on a 0–6 scale (0 = 0%; 1 = <10%; 2 = 10–<30%; 3 = 30–<50%; 4 = 50–<70%; 5 = 70–<90%; 6 = 90–100%). 

The sum of the severity score is then multiplied by the area score and multiplied again by the weight of the respective section (i.e., 0.3 for body). While absolute PASI score is often utilized as an inclusion criterion for clinical trials, it is the percent change in PASI score that is important in measuring efficacy. For example, PASI75 means the percentage of patient that achieved a 75% or more reduction in their PASI score from baseline. For patients with severe psoriasis, clinicians typically consider at least 75% improvement in disease (PASI75) to be a clinically meaningful improvement indicative of success. However, there is strong evidence demonstrating that 50% improvement in PASI (PASI50) is also a clinically meaningful endpoint (7).

The treatment of mild psoriasis typically involves the use of topical treatments, including corticosteroids, vitamin D, and moisturizers such as mineral oil or petroleum jelly. These treatments may be combined with phototherapy, typically psoralen-ultraviolet A (PUVA) therapy, which causes skin cells to grow less rapidly. According to the National Psoriasis Foundation, approximately 25% of individuals with psoriasis are considered to have the moderate-to-severe form of the disease. Based on the numbers above from the prevalence study, this equates to around 1.9 million adults in the U.S. I will focus the remainder of this section on therapeutic treatment options for moderate-to-severe patients. 

Systemic treatments are typically employed for treating moderate-to-severe psoriasis. There are a number of options available, including both small molecule and biologic treatments:

Cyclosporine: Cyclosporine is an immunosuppressive drug that stops the activity of T cells, which slows the growth of skin cells (8). Cyclosporine is taken daily by mouth and is available as a capsule or a liquid. The drug can provide rapid relief from symptoms, with some patients seeing improvement
in symptoms after two weeks of treatment, particularly with larger doses. However, it may take three to four months to reach a complete level of control; and unfortunately, the FDA recommends the drug
not be used for longer than one year because of potential damage to the kidneys. Other possible side effects include an increase in blood pressure, an increase in cholesterol, excessive hair growth, sensitivity, headache, and fatigue (flu-like symptoms).

Methotrexate (MTX): MTX is a powerful immunosuppressant and anti-inflammatory agent. The drug is part of the standard-of-care and traditional first-line systemic therapy for patients with moderate-to-severe psoriasis. However, MTX can cause serious hematologic side effects including myelosuppression and cytopenia. The drug can also cause serious hepatic, pulmonary and gastrointestinal damage. Patients on MTX should be monitored for hepatic enzyme activity, and as with any powerful immunosuppressive drug, risk of serious infection is elevated with long-term MTX use.

Anti-TNF-α biologics: This class of compounds works by blocking the effect of TNF-α and preventing its activation of T cells. There are several TNF-α targeting therapies on the market, including the monoclonal antibodies Humira® and Remicade® and the recombinant TNF receptor Enbrel®. Each of these compounds is approved for treating multiple inflammatory conditions, with psoriasis and psoriatic arthritis being an important two. The anti-TNF-α therapies are highly effective, with PASI75 scores of 70-80% after 12 weeks of treatment; however, this efficacy is associated with sometimes severe side effects, including opportunistic infections, as well as inconvenient dosing. These medications cost approximately $40,000 per year (9). This is among the largest class of pharmaceutical drugs in the world, with total sales in 2014 in excess of $34 billion (10).

Anti-IL-12/23: Ustekinumab (Stelara®) is a monoclonal antibody directed against interleukin-12 and interleukin-23, two cytokines that activate T cells. Stelara® was approved by the FDA in 2009 for the treatment of psoriasis. Similarly to the anti-TNF-α targeting therapies, Stelara® is highly effective (PASI75 between 60-70% after 12 weeks of treatment); however, patients on this medication run the risk of developing serious infections. The compound is administered via subcutaneous injection and costs approximately $50,000 per year (9). 

JAK Inhibitors: Tofacitinib (Xeljanz®) is a janus tyrosine kinase inhibitor that triggers broad anti-inflammatory effects by reducing T cell activity and suppressing pro-inflammatory cytokines including IL-2, 6, 7, 9, 15, 21, and INFs. Xeljanz® is an oral tablet approved in November 2012 for the treatment of rheumatoid arthritis (RA). Pfizer sold $308 million of the drug in 2014, and $96 million in the first quarter 2015. The company submitted a supplemental biologic license application (BLA) looking to expand the label indication into moderate-to-severe plaque psoriasis in February 2015. Based on the Phase 3 data, the drug offers a PASI75 score of 64% after only 12 weeks. Analysts estimated that Xeljanz® will post sales in excess of $1.5 billion by 2020 (11). 

PDE4 Inhibitors: Apremilast (Otezla®) is a small molecule inhibitor of phosphodiesterase 4 (PDE4), an enzyme responsible for the degradation of cAMP. Thus, an inhibitor of PDE4 results in an increase in cAMP, which is similar to CF101’s mechanism of action (discussed below). Otezla® was approved for the treatment of psoriatic arthritis in March 2014, and then for the treatment of psoriasis in September 2014. The drug was approved for the treatment of psoriasis based on two Phase 3 clinical trials involving over 1,200 patients. After 16 weeks of treatment, 33.1% and 28.8% of Otezla® treated patients achieved PASI75, compared to 5.3% and 5.8% treated with placebo. The most common side effects associated with Otezla® are gastrointestinal symptoms including diarrhea, nausea, and vomiting. Otezla® is marketed by Celgene and costs approximately $22,500 per year. Analysts estimate that Otezla® will post sales in excess of $2.5 billion by 2020 (12).

In total, I estimate the global market for psoriasis drugs is in excess of $40 billion, with sales in the U.S. in excess of $20 billion. Oral drugs like Otezla® and Xeljanz® are expected to gain significant market share over the next several years due to the convenience of dose and improved safety relative to the biologics. I believe this is an enormous market opportunity, and although there are nearly a dozen late-stage biologics under development by some rather large players, there is a niche market opportunity for a drug like CF101 given the unique safety profile and convenient oral dosing.

Can-Fite's CF101 - What You Need To Know

Can-Fite is developing CF101 (piclidenoson) for the treatment of psoriasis and rheumatoid arthritis (RA). For the purpose of this analysis, we will focus primarily on psoriasis, with some brief comments on RA at the end.

CF101 is an orally bioavailable A3 adenosine receptor (A3AR) agonist. Scientists have discovered that the Gi protein-associated A3AR is found to be overexpressed in both inflammatory and cancer cells vs. low expression of the receptor if found in normal cells (13). The mechanism of action of adenosine agonism in inflammatory conditions involves the modulation of the Wnt and the nuclear
factor kappa-B (NF-κB) signal transduction pathways, leading to the inhibition of tumor necrosis factor α (TNF-α), interleukin-6 and -12, macrophage inflammatory proteins and receptor activator of NF-κB ligand (RANKL) (14, 15).

As such, Can-Fite believes A3AR may serve as a target for pharmacological intervention in inflammatory states, such as psoriasis or rheumatoid arthritis. The company has as second clinical-stage candidate, CF102, currently under development for the treatment of hepatocellular carcinoma (HCC). Both CF101 and CF102 have high affinity for the A3 receptor with low off-target binding.

Both compounds have been evaluated in a number of preclinical and clinical studies. No safety issues related to toxicology have been reported thus far from studying the compounds in mice, rats, and dogs. Phase 1 clinical testing in humans showed CF101 to have linear pharmacokinetic parameters in doses up to 10 mg. Single oral doses up to 5 mg and repeated doses up to 4 mg every 12 hours were safe and well tolerated. Increased heart rate was noted after a single 10 mg dose. A dose-related increase in heart rate was noted on day 1 of the multiple dose study; however, by day 7 this effect was no longer present. Importantly, no increase in QT interval was noted with the increased heart rate. Based upon the Phase 1 studies, 4 mg total was selected as the highest dose for Phase 2 studies.

…CF101 in Psoriasis…

Can-Fite originally initiated a Phase 2/3 clinical trial of CF101 in patients with moderate to severe psoriasis in 2011 (16). The study enrolled 103 patients into one of three arms: 1 mg of CF101; 2 mg of CF101; or placebo. After 12 weeks of treatment, those in the placebo arm were randomized to receive either 1 or 2 mg of CF101 and the study continued for an additional 24 weeks.

An interim analysis was performed after 24 weeks of treatment, with the results showing a clear difference between patients in the 2 mg cohort and those receiving placebo. Importantly, there was also a marked improvement in patients who were randomized to the 2 mg cohort following the first 12 weeks on placebo in both PGA and PASI75, as shown in the following figure.

However, on March 30, 2015 Can-Fite announced topline results from the Phase 2/3 study showing that treatment with CF101 did not achieve the study’s primary endpoint, which was evaluation of PASI75 after 12 weeks of treatment (17). The proportion of patients treated with CF101 who achieved PASI75 was 8.5% compared to 6.9% in the placebo group. With respect to PGA, 6.4% of patients treated with CF101 achieved a score of 0 or 1 compared to 3.4% of placebo patients. This was quite surprising data given the clear efficacy signal shown in the interim analysis (picture above).

Nevertheless, Can-Fite continued to analyze data from the study out to 32 weeks. On April 27, 2015, Can-Fite announced favorable data from further analysis of the Phase 2/3 study (18). Specifically, after 32 weeks of treatment, 35.3% of patients treated with CF101 achieved PASI75 with a mean percent improvement in PASI score of 57%. This means that, on average, patients who were treated with CF101 saw a 57% reduction in their psoriatic lesions. In addition, 24.7% and 10.6% of patients achieved PASI90 and PASI100, respectively, at week 32. PASI100 indicates a complete clearance of psoriatic lesions. Historically, patients treated with placebo achieve a PASI90 or PASI100 only 1-2% of the time.

A subset analysis was performed examining patients who had previous systemic therapy compared to those who were systemic therapy naïve. PASI90 at week 32 was achieved in 26.9% of patients who had received no prior systemic therapy compared to 13.7% of patients who had received prior systemic therapy. The company believes that this data allows for the opportunity to develop CF101 as a first-line therapy in previously untreated patients with moderate-to-severe psoriasis who do not wish to be treated with currently available systemic therapies (e.g., monoclonal antibodies) due to safety concerns or due to an aversion to injections.

...Putting The CF101 Phase 2/3 Data In Perspective...

Celgene’s Otezla® is likely to be the main competitor for CF101 in psoriasis patients as both molecules are orally bioavailable and the data seem fairly similar. As noted above, Otezla® is a small molecule inhibitor of PDE4, an enzyme responsible for the degradation of cAMP. The inhibition of PDE4 results in an increase in cAMP, which in turn regulates a number of pro-inflammatory and anti-inflammatory mediators including TNF-α, IL-23, and IL-10 (18). The drug was approved for the treatment of psoriasis in March 2014.

The following table compares the data seen with CF101 in the Phase 2/3 study and Otezla®/placebo from the Phase 3 ESTEEM 1 and ESTEEM 2 clinical trials (sections in gray pulled from graphs):

We also present this data in graphic form based on available Phase 3 data from Celgene presented at the American Academy of Dermatology meeting in July 2015 (19).

There are a couple of takeaways from the preceding table and graphs:

1) CF101 2 mg BID clearly has a slower onset of action vs. 30 mg BID Otezla®.  At week 16, only approximately 10.5% of the CF101 patients have achieved PASI75 vs. approximately 30% for Otezla® (pooled data).

2) However, by week 24 the gap has completely disappeared and both drugs appear to be equal in terms of percent of patients achieving PASI75. By week 32, CF101 appears to be the superior drug, with the trend line continuing to rise (graph on the right) vs. the Otezla® line that looks to have peaked at 16 weeks and is in slow decline.

3) CF101 offers a more durable response. CF101 appears to generate a higher PASI90: 25% of CF101 treated patients experienced PASI90 at week 32 (with that increasing linearly from week 12) compared to 19% of Otezla® treated patients at week 32 (based on Phase 2 data).

Can CF101 Find A Niche In Psoriasis?

Otezla® costs approximately $22,500 per year. Celgene reported Otezla® sales in the second quarter 2015 of $90 million ($85 million U.S. / $5 million Row). By 2020, analysts believe that the drug will post sales in excess of $2.5 billion (12). Based on the available data, although slower in onset, CF101 looks to be superior to Otezla® in efficacy and durability of response. We have yet to see a comprehensive safety presentation on CF101, although data generated to date leaves little area for concern. The most common side effects of Otezla® are diarrhea (5-7% placebo adjusted), nausea (4-6%), headache (3-4%), and vomiting (2-3%). The prescribing insert suggests the drug be titrated over the first five days to minimize adverse reactions.

The fact that CF101 is orally bioavailable has a meaningful impact on potential uptake. It seems clear that the drug is going to have a difficult time competing for any patient that would consider or are encouraged by their physician to take an injectable biologic treatment. Week 16 data for Humira® shows that approximately 71% of patients achieve PASI75. CF101, or even Otezla® for that matter, is not going to compete with a big gun like Humira®. But because the drug looks safer (far lower risk of developing serious infections) and is more convenient (oral vs. injectable), Can-Fite may be able to find a niche in MTX or biologic naive patients not wanted to jump directly into the injectables, or in immunocompromised patients that should avoid the stronger biologic drugs.

Assuming Pfizer receives approval for Xeljanz® in October 2015 (achieved 64% PASI75 at week 12 based on Phase 3 data), patients are going to have a number of oral options available ahead of the injectable biologics. As noted above, Xeljanz® is already on the market for rheumatoid arthritis and analysts estimated the drug post sales in excess of $1.5 billion by 2020 (11).

I'm not suggesting that CF101 will have peak sales in the area of Xeljanz® or Otezla®, but the market is currently ignoring the drug given that Can-Fite currently trades with a market capitalization of only $21 million! I think the efficacy of CF101 in psoriasis looks decent, at least compared to Otezla®, and the safety profile looks superior. For example, Can-Fite saw no discontinuations due to safety or otherwise in the Phase 2/3 study. Based on Phase 3 data on Xeljanz® and Otezla®, discontinuation rates for those drugs average around 5% after only three months on therapy. Risk of serious infection is a primary concern for these powerful immunosuppressors.

Above we noted 7.4 million Americans with some form of psoriasis, approximately 1.9 million of which are classified as moderate-to-severe. I think a drug like CF101 could be an effective treatment option for the 5% of these patients that seek oral medications but are discontinuing drugs like Xeljanz® or Otezla® for whatever reason. That's about 100,000 patients Can-Fite can effectively target with CF101.

Otezla® costs $22,500 per year. That's pretty steep, but the biologics like Humira® and Enbrel® cost even more (estimated at $30-40,000 per year). Similarly, a one year supply of Xeljanz® will cost you over $30,000. If approved, I think Can-Fite can effectively charge around $20,000 for CF101. That puts the peak market potential at $2 billion. With 10% penetration, Can-Fite has a $200 million drug in the U.S. alone. The market in Europe is about 75% of the size, meaning there's another $150 million sales potential overseas.

I'll remind investors that in March 2015, Can-Fite signed a distribution agreement with Cipher Pharmaceuticals (NASDAQ:CPHR) to distribute CF101 for the treatment of rheumatoid arthritis and moderate to severe psoriasis in the Canadian market upon receipt of regulatory approval (20). Under terms of the agreement, Cipher made an upfront payment of CDN$1.65 million, with additional payments of up to CDN$2.0 million upon the achievement of certain milestones plus royalty payments of 16.5% of net sales of CF101 in Canada.

CF101 in Rheumatoid Arthritis

In December 2014, Can-Fite announced the completion of the study design for the RA Phase 3 clinical trial (21). It will be a multicenter, randomized, double blind, placebo controlled parallel group study that will investigate the efficacy of CF101 dosed at 2 mg and 3 mg BID compared to placebo for 12 weeks. Approximately 300 patients are expected to be enrolled and the study is likely to be initiated by the end of 2015.

The opportunity for CF101 in RA is quite large even though CF101 is likely to have somewhat less efficacy compared to biologic treatments (based upon the results from the Phase 2 monotherapy clinical trial), with the trade-off being a much more benign side effect profile. However, what intrigues me the most about the opportunity in RA is the potential to select patients most likely to respond to CF101 treatment through the use of a blood based biomarker test, an approach that was validated by the statistically significant increase in ACR20 in a population of patients pre-selected based on A3AR expression level. In addition, the data showing increased efficacy in patients with no prior systemic treatment is quite exciting, and it will be interesting to see whether these results can be replicated in a larger patient population.

The opportunity for CF101 in RA lends itself to another feature article, which I will look to publish in the near future. That being said, similar to psoriasis, I believe CF101 has niche market opportunity in RA and if approved will likely generated another $300-400 million in global sales.


The company exited the first quarter of 2015 with roughly $9.0 million on cash and cash equivalents, compared to $9.1 million on Dec. 31, 2014. The slight decrease in cash was due to an upfront payment of $1.3 million received from Cipher Pharma offset by operating expenses. When the company reports financial results for the second quarter 2015 I believe cash will still be north of $7 million, with burn averaging only around $0.5 per month.

As noted above, the market capitalization of only $21 million seems exceedingly low and vastly undervalues the opportunity for CF101 in psoriasis and RA, as well as the potential for CF102 in HCC (not discussed in this article). The company also has a preclinical candidate in CF602 for sexual dysfunction (23) and an expanding subsidiary focusing on ophthalmic disorders called OphthaliX (24).

Above I outlined how CF101 could achieve $350 million in sales between the U.S. and Europe given 0.5% market share of the moderate-to-severe psoriasis patients. I'll run down this again quickly. There are approximately 13.1 million sufferers from psoriasis in the U.S. (7.4 million) and EU (5.7 million), of which approximately 25% (3.3 million) are classified as moderate-to-severe. We think the vast majority of these patients would prefer an oral medication over an injectable, assuming they could start out on oral and progress to the stronger biologic drugs as their disease progresses or they fail oral medications. The discontinuation rates for the existing oral drugs like Xeljanz® and Otezla® is around 5%. If Can-Fite can penetrate 10% of these patients, with a cost of roughly $20,000 per year, the market in psoriasis alone equates to $350 million.

If we assume that Can-Fite can initiate a Phase 3 psoriasis trial in 2016, file for FDA approval in 2018, and gain approval in 2019, then we can start to get a sense of the future potential cash flows the company can generate from the sale of the drug to commence in 2020. Typically, and based on the projected trajectory for Xeljanz® and Otezla®, it takes five to six years to achieve peak sales. If we model out the potential revenues and cash flows (50% EBITDA margin) off the sale of CF101, with a 25% discount rate and only a 15% probability of success (which is rather low for a Phase 3 asset), the NPV of CF101 today is over $40 million! Heck, even a 7.5% probability of success puts the value of CF101 at over $20 million, which is about what you can buy the entire company for today, and get the existing cash balance, plus CF102, CF602, and interest in OphthaliX for free. As such, I conclude that Can-Fite offers an attractive risk / reward for investors given the depressed stock price and heavily discounted opportunity for CF101.

No comments:

Post a Comment