In February 2013, Tonix Pharmaceuticals (NASDAQ:TNXP) management met face-to-face with the U.S. FDA to discuss the design of the clinical program for TNX-102SL, the company's low-dose sublingual formulation of cyclobenzaprine for the treatment of fibromyalgia syndrome. Management believes the results of the meeting were positive, and that a clear path to approval has been outlined. We discuss the proposed plan below:
During the third quarter 2013, Tonix plans to initiate a prospective Phase IIb study to assess the safety and efficacy of TNX-102SL in patients with fibromyalgia. The double-blind / placebo-controlled study will seek to enroll 100-200 patients with fibromyalgia at 10 centers all around the U.S. Patients will be randomized to receive either TNX-102SL (2.8mg) or placebo at bedtime nightly for 12 weeks. The primary endpoint will be pain severity at week 12 as assessed by the Numeric Rating Scale (NRS).
We note this endpoint is similar to that utilized by Pfizer (NYSE:PFE) and Eli Lilly (NYSE:LLY) to gain approval for Lyrica and Cymbalta, respectively, for the treatment of fibromyalgia syndrome. Placebo response will be a key issue to watch. Tonix plans to include a 1-2 week wash-out period prior to randomization. We note the Pfizer Phase III study with Lyrica showed a placebo response rate of nearly 50%.
The company will also collect information on secondary endpoints, including NRS scores at other time points during the 12 week study, the Fibromyalgia Impact Questionnaire (FIQ), and the Patient Global Impression of Change (PGIC). We suspect that the trial will take about 12 months to enroll and report top-line data, at an all-in cost of around $4-5 million.
If the proposed Phase IIb trial noted above is successful, Tonix plans to proceed into confirmatory Phase III registration study. We suspect that the Phase III study will follow a similar design to the proposed Phase IIb study, only slightly larger at 300 patients, and incorporate the same change in NRS from baseline to week 12 primary endpoint, and NRS, FIQ, and PGIC secondary endpoints. Assuming this study initiates late 2014, we believe data can be reported by the middle of 2016, allowing for a U.S. NDA filing in early 2017.
…Pharmacokinetic Reference Study…
Besides the two registration efficacy studies noted above, Tonix plans to conduct a multi-dose pharmacokinetic study with TNX-102SL in comparison to generic cyclobenzaprine in a multiple-day dosing (once daily) study. In the study, peak and trough blood levels of cyclobenzaprine will be measured in subjects that receive either TNX-102SL or generic cyclobenzaprine immediate release for 4-5 consecutive days. The goal of the study is to show a bridge between 2.8mg TNX-102SL and generic 5mg cyclobenzaprine that will allow for use of the generic tablet as a reference product in the planned 505(b)(2) NDA submission.
…Next Day Drowsiness / Cognitive Effects Study…
In order to differentiate TNX-102SL from the readily available generic cyclobenzaprine, Tonix plans to conduct a small study designed to evaluate next morning drowsiness and other cognitive measures following the bedtime dosing of either TNX-102SL or generic cyclobenzaprine immediate release. We believe it is imperative for Tonix to clearly show a benefit to dosing 2.8 mg TNX-102SL versus generic 5 mg cyclobenzaprine on measures such as next morning drowsiness and on other cognitive functions for else market penetration for the product will be negligible.
The commercial failure of Silenor (3 mg or 6 mg doxepin) is a direct result of the lack of differentiation between Silenor and generic 5 mg or 10 mg doxepin. To date, sales of Intermezzo (1.75 mg or 3.5 mg zolpidem tartrate SL) have been a significant disappointment to Transcept Pharmaceuticals and commercial partner, Purdue Pharma L.P. We believe this creates a significant perception hurdle for Tonix with investors, because Transcept did studies to show differentiation between Intermezzo 3.5 mg and generic 10 mg zolpidem (Ambien®), and uptake is still far below what both Transcept and Purdue had hoped.
…Safety Extension Study…
To gain approval of TNX-102SL for chronic use, the U.S. FDA will require long-term safety data in the NDA that includes exposure of the drug in at least 300 patients, with at least 100 patients receiving TNX-102SL for six months and at least 50 patients for one year. We expect that Tonix will start enrolling this extension study following the completion of the Phase IIb trial. At this point, it remains to be seen whether or not Tonix can satisfy the long-term safety exposure requirement from just the Phase IIb study, or whether additional patients from the Phase III study will be required.
Fulfilling the safety exposure requirement with just the patients from the Phase IIb study would put the company on track to file the NDA late 2016. However, if Tonix requires additional exposure from the Phase III study, the NDA may be delayed until the first half of 2017, which is what we currently model.
Quick Review Of Phase IIa Data
In December 2011, data from a Phase IIa study with an oral formulation of cyclobenzaprine was published by the Sleep Disorders Clinics of the Centre for Sleep and Chronobiology, University of Toronto, Ontario, Canada (supported by Tonix Pharmaceuticals Inc.) in the Journal of Rheumatology (38(12):2653-63). The paper is called, "Effects of bedtime very low dose cyclobenzaprine on symptoms and sleep physiology in patients with fibromyalgia syndrome: a double-blind randomized placebo-controlled study."
The trial was a randomized, double-blind, placebo-controlled, dose-escalating, parallel-design study in patients with FM and disrupted sleep. A total of 36 patients (1:1 very low dose (VLD) cyclobenzaprine vs. placebo) were treated for 8 weeks. Doses were taken between dinner and bedtime. The dosage was 1 mg for the first 7 days, after which, if clinically indicated and according to tolerability, the daily dose could be increased up to 4 mg (mean dose was 3.1 mg at week 8). We present a summary of the data below:
- Tolerability: VLD cyclobenzaprine was well tolerated, with only 1 severe AE (headache) compared to five SAEs in the placebo group. The most common treatment related adverse events can be found in Figure 1, and was comparable between VLD cyclobenzaprine and the placebo. There were no serious adverse events in either group.
- Pain: Musculoskeletal pain was rated on a 7-point scale (0 - 6) and showed patients on VLD cyclobenzaprine had a 26% reduction at week 8 (1.7 vs. 2.3, p=0.010) vs. no change for the placebo (p=0.044 vs. placebo).
- Fatigue: Fatigue was rated on a 7-point scale (1 - 7) and showed patients on VLD cyclobenzaprine had a 14% reduction at week 8 (4.3 vs. 5.0, p=0.039) vs. a 4% (not significant) worsening for the placebo (4.9 vs. 4.7) (p=0.126 vs. placebo).
- Tenderness: Application pressure was applied to specific body regions to assess tenderness. Patients on VLD cyclobenzaprine showed a 30% improvement in tenderness (18.6 vs. 14.3, p=0.006) vs. 1% (not significant) for the placebo (16.1 vs. 15.6) (p=0.029 vs. placebo).
- Mood: Mood scores were assessed by the Hospital Anxiety and Depression Scale (HAD). Patients on VLD cyclobenzaprine showed a 24% reduction in HAD score (10.4 vs. 13.7, p=0.012) and 22% reduction in HAD depression (4.9 vs. 6.3, p=0.017) vs. a 4% reduction in HAD score (15.1 vs. 15.7) and 10.4% worsening in depression (7.4 vs. 6.7) for the placebo at week-8. Scores for HAD (p=0.067) and HAD depression (p=0.023) vs. placebo showed encouraging trends.
- Subjective Ratings: The trial assessed both clinician-rated (CGIC) and patient-rated (PGIC) improvement scales for fatigue. Fatigue was evaluated using a scale of 5 ("worse") to 1 ("marked improvement"). VLD cyclobenzaprine treatment was associated with a statistically significant improvement in CGIC and PGIC in fatigue (p=0.003 and p=0.001, respectively). In contrast, placebo treatment did not result in statistically significant changes in any of these measures. However, relative to placebo, VLD cyclobenzaprine treatment did not significantly improve either measure (p=0.133 and p=0.257, respectively).
- Effects On Sleep: Effects on sleep was assessed by using polysomnogram (OTC:PSG) recordings. PSG was performed at screening, at baseline, and typically at Weeks 2, 4, and 8. In the VLD cyclobenzaprine group from baseline to Week 8, total time awake decreased 38.5% (0.8 to 1.3 hrs, p=0.011), while total sleep time increased 12.3% (5.7 to 6.4 hrs, p=0.005), and sleep efficiency 15.6% (73.6% to 85.1%, p=0.023). In contrast, placebo treatment did not result in statistically significant changes in any of these measures. Compared to placebo, VLD cyclobenzaprine treatment did not significantly change total time awake, total sleep time, or sleep. Within the VLD cyclobenzaprine group from baseline to Week 8, VLD cyclobenzaprine treatment did not significantly change the percentages of Stage 1, 2, 3, or 4 or REM sleep.
The Benefits of SL
The SL formulation dissolves quickly under the tongue and provides rapid absorption of cyclobenzaprine across the mucous membrane into the bloodstream. As noted above, this avoids the pH variability and food effect seen with generic oral cyclobenzaprine. It also avoids first-pass metabolism in the liver. This is very important because cyclobenzaprine, when taken orally, is absorbed into the small intestines and transported directly to the liver where it is metabolized into norcyclobenzaprine.
According to the literature, norcyclobenzaprine is a long-lived, psychoactive metabolite believed to interfere with cyclobenzaprine's efficacy over time, potentially contributing to the reported decrease of benefit of oral cyclobenzaprine in fibromyalgia when taken chronically. We note the company has generated pharmacokinetic data showing 2.8 mg of the SL formulation is equivalent to 3.1mg of the oral formulation (mean dose from Phase IIa study).
If Tonix can gain approval for TNX-102SL under the FDA's 505(b)(2) pathway, the drug will be granted three years of marketing exclusivity under Hatch-Waxman law. However, Tonix has been working diligently over the past few years to build patent protection around TNX-102SL beyond FDA granted exclusivity. The company has a number of methods of use and formulation patients granted and filed. For example, management has recently filed patent applications on TNX-102SL which, if issued, it expects to provide protection from generic substitution until 2033.
Tonix is pursuing a multi-pronged patent strategy by seeking intellectual property protection on several aspects TNX-102SL. These include, methods of use for certain known active pharmaceutical ingredients, formulation technologies incorporated into TNX-102SL, and performance characteristics in the human body. We note this is a similar strategy to what Purdue Pharma, L.P. took with OxyContin (extended-release oxycodone) and what Transcept Pharmaceuticals (NASDAQ:TSPT) did with Intermezzo. It is not surprising then that Purdue, already familiar with this strategy, is the commercial partner for Transcept with Intermezzo.
Specifically, with respect to methods of use patents, the invention of bedtime very low dose cyclobenzaprine as a treatment for FM is novel and unexpected, thus Tonix is seeking additional patents to cover the use. Additional protection may stem from novel routes of delivery for cyclobenzaprine, such as the sublingual formulation and delivery with TNX-102SL. With respect to patents related to the performance characteristics, Tonix plans to seek protection for novel pharmacokinetic properties of cyclobenzaprine, as well as its active metabolites, at certain times after administration.
Sales and Commercial Potential
Averaging the data obtained from the NIH, AFSA, and NFA, we estimate there are approximately 8 million Americans living with FM. Over 90% of these patients have sleep problems. Approximately 70% of FM patients report difficult in conducting normal daily tasks, such as light housework due to chronic fatigue. According to Decision Resources, U.S. sales of prescription drugs specifically for the treatment of FM totaled $1.4 billion in 2011. This figure includes sales of Cymbalta, Lyrica, and Savella of $595 million, $504 million, and $110 million, respectively. Despite the availability of FDA approved products, we believe the current treatment options for FM continue to leave many patients dissatisfied. Existing approved FM medications such as Lyrica and Cymbalta, which focus on reducing FM-associated pain and mood disorder, respectively, do little to improve sleep quality. Insomnia medications such as Ambien and Lunesta improve total sleep time, but do little to improve the chronic fatigue associated with FM.
If Tonix can gain approval for TNX-102SL in FM using standardized pain (NRS) endpoints in its Phase IIb and Phase III program, while also demonstrating improvement in symptoms of fatigue and sleep quality, we believe a meaningful market opportunity exists. Frost & Sullivan estimate that 50 million tablets of cyclobenzaprine were sold specifically for FM in 2010. We believe at least a third of FM patients would actively seek out prescription therapy, either as a monotherapy or an adjunctive therapy to Lyrica or Cymbalta, with their physicians support, and try a novel sublingual formulation of VLD cyclobenzaprine. Tonix has successfully designed a superior 2.4 mg sublingual formulation of cyclobenzaprine to the readily available oral 10 mg generic. Does that mean there will be no, or even limited, generic substitution or competition once TNX-102SL hits the market? No.
Based on the generic market, the target population for Tonix with TNX-102SL is over 2 million patients. We think that the sublingual formulation clearly contains meaningful advantages over the generic oral formulation. These include rapid absorption and minimal next-day residual effects - ideal for a bedtime dose, and avoidance of first-pass metabolism and build-up of norcyclobenzaprine - ideal for chronic dosing. Of course, all this will need to be confirmed in clinical trials.
For the purpose of our model, we assume Tonix (and its partner) can capture 5% share. That's one out of every twenty patients currently on generic oral cyclobenzaprine for FM switching to TNX-102SL. At a cost of $6-7 per pill ($2,500 per year by 2020), with decent tier-2 and tier-3 coverage, we see TNX-102SL as a $250 million peak drug. We expect that by 2017, both Lyrica and Cymbalta will be generic, and that Tonix' commercial partner will have one of the only (if not the only) branded prescription medications for FM available.