Tonix Pharmaceuticals — Update 3 October 2016

Tonix announced concurrently with the release of the results from the Phase III AFFIRM clinical trial for fibromyalgia that it would be devoting the entirety of its resources to the development of TNX-102 SL (cyclobenzaprine sublingual tablets) for PTSD. This recalibration will allow the company to conserve resources while pursuing the development of the drug for PTSD, an indication that recently showed significant results in the Phase II AtEase clinical trial.

The company reported results from the Phase III AFFIRM study of TNX-102 SL for the treatment of fibromyalgia on 6 September 2016. Although results were positive for an array of pain, fibromyalgia symptom and global function analyses, the trial failed to demonstrate statistical significance for the primary endpoint of an increase in the proportion of patients with 30% or greater improvement in pain (p=0.095). This is surprising considering that this metric did reach statistical significance in the earlier BESTFIT Phase IIb clinical trial (p=0.030). The company identified a disproportionate number of discontinuations in the active arm (22.5% vs 13.6%) as the cause for failure to reach significance, especially those categorized as “withdrawal of consent” (5.7% vs 1.2% for placebo). which can occur for any reason including patients moving away. The result was significant (p=0.012) when the data in the 30% responder analysis was interpreted by imputing missing data with the last available data for those patients who dropped out for reasons other than lack of efficacy or an adverse event (as opposed to imputing baseline data as prespecified for the primary endpoint).

The results were significant by a range of other metrics and analyses (Exhibit 1), indicating that the drug is active for fibromyalgia. However, even if the RE-AFFIRM trial (the Phase III trial that had been initiated in July) did hit the primary endpoint, it is likely that the FDA would have required a third Phase III trial in fibromyalgia for approval, which given the company’s limited resources may have been too high a hill to climb.

Tonix reported positive results for the Phase II AtEase study of TNX-102 SL for the treatment of military-related PTSD in May 2016. The goal of the study was to identify the proper treatment dose of the drug as measured by the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) compared to placebo, and the higher dose (5.6mg) showed a significant improvement in CAPS-5 (p=0.038 by analysis of covariance [ANCOVA]) and other metrics of severity and disability (Exhibit 2).

Subsequently, the company has released more detailed information from the AtEase data set at scientific conferences, which further supports the efficacy of TNX-102 SL for the treatment of military PTSD. Among these data, the company presented a time-course of CAPS-5 scores following treatment, which showed that the 5.6mg dose of TNX-102 SL begins to relieve PTSD symptoms quickly (by two weeks) and that this improvement was persistent through the 12-week study period (Exhibit 3). The improvement showed increased significance among patients with a firmer diagnosis and non-mild symptoms (Exhibit 4). This fast onset is in contrast to SSRIs, the current first line pharmacological treatment for PTSD, which may take up to six weeks to show a clinical response.1

Additionally, the company released the first analysis of the number of patients who clinically remitted following treatment (as defined as a CAPS-5 score under 11). Approximately one-third (28.9%, p=0.10) of patients with a firmer diagnosis and non-mild symptoms (CAPS-5 ≥ 33) were in remission by the end of the clinical trial when treated with the 5.6mg dose (Exhibit 5). This is more than twice the remission rate for patients on the placebo arm (14.3%).

Also, across the different symptom clusters, effect size (based on the Cohen’s d statistic) was greater in those with the baseline CAPS-5 ≥ 33 (see Exhibit 6) with the total effect size equal to 0.53. Note that with Cohen’s d, an effect size of 0.2 is considered a small effect size, 0.5 is considered moderate and 0.8 is considered large.2

The adverse event (AE) profile of the treatment is largely within lines of prior clinical experience with cyclobenzaprine (Exhibit 7). The most common AEs associated with the treatment were local effects on the mouth. Hypoaesthesia (numbness) and paraesthesia (pins and needles) were common in the treatment arms. Other adverse events include somnolence and sedation, which are well known effects of the drug. These AEs did not appear to contribute to discontinuations in the trial as the 5.6mg arm had a higher completion rate (84%) than both the 2.8mg arm (79%) and the placebo arm (73%). The resolution of PTSD symptoms appears tied to improvement in sleep quality and the administration of the drug before bed may have limited the impact of somnolence and sedation on the trial outcome.

The company reported that the FDA accepted its proposed Phase III trial design. An important aspect of this design include increasing the enrollment threshold for CAPS-5 scores to 33, including those patients who showed the largest effect in the Phase II study. Future studies will use the 5.6mg dose and will consist of two planned Phase III trials: a military-related PTSD trial beginning in Q117 followed by a predominantly civilian trial. Topline data from the military-related trial is expected approximately one year later.

The company also stated that a Breakthrough Therapy Designation is possible for military-related PTSD which could expedite its review and also allow for intensive guidance from the FDA. According to the FDA, the qualifying criteria are “a drug that is intended to treat a serious condition and preliminary clinical evidence that the drug may demonstrate substantial improvement on a clinically significant endpoint(s) over available therapies.” Based on the available data and the paucity of positive clinical trials in this indication, TNX-102 SL may qualify for this designation.

We have reduced our valuation to $208m or $8.05 per basic share, from $353m or $18.67 per basic share. We have removed the fibromyalgia program from our valuation, which previously contributed $150m in value. There is the possibility of some residual value from the program if Tonix is able to out license the asset in the future, although we believe that decision will be tied to the future prospects of PTSD program, and we are not including it in our valuation at this time. The valuation was also negatively affected by the June 2016 share offering and the resulting dilution (note that our valuation uses the share count following the overallotment option, which occurred after the end of Q216). These effects are partially offset by rolling over our NPVs to Q216. We expect to update our valuation with the advancement of the PTSD program.

Tonix reported a loss of $9.8m for Q216, predominantly associated with clinical development expenses ($7.5m in R&D spending). We expect spending to decrease throughout the year to reflect fewer R&D commitments ($27m for the year) following the suspension of fibromyalgia development and the completion of the PTSD clinical trial in May. We have reduced our expected spending for 2016 to $39m (from $43m) and to $31m (from $50m) for 2017. The company ended the quarter with $31.2m in cash following the June 2016 equity offering of $10m. This was shortly followed by a full overallotment of an additional $1.4m following the end of the quarter. We currently forecast that the company will need an additional $70m before profitability in 2023. This is a reduction from previous estimates of $140m because of the recent financing and the reduction in R&D spending.