Tonix Pharma (TNXP) Says TNX-102 SL Phase 3 Did Not Meet Primary Endpoint in Fibromyalgia

September 6, 2016 7:11 AM EDT

Get instant alerts when news breaks on your stocks. Claim your 2-week free trial to StreetInsider Premium here.

Tonix Pharmaceuticals Holding Corp. (Nasdaq: TNXP) (Tonix) announced preliminary topline results from its Phase 3 clinical study, AFFIRM, designed to evaluate the safety and efficacy of TNX-102 SL (cyclobenzaprine HCl sublingual tablets), 2.8 mg, in patients with fibromyalgia. AFFIRM was a 12-week randomized, double-blind, placebo-controlled trial of TNX-102 SL taken daily at bedtime, in which 519 participants were enrolled at 35 centers in the U.S. Fibromyalgia is a multi-symptom disorder that originates in the central nervous system and is characterized by widespread pain, non-restorative sleep, fatigue, and disability.

The AFFIRM data did not achieve statistical significance in the primary efficacy endpoint: the proportion of patients who reported a 30 percent or greater reduction in pain from baseline to the end of the 12-week treatment period based on the pre-specified primary analysis (p=0.095, Table 1). However, TNX-102 SL did show statistically significant effects on pain when analyzed by other standard statistical approaches (Table 1). TNX-102 SL activity in fibromyalgia was cross-validated by two additional endpoints, Patient Global Impression of Change (PGIC) and Fibromyalgia Impact Questionnaire-Revised (FIQ-R) (Table 2). These endpoints assess global improvement and a range of fibromyalgia symptoms and function. TNX-102 SL showed strong effects on improving sleep quality by the daily diary and the PROMIS sleep disturbance scale (Table 2). The internal consistency of these results provides clear evidence of beneficial effect of TNX-102 SL for the treatment of fibromyalgia.

Seth Lederman, M.D., president and chief executive officer of Tonix, commented, "TNX-102 SL showed broad beneficial effects across key fibromyalgia symptoms and was well-tolerated in the AFFIRM study. Despite achieving clinically meaningful results from AFFIRM, we have greater clarity on the regulatory path forward in our PTSD program. We will therefore discontinue the fibromyalgia program in order to fully focus Tonix’s resources on advancing our potential breakthrough PTSD program to Phase 3. We owe it to our investors, and to patients who are waiting for meaningful clinical innovation, to steward our resources effectively.” Dr. Lederman continued, “We thank those who contributed to the AFFIRM trial, from the clinical teams to the patients and their families. They helped us evaluate this potential new therapy and their involvement provided valuable clinical and scientific information.”

An unexpected imbalance in patient discontinuations for reasons unrelated to efficacy or tolerability (for example, a patient relocating away from the clinical site) (Table 3), created a negative bias in the primary responder analysis because any patient who left the study, for any reason prior to completion, was labeled a non-responder despite their results up to that point. Another standard statistical method for assessing the 30 percent responder analysis that considers the reason for discontinuation showed statistical significance in the primary pain data (Table 1, P=0.012).

Overall, TNX-102 SL was well-tolerated in the AFFIRM study and the adverse events reported were similar to those seen in other TNX-102 SL clinical studies (Table 4). There were seven serious adverse events (SAEs) reported during the study: four in the placebo group and three in the active group. No new safety signals were observed; multiple causal factors were involved in each SAE, and all were resolved quickly and without sequelae.

Table 1. Primary and Other Standard Analyses of Pain

Analysis Method Imputation Result
30% Responder Analysis Pre-specified BOCF all discontinuations P=0.095
30% Responder Analysis BOCF for LOE and AE; LOCF for others P=0.012
ANCOVA; MCFB Multiple imputation; LOE, AE and ID considered MNAR P=0.009
ANCOVA; MCFB Multiple imputation; all MNAR except LTF P=0.042
MMRM of MCFB None P<0.001
50% Responder Analysis BOCF all discontinuations P=0.035

AE- Adverse Event; ANCOVA- Analysis of Covariance; BOCF- Baseline Observation Carried Forward; ID- Investigator Decision; LOCF- Last Observation Carried Forward; LOE- Lack of Efficacy; LTF- Lost to Follow-up; MCFB- Mean Change from Baseline; MMRM- Mixed Models Repeated Measures; MNAR- Missing Not at Random

Table 2. Key Secondary Efficacy Data

Measure Analysis Method Imputation Result
PGIC Responder AnalysisBOCF 0.038
FIQ-R Total Score MMRM of MCFBNone <0.001
FIQ-R Symptom Domain MMRM of MCFBNone <0.001
FIQ-R Function Domain MMRM of MCFBNone <0.001
Clinic 7-day pain recall MMRM of MCFBNone 0.003
FIQ-R Pain Item MMRM of MCFBNone <0.001
PROMIS Fatigue MMRM of MCFBNone <0.001
Daily Sleep Quality Diary MMRM of MCFBNone <0.001
PROMIS Sleep Disturbance MMRM of MCFBNone <0.001
FIQ-R Sleep Quality Item MMRM of MCFBNone <0.001

Table 3. Reasons for Patient Dropouts/Discontinuations

Reason TNX-102 SL Placebo
Occurrence of an Adverse Event20 (7.6%) 11 (4.3%)
Withdrawal of Consent15 (5.7%) 3 (1.2%)
Investigator Decision6 (2.3%) 0 (0%)
Lack of Efficacy6 (2.3%) 5 (1.9%)
Lost to Follow-up11 (4.2%) 15 (5.8%)
Other1 (0.4%) 1 (0.4%)
Total 59 (22.5%) 35 (13.6%)

Among subjects randomized to the TNX-102 SL and control arms, 77.5 percent and 86.4 percent, respectively, completed the 12-week dosing period. As observed in other TNX-102 SL clinical studies, the rate of tongue numbness was higher in the active treatment group (40.2 percent vs. 0.8 percent). Transient tongue numbness, the most frequent adverse reaction, is a local effect related to TNX-102 SL sublingual administration and it did not appear to bias efficacy results. The most common systemic adverse reactions occurring in greater than or equal to 3 percent of patients in TNX-102 SL group and greater than placebo, are listed in Table 4.

Table 4. Most Common Systemic Adverse Reactions Occurring in > 3% of Patients in the TNX-102 SL Group and Greater than Placebo

Preferred term TNX-102 SL (N = 261)*Placebo (N=257)*
Fatigue15 (5.7%)6 (2.3%)
Somnolence8 (3.1%)4 (1.6%)

*Safety Population = 518 patients

Serious News for Serious Traders! Try Premium Free!

You May Also Be Interested In

Related Categories

Corporate News, FDA, Management Comments

Add Your Comment