By Daniel MacGowan, MD, MRCPI
Assistant Professor of Clinical Neurology, Weill Cornell Medical College, Cornell University
Dr. MacGowan reports no financial relationships relevant to this field of study.
SYNOPSIS: A novel drug combination was taken from an animal model and tested in a Phase II trial of patients with amyotrophic lateral sclerosis. Although there was a trend toward benefit, there were no statistically significant endpoints at the end of the trial. Future trials need to enroll a larger number of patients.
SOURCE: Paganoni S, Macklin EA, Hendrix S, et. al. Trial of sodium phenylbutyrate-taurursodiol for amyotrophic lateral sclerosis. N Engl J Med 2020;383:919-930.
Paganoni et al have published results of a six-month, multicenter, double-blind, placebo-controlled treatment trial for patients with definite amyotrophic lateral sclerosis (ALS). Using a combination of 3 grams of sodium phenylbutyrate with 1 gram of taurursodiol, they randomized patients at a 2:1 ratio vs. a taste-matching placebo sachet once daily for the first three weeks of a 24-week trial, followed by twice daily for the remaining 21 weeks.
The primary outcome was the rate of decline (slope) of the ALS Functional Rating Scale (ALSFRS) score at 24 weeks. Patients had to meet El Escorial criteria for definite ALS (clinical upper and lower motor signs in at least three regions) within 18 months of the first symptom onset and a slow vital capacity (SVC) of > 60% of expected for gender, height, and age. Patients were allowed to be cotreated with the Food and Drug Administration-approved agents riluzole and/or edaravone.
Secondary efficacy outcomes were the rate of decline in isometric muscle strength as measured by the Accurate Test of Limb Isometric Strength (ATLIS) device score, the rate of decline in plasma levels of the phosphorylated axonal neurofilament H subunit (pNF-H, a potential biomarker of motor neuron degeneration), the rate of decline in the SVC, time to death, tracheostomy, permanent assisted ventilation, or any hospitalization. Sodium phenylbutyrate is a histone deacetylase inhibitor that has been shown to upregulate heat shock proteins and act as a small molecular chaperone, thereby ameliorating toxicity from endoplasmic reticulum stress. Taurursodial prevents translocation of the Bax protein into the mitochondrial membrane, thus reducing mitochondrial permeability and increasing the apoptotic threshold of the cell. Both compounds have been demonstrated to be neuroprotective through these mechanisms in superoxide dismutase 1G93A transgenic mice.
A shared-baseline, mixed effects regression model with no added covariates, using data from an old, larger ALS ceftriaxone trial, was used to calculate that a cohort of 131 recruited patients, randomized 2:1 between active drug and placebo, followed over the complete six-month trial period, would be needed to provide 80% power to detect a 30% difference in the slope of ALSFRS decline. Of 177 patients screened, 137 were eligibly randomized, with 89 to active drug and 45 to placebo. Two patients in the active drug group died before their first post-treatment evaluation at three weeks, reducing the active drug cohort to 87. Twenty patients taking the drug (23%) and 10 taking the placebo (21%) discontinued follow-up. Fourteen patients taking the drug (14%) and six patients taking the placebo (12%) terminated participation, while three patients taking the drug died (3%) and three patients taking the placebo (6%) died or underwent tracheostomy. One patient in the placebo group was lost to follow-up. This resulted in only 67 patients in the active drug group (77%) and 38 in the placebo group (79%) completing the 24-week treatment trial, although only 60 patients taking the active drug (69%) and 37 taking the placebo (77%) were available for a seven-month follow-up telephone interview. Therefore, the study ultimately did not provide enough statistical power to detect a statistically significant decline in ALSFRS between the combination of these drugs and placebo.
The authors acknowledge an average 20% dropout rate in many ALS trials, meaning that it would come as no surprise that of the 137 patients enrolled, only 105 patients completed the six-month trial, a drop-out rate of 23%. This would have been prevented by enrolling 170 rather than 137 patients. As it turned out, the intention to treat analysis of the 87 actively and 45 placebo-treated patients returned a positive trend favoring the drug combination of reducing the decline of ALSFRS over 24 weeks by a squared mean score of 2.32 (0.18 to 4.47, 95% confidence interval [CI], P = 0.03). This mean square decline of 2.32 is a small effect given the ALSFRS scale is a sum-score of 12 functional domains, each measuring 0 to 4, range 0 to 48. There were similar mild trends favoring the drug combination over placebo in the total ATLIS upper and lower extremity muscle strength score represented as a percentage of the normal expected value (2.82%, -0.67% to 6.31%, 95% CI), the SVC as a percentage of normal predicted value (5.11%, -0.54% to 10.76%, 95% CI), a hazard ratio risk reduction of death or tracheostomy favoring the drug combination of 0.63 (0.11 to 3.92, 95% CI), and reduced risk of hospitalization (0.54, 0.27 to 1.12, 95% CI). Paradoxically, the active drugs were associated with an 8% average increase in plasma pNF-H levels at 24 weeks, an unexpected finding arguing against a neuroprotective effect. The coadministration of riluzole or edaravone had no effect on the primary or secondary outcomes. The combination of 3 grams of sodium phenylbutyrate and 1 gram of taurursodial twice a day in this trial was associated with a statistically significant increase in adverse effects of diarrhea, nausea, sialorrhea, and abdominal discomfort vs. placebo, leading many subjects to drop out and reduce the dose.
COMMENTARY
In summary, this trial was underpowered from the beginning by a failure to enroll an adequate number of patients to compensate for the 23% expected drop-out rate. The clinical trends of declines in the slope of ALSFRS decline, SVC decline, total isometric ATLIS muscle strength score, risk of death, tracheostomy, or hospitalization favoring the combination of sodium phenylbutyrate and taurursodial were all minor and not clinically or statistically significant. In addition, sodium phenylbutyrate and taurursodial were not associated with an expected decline in pNF-H level. An Italian Phase III randomized, double-blind, controlled trial of taurursodial alone in ALS is expected to release its results in early 2021. Of note, taurursodial is available in 300 mg capsules as a nutritional supplement and is associated with diarrhea at high doses of 1 gram or more. ALS patients will need to take this into account if considering self-dosing with commercially available taurursodial.