Cymabay Therapeutics Inc.'s chief scientific officer, Charles McWherter, told BioWorld that the concept of liver enzyme scores as a powerful indicator of efficacy vs. liver fat reductions in nonalcoholic steatohepatitis (NASH) "is not an entirely made-up idea, just because we got this result" in the phase IIb study of peroxisome proliferator-activated receptor (PPAR) delta agonist seladelpar. "If you had to choose between reducing fat, which can be harmful or can be inert, or showing evidence of improving the underlying lesions and injury" as a way of showing a drug's worth in NASH, he said, researchers might choose the latter as "more proximal to the injury," McWherter said. As Newark, Calif.-based Cymabay's chief medical officer, Pol Boudes, put it, "the fat story is not the end of the story" in NASH.
Shares of Cymabay (NASDAQ:CBAY) plunged 45.5%, or $5.04, to close Tuesday at $6.05 as Wall Street recoiled from the 12-week top-line NASH results from the 52-week experiment. Reductions in liver fat proved minimal and not significant compared to placebo, but drops in markers of liver injury turned up robust and clinically meaningful. CEO Sujal Shah said officials at the firm "have to own the fact that we went into this study with the assumption that we would see both fat reduction as well as reduction in biochemical markers of liver inflammation and injury," telling BioWorld the half-and-half outcome is "a learning for us, certainly. In the end, irrespective of fat reduction and even alanine transaminase [ALT], the true endpoint to progress to phase III or even in phase III in the setting of NASH comes from a read on histology. It's either NASH resolution or a reduction in fibrosis. You want to get to the biopsy" and allow histological results to speak, as they will at the 52-week mark.
McWherter, addressing the fat endpoint, noted that "30 to 50 percent of folks in the U.S. have fatty liver disease. They're not all going to get NASH. They're not all going to get cirrhosis. The other fact that's been a little confusing, even for people who work in the field, is that if you lose weight or you have bariatric surgery, that's the only intervention so far that really has an impact on NASH. But you wouldn't want to treat everyone that way. Liver injury [such as would be proven in histology] is the setting where the risk is high enough that it warrants a drug intervention." Preclinical data directed Cymabay to the liver fat endpoint, he said. "We were strongly influenced by the results we saw in mice."
The double-blind, placebo-controlled study randomized 181 subjects with biopsy-confirmed NASH and a liver fat content (LFC) greater than 10% to receive either placebo or seladelpar 10 mg, 20 mg or 50 mg once daily. Subjects had NASH with a mean NAFLD Activity Score of 5.2 at baseline, and 83% showed stage 2 or stage 3 fibrosis. Other baseline characteristics included a mean LFC by magnetic resonance imaging-proton density fat fraction (MRI-PDFF) of 21% as well as elevated mean ALT and aspartate transaminase levels of 62 U/L and 46 U/L, respectively. About half of the subjects enrolled had a diagnosis of type 2 diabetes. The primary endpoint was the relative change in LFC from baseline to 12 weeks.
Seladelpar knocked down markers associated with liver injury; ALT declined up to 37.5% or 32 U/L in 12 weeks. Such reductions in ALT are significantly greater than the 17 U/L threshold that has been correlated with histologic improvement in NASH. Gamma glutamyl transferase also fell significantly, which suggests a lowering of hepatocellular oxidative stress. Significant cutbacks in alkaline phosphatase at 12 weeks were observed, too, supportive of a decrease in hepatocellular bile acids. Changes in liver enzymes collectively point to the potential to impact ballooning and lobular inflammation, key components of NASH resolution.
"I can't recall a drug used to treat NASH where there was a positive histopathologic outcome where ALT did not drop," said Stephen Harrison, principal coordinating investigator of the trial, during a conference call Tuesday morning with investors. "We're just getting a look at the 12-week data, and I can't drill down to the level where we can begin to discern what's really happening relative to an individual patient. It's hard to say. There's clearly a dose-response relationship between placebo 10, 20, 50 with seladelpar relative to ALT. So, why is that happening? I think that's just as an important question as why you didn't see a change in fat occurring. What we're seeing in the placebo group at week 12 – I'm interested to see if that's going to be sustainable at 26 and at 52. Because, quite frankly, that placebo response on PDFF is almost twice than we would traditionally expect to see on a placebo arm with that reduction."
SVB Leerink analyst Pasha Sarraf went along with Harrison's optimism for later outcomes, writing in a report that he was "disappointed that Cymabay's experiment didn't work on fat reduction, but we have high conviction based on the mechanism and molecule that we will see meaningful NASH results at 52 weeks, particularly given the statistically significant reductions in ALT."
Long live Ocaliva: analyst
Seladelpar reduced low-density lipoprotein cholesterol (LDL-C) and triglycerides. At 12 weeks, the median percent changes in LDL-C were 7.8, -7.5, -8.4 and -14.4 in the placebo, seladelpar 10-mg, 20-mg and 50-mg groups, respectively. The median percent change in triglycerides were 14.4, -9.1, -4 and -10 in the placebo, seladelpar 10-mg, 20-mg and 50-mg groups, respectively. There were no significant changes in high-density lipoprotein cholesterol. High sensitivity C-reactive protein, a marker of inflammation and cardiovascular risk, decreased by a greater magnitude in seladelpar-treated subjects, with median percent changes of -3.1, -16.7, -20.7 and -22.6 in the placebo, seladelpar 10-mg, 20-mg and 50-mg groups, respectively.
The drug yielded a favorable safety and tolerability profile at all doses. The most common (>5%) treatment emergent adverse events (AEs) included nausea, constipation, dizziness, headache, gastroesophageal reflux disease and upper abdominal pain. Treatment-emergent AEs were mostly mild to moderate in severity and deemed unrelated to seladelpar. Two serious AEs occurred after randomization through week 12, neither of which were believed to be caused by the drug.
Summarizing the data during Cymabay's conference call with investors, Piper Jaffray analyst Tyler Van Buren wanted to know if it was "fair to say that seladelpar is, given the lack of fat reduction but beneficial effect on lipid markers, more similar to elafibranor than maybe we anticipated." Genfit SA, of Lille, France, about a week ago started a phase II trial with elafibranor, which targets the alpha/delta PPAR pathways, evaluating its activity on liver fat quantity and fat composition in nonalcoholic fatty liver disease patients. Phase II results indicate the drug may be an effective treatment for primary biliary cholangitis (PBC), where Genfit began a phase III study earlier this year. (See BioWorld, April 16, 2019.)
McWherter said that, "as we take a step back and look at this data, we might conclude that the fat burning effects of PPAR delta and PPAR alpha or seladelpar vs. elafibranor may in fact be somewhat comparable." What's different, though, are the changes in biomarkers of liver injury. CEO Shah said Cymabay has generated "very strong, compelling data in PBC [with seladelpar] that we believe can position it as being the preferred second-line treatment of choice" and the latest findings in NASH add to the PBC package.
Jefferies analyst Michael Yee said in a report that, "while positive changes in other markers for liver inflammation are noted and look good [with seladelpar], the results will be viewed as unfavorable and a modestly negative read-through to the phase III trial" with Genfit's candidate. "For the details-oriented investors, we do see Cymabay data as mostly negative on just liver fat however (while ALT and other markers were good) and fat isn't implicated directly with the registration endpoint of NASH resolution, so Genfit's drug probably still has about a 35% chance of working. The next competitor [Conshohocken, Pa.-based Madrigal Pharmaceuticals Inc.'s MGL-3196 (resmetirom), a thyroid hormone receptor beta agonist that recently started a phase III trial] does have significant fat reduction and positive additional markers but is years off from data," expected in 2022 or after, "so [Intercept Pharmaceuticals Inc.'s] could be the only pill on the market for the next three to four years, particularly with the high hurdle of a fibrosis benefit." The NASH space "has been semi-challenging for investors as they sort out what is good and what is not and how much confidence one can have in phase II data and read-through to phase III," Yee noted.
In February, New York-based Intercept said its obeticholic acid achieved its phase III primary endpoint, demonstrating statistically significant improvement in liver fibrosis without worsening of NASH at 18 months (p=0.0002). The company aims to file for regulatory approval in that indication in the U.S. and Europe during the second half of this year. Branded Ocaliva, obeticholic acid was approved by the FDA for PBC several years ago. (See BioWorld, June 1, 2016.)
RBC's Brian Abrahams said he did not believe expectations for the Cymabay study "were a major overhang on Intercept shares, and seladelpar could still possibly have an effect on fibrosis longer-term that may be measurable on 52-week biopsy or in phase III. Nonetheless, we believe these data do clearly reflect the underappreciated complexity and challenges in NASH, and – in a disease where multiple midstage treatments have failed – reaffirm to us that concerns over competitive products rapidly displacing Intercept's Ocaliva, once it is potentially on the market for NASH, are substantially overdone," he added in a report.