Intrigued by results from a 13-patient, phase II pilot study with peroxisome proliferator-activated receptor (PPAR) delta agonist MBX-8025 for homozygous familial hypercholesterolemia (HoFH), Cymabay Therapeutics Inc. has begun exploring the prospect of combining it with a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor to make the drug work even better.

"To the best of our knowledge, nobody has ever connected the regulation of PCSK9 with the PPAR delta mechanism," CEO Harold Van Wart said during a conference call with analysts to discuss the top-line phase II results, which showed a wide variability of response despite the small size, including a PSCK9 uptick. "I would have to characterize this as a new finding. It's very fortunate that we had the foresight to measure PCSK9 levels," he added, pointing to "a reasonable hypothesis that if you could prevent the increase in PCSK9, which may be offsetting the LDL cholesterol lowering, [then research deploying] a concomitant PCSK9 inhibitor treatment could enhance" the sought result and make the counts go down farther. In any case, Wall Street wasn't thrilled about the outcome with MBX-8025 by itself. Shares of Newark, Calif.-based Cymabay (NASDAQ:CBAY) closed Friday at $1.29, down 41 cents, or 24 percent.

The study was an open label, dose-escalation effort lasting 12 weeks, conducted at five centers in Europe and Canada. All patients had genetically confirmed HoFH, including with functionally negative mutations in their LDL receptor (LDL-R) genes, and were taking ezetimibe (Zetia, Merck & Co. Inc.) plus maximum statin therapy. None received lomitapide (Juxtapid, Aegerion Pharmaceuticals Inc.), mipomersen (Kynamro, Sanofi SA), or a PCSK9 inhibitor. Eight were undergoing apheresis on a weekly or biweekly schedule. Despite being on maximal conventional therapy, the subjects tallied an average baseline LDL-C was 368 mg/dL.

Patients received once-daily 50 mg of MBX-8025 for four weeks, after which the dose was escalated to 100 and 200 mg in successive four-week periods, in order to evaluate the effect on LDL-C as well as a spectrum of other lipid-related parameters, including PCSK9 levels, and collect safety information.

Two per-protocol analyses were performed on 12 subjects. (The data for one subject was excluded because of multiple missed apheresis visits, causing LDL-C fluctuations.) A responder analysis was carried out which reflects the largest decrease in LDL-C found during treatment for each of those remaining. Three (25 percent) showed a greater than or equal to 30 percent decrease. Five (42 percent) had a greater than or equal to 20 percent decrease, including one patient that was receptor-negative, and seven (58 percent) had a greater than or equal to 15 percent decrease. Five (42 percent) had a less than 15 percent decrease, and the average maximum decrease in the study was 19 percent.

Because of the high baseline LDL-C levels in those enrolled, the percentage decreases correspond to significant absolute decreases in LDL-C (mean decrease of 109 mg/dL for the subjects with a greater than or equal to 15 percent decrease), the company noted, conceding that although reductions in LDL-C tended to be greater at the higher doses, no clear dose response was observed.

In a second analysis, the mean change in LDL-C for each subject was measured by averaging values across all doses and dosing periods while on treatment. The overall mean change for all 12 subjects was a decrease of 10 percent. Eight of these subjects had a mean decrease in LDL-C of 16 percent, including three with greater than 20 percent, with one receptor-negative patient in the mix. The result was offset by four patients who turned up a mean increase of 4 percent.

Mean PCSK9 was elevated at baseline (544 +/- 133 ng/mL), about what's expected in patients with HoFH, and increased significantly during treatment by a mean of 43 percent. Also noted were decreases in mean levels of alkaline phosphatase (30 percent), gamma glutamyl transferase (27 percent) and total bilirubin (22 percent) – all markers of cholestasis. Three severe adverse events were recorded, none drug-related, and three treatment dropouts for adverse events that may have been due to MBX-8025.


Also on the call was Evan Stein, consultant to Cymabay and one of the chief researchers with Repatha (evolocumab, Amgen Inc.), the world's first approved PCSK9 inhibitor. "If you had a PCSK9 monoclonal antibody [involved in a future trial], what you will do is neutralize the effect of the increased PCSK9, and that would help determine whether we would see additional LDL reduction on top of what we're seeing in this study," he said. "Another 10 percent could make this drug – if we could get an average of 20 percent – a viable option in these patients."

The wide differences in response, especially in homozygous patients "goes back almost 25 years" to statins, Stein said. "With statins, that's been related to their residual LDL receptor activity [and] with PCSK9 as well. In the bigger studies we did with PCSK9, we documented that patients who had a defective rather than a completely absent LDL receptor had a much better response. Those with no LDL receptors [had] essentially no response. So you see a broad range, even with drugs which don't work through the LDL receptor but work by inhibiting synthesis and release of LDL."

Another factor is trial size, Stein said. "In the first pilot study we did with [Repatha], it was an eight-patient homozygous [trial and] we had two complete non-responders who had LDL receptor-negative status, and six good responders. The average reduction was about 16-19 percent. When the big phase III trial was done with more patients [who had] a lot more LDL receptor defects that cause homozygous FH, compared to placebo the reduction was 31 percent." The results " obviously led to FDA approval, putting the drug between mipomersen and lomitapide in terms of response," he said.

CEO Van Wart estimated that an experiment pairing a PCSK9 inhibitor with MBX-8025 wouldn't strain the budget. "I'd like to highlight that this current study cost less than $2 million," he said. "We believe we can carry out that [PCSK9] study for certainly less, probably at one or at most a small number of centers. This is not a big resource issue, and I do believe that it will give us the kind of information we need to decide whether or not the size of the effect and the broadening of the response would be sufficient to continue development, go talk to the FDA about what a phase III trial might look like, and move the ball down the field."

Piper Jaffray analyst Edward Tenthoff said that, "despite some disappointment with the magnitude of LDL-C lowering, we believe MBX-8025 is active. More importantly, the drug continues to show a beneficial hepatic profile, improving three metabolic parameters, which are key outcomes of the ongoing phase II in primary biliary cholangitis," with data expected in the second half of this year. He maintained an "overweight" rating on the company with a $4 price target.

Tenthoff said a partnership for phase III trials with arhalofenate in gout may stimulate the shares. The FDA has agreed to a pivotal program that includes two trials with chronic gout patients and one trial in tophaceous gout, characterized by masses of uric acid crystals, or tophi, accumulating in soft tissues. Co-primary endpoints are the percent of patients achieving serum uric acid of less than 6mg/dL in chronic gout or less than 5mg/dL in tophaceous gout, and flare rate following 12 months of therapy. "Importantly, arhalofenate will not have to achieve superiority over colchicine [Colcrys, Takeda Pharmaceutical Co. Ltd., but also available in generic form] in flare control," he wrote in a research report. A total of 1,300 patients will be enrolled across the trials, which the company said will go ahead once a partner is signed. Although London-based Astrazeneca plc's gout drug Zurampic (lesinurad) beat arhalofenate to market, the compound is hobbled by renal safety concerns that led to an FDA black box warning on the label. (See BioWorld Today, Dec. 24, 2015, and Jan. 21, 2016.)

No Comments