Synlogic scraps ammonia-lowering drug after early phase fail

Headshot of Aoife Brennan, CEO of Synlogic
Synlogic CEO Aoife Brennan (Synlogic)

Synlogic has stopped development of SYNB1020 after the drug failed to lower blood ammonia levels in a phase 1b/2a trial. The engineered stain of Escherichia coli performed no better than placebo in the small trial, leading Synlogic to conclude its money is better spent on other assets.

Synthetic biology startup Synlogic designed SYNB1020 to consume ammonia in the colon, the main source of the compound in the body. SYNB1020 came through a trial in healthy volunteers, giving Synlogic increased confidence in the safety and tolerability of the drug and its broader pipeline of engineered microbes. But the candidate has come unstuck at the first test of its efficacy.

In a trial of 23 patients with cirrhosis and elevated blood ammonia, Synlogic found “no evidence of blood ammonia lowering or changes in other exploratory endpoints relative to placebo,” despite an increase in plasma and urinary nitrate suggesting the therapeutic strain was active.

Synlogic is holding off on the release of detailed results until a future scientific or medical conference, but the limited information available, plus the company’s response to the results, suggests the drug totally failed to move the needle. The sole positive highlighted in the release is the fact the drug was well tolerated.

The question now is whether the lack of efficacy seen in the trial is limited to SYNB1020 or is an early sign of a weakness of Synlogic’s core concept. Synlogic has data linking another asset, SYNB1618, to a statistically significant increase in biomarkers related to its ability to consume phenylalanine. But, as SYNB1020 was among its most advanced candidates, Synlogic has little other clinical efficacy data.

Synlogic plans to generate more evidence on its pipeline programs by funneling money formerly earmarked for SYNB1020 into the development of SYNB1618 and SYNB1891, a treatment for solid tumors that is moving toward the clinic. With almost $150 million in the bank at the last count, Synlogic has the means to find out whether the candidates work better than SYNB1020.