Corbus reports 5 percent weight loss for its peripheral CB1 pill in Phase 1b

Corbus Pharmaceuticals reported that its experimental oral pill CRB-913 produced an average of 5.0 percent weight loss after 12 weeks at the highest dose tested, compared with 0.0 percent for people taking a placebo, in a Phase 1b study called CANYON-1 ^[1]. The company plans to share fuller results at a late-breaking session at ObesityWeek on November 14, 2026, and aims to start a Phase 2 study in the first half of 2027 ^[1][2].

CRB-913 works differently than semaglutide (Ozempic, Wegovy, Rybelsus) or tirzepatide (Mounjaro, Zepbound). Instead of mimicking gut hormones like GLP-1, it blocks a receptor called CB1, part of the body's endocannabinoid system, which is involved in appetite and metabolism ^[1]. Corbus designed the drug to act mainly outside the brain, since earlier CB1 blockers were pulled from the market years ago after being linked to depression and other psychiatric problems ^[1][2].

The trial enrolled 254 adults with obesity who did not have diabetes, at 15 U.S. sites, and ran for 16 weeks total, including 12 weeks of dosing and 4 weeks of follow-up ^[1]. Participants were randomly assigned to placebo or to one of three CRB-913 doses -- 20 mg, 40 mg, or 60 mg once daily -- all starting at 20 mg and increasing every two weeks ^[1]. At the 60 mg dose, weight loss reached 5.0 percent versus 0.0 percent on placebo, a statistically significant difference (p<0.0001), and the lower doses also beat placebo, at 2.8 percent for 20 mg and 3.3 percent for 40 mg ^[1]. Among people who finished the study on the 60 mg dose, all lost some weight, 44.4 percent lost at least 5 percent of their body weight, and the largest individual loss recorded was 13.4 percent ^[1]. The company said weight loss had not leveled off by week 12 at any dose, suggesting further loss might occur with longer treatment ^[1].

On safety, Corbus said CRB-913 was "generally safe and well tolerated," with study dropout rates due to side effects (3.1 to 13.1 percent) similar to those seen in published data for approved oral GLP-1 drugs (6.9 to 20.7 percent) and notably lower than a rival CB1 drug called monlunabant (13 to 42 percent) ^[1]. A separate report on the announcement said psychiatric side effects were "broadly in line" with GLP-1 drugs and that the treatment appeared to avoid the depressive effects tied to older, brain-penetrating CB1 medicines, though specific numbers for psychiatric events were not disclosed in the sources reviewed ^[2].

Why it matters for patients

For people who cannot tolerate GLP-1 drugs' gastrointestinal side effects, or whose weight loss stalls on existing therapies, a working pill with a different mechanism could someday be another option. RTTNews noted that more than 60 percent of patients discontinue GLP-1 treatment within the first year, citing the ongoing need for alternatives ^[2]. Corbus's CEO said the results support the idea that a peripherally restricted CB1 drug could be a "safe, tolerable, and competitive oral alternative" to GLP-1 therapies, and the company is also exploring combining CRB-913 with GLP-1 treatment ^[2].

However, this is a small, early Phase 1b trial with 12 weeks of dosing, far short of the year or more of data regulators typically require, and the 5 percent weight loss reported here is modest compared with published results for approved GLP-1 and dual-agonist drugs. Because CB1 drugs as a class were withdrawn in the past over psychiatric safety concerns, longer and larger studies will be needed to confirm this newer, more peripherally targeted version remains safe over time ^[1][2].

What happens next

Full CANYON-1 data will be presented at ObesityWeek on November 14, 2026 ^[2]. Corbus expects to begin a Phase 2 monotherapy study in the first half of 2027, and is also evaluating a CRB-913 plus GLP-1 combination approach ^[1][2]. Until Phase 2 and later trials are completed, it is not yet known whether CRB-913 will show similar effectiveness or safety in larger, longer studies, or whether it will reach approval.