Johnson & Johnson (JNJ) reported data Monday night from cardiovascular trials of its diabetes treatment Invokana, showing arguably mixed results as the drug lowered disease risk while increasing the risk of amputations, an outcome which Wall Street said could lift similar but safer drugs in the space.

INVOKANA DATA: Last night, Johnson & Johnson reported results from the “CANVAS” trials of its already-approved SGLT-2 inhibitor Invokana, saying the drug significantly lowered risk of cardiovascular death, myocardial infarction, and nonfatal stroke in type 2 diabetes patients at risk or with history of heart disease, while also registering a potential kidney benefit. However, the two trials also showed nearly double the risk of amputation, with the company cautioning that those safety findings “have been shared by the U.S. FDA and will be reflected in the U.S. prescribing information.”

PLUS FOR LILLY, MINUS FOR MERCK: Credit Suisse analyst Vamil Divan argues that CANVAS nearly paints a best case scenario for Eli Lilly (LLY) and its Jardiance drug. Invokana validates the previous efficacy data for Jardiance as well as hospitalization rate and possible kidney protection, which “bodes well” for Lilly’s ongoing studies in those populations. Meanwhile, the increased amputation risk — while potentially outweighed by Invokana’s benefits — and consequent label warnings “will likely lead to a significant commercial advantage” for Eli Lilly, Divan contends. Conversely, Merck’s (MRK) Januvia product, which belongs to a different class of diabetes medicines, may face “incremental pressure” over time as enthusiasm around the CANVAS efficacy data likely pushes SGLT-2 inhibitors to the forefront of the treatment cycle.

COWEN POSITIVE: Keeping an Outperform rating and $143 target on Johnson & Johnson, Cowen analyst Joshua Jennings calls Invokana’s positive efficacy data “crucial” for its competitive positioning and supportive of his current forecasts, while also boding well for the larger SGLT-2 drug class. He remains “optimistic” that digestion of the data will yield a broad label, similar to Jardiance.