Eli Lilly’s tease that it wasn’t yet done with its experimental Alzheimer’s medication solanezumab has moved forward: the drug maker announced Wednesday that it was giving the treatment another go at Phase III trials.

The difference this time around is that the target audience is patients with mild Alzheimer’s. Previous tests were among mild-to-moderate patients, and failed to trigger results on both cognitive and global function targets.

Lilly is discussing the design and length of the new study with the FDA, but expects to kicks things off by the third quarter of next year.

Bernstein analyst Tim Anderson wrote in his Wednesday research note that although Lilly has told his group they don’t want to up the solanezumab dose, he thinks it would make sense to change the rules for this next round of testing, in part, because he attributed Pfizer’s bapineuzumab failure to dosing, writing the drug “likely failed in part because only very low dosing was possible.” Anderson wrote Wednesday that Lilly used 17 times the amount of solanezumab in its Phase III trials than Pfizer did of bapineuzumab. Lilly said in the fall that it was able to tease out some positive results, such as slowed cognitive decline, even though it didn’t hit the trial’s goals. Anderson speculated Wednesday that the higher-than-babineuzumab dose “may explain why modest efficacy was seen” in Lilly’s trials.

Hopes were never high for either of these drugs, but a December 4 research note by Anderson asks if Lilly’s Alzheimer’s pipeline isn’t already in a good place, for reasons having nothing to do with solanezumab. Instead, Anderson’s note compares Lilly’s nascent BACE inhibitor, LY2886721, which is already in Phase II trials, with Merck’s MK-8931, which only just started its Phase II trials. Both are oral, once-a-day medications that would seek to reduce the accumulation of beta-amyloid plaques which are associated with the disease, but little is known about either of these experimental therapies.