Creatine has been tested over long time spans already so this comment is not applicable here.

The better rebuttal is that the paper didn’t follow RCT protocols.