If the effect is not stable and at least some patients tend to return to insulin injection or need to repeat the procedure annually I'd recommend the researcher

If the effect is not stable and at least some patients tend to return to insulin injection or need to repeat the procedure annually I'd recommend the researcher to try supplementing α-lipoic acid, sodium butyrate and, perhaps, inulin (not insulin). I believe this may help (or may not, would be cool to check with some sort of clinical trials).