Your argument overlooks dosing differences between the various IL-1 drugs and the various trials!
1) The point here is that the FDA is ultra sensitive. At the very least for an indication like acne they will need a very large database (i.e. expensive and time consuming) and almost any excess risk is likely to trigger rejection (i.e. risky since, given its MOA, it almost certainly does create some excess infections - just a question of what rate.)
2) Anti-TNF data says that the therapuetic window (infections vs efficacy) can be optimized but not perfected. Some excess infections detectable in trials of several 1000 patient years are likely unless you give up on efficacy.
All told, if you have a drug that you know is likely to be reasonably suspected of having infection risk... it is odd to pursue an indication near the 'cosmetic' end of the spectrum.