I've tried a few de-essers, but seem to keep coming back to doing it using old-style side-chained compression.
Clone the offending vocal track, set the output to an aux channel. Eq that track to emphasise the sibilance, preferably so it contains as little content other than sibilance as possible. Then mute the aux.
Put a compressor on the "real" vocal track and side-chain it to the sibilance track, then adjust the compressor and sibilance track fader until the result is as good as it can be. A few more steps than doing it all in a single plugin, but allows the choice of any compressor that has a side-chain.
The even more old-fashioned 1930s-50s way of dealing with sibilants was for singers to change how they vocalised words that begin with an s, singing almost an "sh" sound instead, which puts the tongue at the bottom of the mouth not the top (where it is with an "s") and greatly reducing the whistling caused by directing air out of the top of the mouth and through the teeth.
Crooners seem to have suffered most from sibilance, maybe as a side-effect of the style, and although neother crooned, extreme use of the technique resulted in the shound that is often ashoshiated with shingersh like Jimmy Durante, or Richie Barrett's rendering of "Shome Other Guy".