There are likely to be small changes in several areas, but whether you will be able to hear any difference is debatable.

Firstly, the use of a digital connection instead of an RCA cable will remove any attenuation and distortion in the connection between Port and each LS50.

Secondly, you will be using a different DAC - the one in each LS50 rather than the one in the Port, and this may have a different characteristic sound.

Thirdly, you may (depending on how you implement a volume control) introduce or remove distortion when you change the volume (because as I understand it, you could either change the volume in the digital domain or in the analogue domain). Digital volume controls can affect the levels of quantisation distortion, while analogue volume controls do not (though they may also introduce other forms of distortion instead).

There are probably other factors that I haven’t thought of, but I suspect that most will be undetectable in practice. The one thing that is likely (IMO) to make the most difference is the “characteristic sound” of the KEF DAC versus the Sonos DAC. That’s because there are aspects of DAC design that can be implemented in different ways  - there is no “correct” way, and each is likely to sound very slightly different.

As in the case of stereo amps, it makes the most sense for DAC makers to have their DAC to be as transparent as possible; any house sound is better taken care of by user operated filters/tone controls, so as to not limit the market to those that prefer the house sound.

IMO, the bigger problem may be the converter that will be needed. As one more element in the signal path, at best, it will not degrade it.

I agree that the most likely outcome is that nothing different will be heard.

very helpful explanations, thank you very much.

You will also bypass the ADC of the KEFs, which may slightly improve the noise floor, but probably not enough to notice.