Some therapies (antibiotics, eg) select for resistant bacterial or tumor populations. That is not true of the CFTR modulators Trikafta or Alyftrek, which work by helping the mutant chloride channel protein fold, getting it to the surface of the cell, and increasing the amount of time the channel stays open. There's no apparent theoretical advantage to the host cell to keep the modulator from entering the cell where it is needed.
That's not to say the modulators prevent all pulmonary exacerbations (PE) long term - they don't. But PE are greatly reduced. It is quite likely that the bacteria colonizing the lung adapt to the changing conditions resulting from a vastly different extracellular environment post-modulator treatment and can flare up, often in response to viral insults. However, data do show that the bacterial burden in the CF lung (the only population studied to date) remains orders of magnitude lower even after several years of effective modulator treatment.
I hope that helps.