Human carbonic anhydrase (CA; EC 18.104.22.168) isoforms II and VII are implicated in neuronal excitation, seizures and neuropathic pain (NP). Their selective inhibition over off-target CAs is expected to produce an anti-NP action devoid of side effects due to promiscuous CA modulation. Here a drug-design strategy based on the observation of (dis)similarities between the target CA active sites was planned with benzenesulfonamide derivatives and, for the first time, a phosphorus-based linker. Potent and selective CA II-VII inhibitors were identified among the synthesized phenyl(thio)phosphon(amid)ates 3-22. X-ray crystallography depicted the binding mode of phosphonic acid 3 to both CA II and VII. The most promising derivatives, after evaluation of their stability in acidic media, were tested in a mouse model of oxaliplatin-induced neuropathy. The most potent compound racemic mixture was subjected to HPLC enantioseparation and the identification of the eutomer, the (S)-enantiomer, allowed to halve the dose totally relieving allodynia in mice.