This article describes several approaches to a selective therapy of virus infections: (E)-5-(2-bromovinyl)-2'-deoxyuridine (BVDU [brivudin]) for the therapy of herpes simplex virus type 1 and varicella-zoster virus infections: (S)-9-(3-hydroxy-2-phosphonylmethoxypropyl)cytosine (HPMPC [cidofovir]) for the therapy of various DNA virus (i.e., herpesvirus, adenovirus, papillomavirus, polyomavirus, and poxvirus) infections; 9-(2-phosphonylmethoxyethyl)adenine (PMEA [adefovir]) for the therapy of retrovirus, hepadnavirus, and herpesvirus infections; (R)-9-(2-phosphonylmethoxypropyl)adenine (PMPA) for the therapy and prophylaxis of retrovirus and hepadnavirus infections; and nonnucleoside reverse transcriptase inhibitors (NNRTIs), such as tetrahydroimidazo[4,5,1-jk][1,4]-benzodiazepin-2(IH)-one and -thione (TIBO), 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT), alpha-anilinophenylacetamide (alpha-APA), and 2',5'bis-O-(tert-butyldimethylsilyl)-3'-spiro-5"-(4"-amino-1",2"-oxat hiole- 2",2"-dioxide)pyrimidine (TSAO) derivatives, and thiocarboxanilides for the treatment of human immunodeficiency virus type 1 (HIV-1) infections. For the clinical use of NNRTIs, some guidelines have been elaborated, such as starting treatment with combinations of different compounds at sufficiently high concentrations to effect a pronounced and sustained suppression of the virus. Despite the diversity of the compounds described here and the different viruses at which they are targeted, they have a number of characteristics in common. As they interact with specific viral proteins, the compounds achieve a selective inhibition of the replication of the virus, which, in turn, should be able to develop resistance to the compounds. However, as has been established for the NNRTIs, the problem of viral resistance may be overcome if the compounds are used from the start at sufficiently high doses, which could be reduced if different compounds are combined. For HIV infections, drug treatment regimens should be aimed at reducing the viral load to such an extent that the risk for progression to AIDS will be minimized, if not avoided entirely. This may result in a real "cure" of the disease but not necessarily of the virus infection, and in this sense, HIV disease may be reduced to a dormant infection, reminiscent of the latent herpesvirus infections. Should virus replication resume after a certain time, the armamentarium of effective anti-HIV and anti-herpesvirus compounds now available, if applied at the appropriate dosage regimens, should make the virus return to its dormant state before it has any chance to damage the host. It is unlikely that this strategy would eradicate the virus and thus "cure" the viral infection, but it definitely qualifies as a cure of the disease.