Archival cervical smears: a versatile resource for molecular investigations Archival cervical smears represent a huge resource of pathological specimens. This, together with long clinical follow-up data, makes archival smears the most valuable resource for cervical cancer research. Despite this huge potential, only a few molecular investigations have been carried out based on archival smears. It has been shown that archival smears can be used for amplification of genomic sequences by polymerase chain reaction (PCR). However, it is unknown whether PCR can be applied to minute dyskaryotic cells microdissected from archival cervical smears and whether these archival materials are suitable for reverse transcription PCR (RT-PCR). To address these issues, we prepared DNA and RNA samples from dyskaryotic cells microdissected from archival cervical smears with a storage time of 11 years and systematically tested the extent that these materials can be used for PCR-based molecular investigations at both DNA and RNA levels. Our results showed that a crude DNA preparation simply by proteinase K digestion was suitable for PCR amplification of genomic sequences. By targeting the amplified genomic sequence to 250 bp or less, most if not all archival smears could be used for PCR and are therefore suitable for screening gene mutations and loss of heterozygosity, human papillomavirus typing, etc. Purified DNA samples from microdissected dyskaryotic cells were adequate for restriction enzyme digestion and could be used for a PCR-based clonality analysis of the androgen receptor gene. Finally, RNA samples extracted from dyskaryotic cells microdissected from archival smears were adequate for RT-PCR as long as a gene-specific primer was used for the RT reaction and the target sequence was restricted to 150 bp or less. In summary, our results demonstrated that archival cervical smears are suitable for a range of molecular investigations at both DNA and RNA levels. The potential gain of knowledge on cervical cancer by the molecular study of archival smears is immense.