The validity of molecular studies using DNA and RNA extracted from decades-old formalin-fixed and paraffin-embedded tissue blocks has been demonstrated. The quality and usability of DNA and RNA from archival tissues are modified by various factors, such as the fixative, the fixation time, and the postmortem time. However, in contrast to DNA, there are no comprehensive studies quantitatively addressing the feasibility of RNA from old (more than 10 years) archival samples. This study examined the integrity of RNA extracted from 738 autopsy liver and 63 autopsy thyroid cancer tissue blocks procured during a span of nearly four decades, beginning in 1952 and ending in 1989, from the atomic bomb survivors. The integrity of RNA was assessed by amplification of c-BCR messenger RNA (mRNA) between two sequential exons with an intervening intron by reverse-transcription polymerase chain reaction (RT-PCR). The integrity of RNA was influenced by the age of the samples and the postmortem time, but not by the formalin-fixation period. It was possible to amplify more than 60% of the samples. Using these RNAs, the HCV genome in liver cancers and the H4-RET gene in thyroid cancers were detectable. This study illustrates the possibility of molecular studies using RNA from routinely prepared paraffin blocks stored for long periods and provides the statistics and critical factors to consider in assessing the feasibility of such contemplated studies.