The p53 tumor suppressor gene has been found to be altered in almost all human solid tumors, whereas K-rasgene mutations have been observed in a limited number of human cancers (adenocarcinoma of colon, pancreas, and lung). Studies of mutational inactivation for both genes in the same patient's sample on non-small-cell lung cancer have been limited. In an effort to perform such an analysis, we developed and compared methods (for the mutational detection of p53 and K-rasgene) that represent a modified and universal protocol, in terms of DNA extraction, polymerase chain reaction (PCR) amplification, and nonradioisotopic PCR-single-strand conformation polymorphism (PCR-SSCP) analysis, which is readily applicable to either formalin-fixed, paraffin-embedded tissues or frozen tumor specimens. We applied this method to the evaluation of p53 (exons 5–8) and K-ras(codon 12 and 13) gene mutations in 55 cases of non-small-cell lung cancer. The mutational status in the p53 gene was evaluated by radioisotopic PCR-SSCP and compared with PCR-SSCP utilizing our standardized non-radioisotopic detection system using a single 6-μm tissue section. The mutational patterns observed by PCR-SSCP were subsequently confirmed by PCR-DNA sequencing. The mutational status in the K-rasgene was similarly evaluated by PCR-SSCP, and the specific mutation was confirmed by Southern slot-blot hybridization using32P-labeled sequence-specific oligonucleotide probes for codons 12 and 13. Mutational changes in K-ras(codon 12) were found in 10 of 55 (18%) of non-small-cell lung cancers. Whereas adenocarcinoma showed K-rasmutation in 33% of the cases at codon 12, only one mutation was found at codon 13. As expected, squamous cell carcinoma samples (25 cases) did not show K-rasmutations. Mutations at exons 5–8 of the p53 gene were documented in 19 of 55 (34.5%) cases. Ten of the 19 mutations were single nucleotide point mutations, leading to amino acid substitution. Six showed insertional mutation, and three showed deletion mutations. Only three samples showed mutations of both K-rasand p53 genes. We conclude that although K-rasand p53 gene mutations are frequent in non-small-cell lung cancer, mutations of both genes in the same patient's samples are not common. We also conclude that this universal nonradioisotopic method is superior to other similar methods and is readily applicable to the rapid screening of large numbers of formalin-fixed, paraffin-embedded or frozen samples for the mutational analysis of multiple genes.