Reliable identification of cancer-related mutations in TP53 is often problematic, as these mutations can be randomly distributed throughout numerous codons and their relative abundance in clinical samples can fall below the sensitivity limits of conventional sequencing. To ensure the highest sensitivity in mutation detection, we adapted the recently described coamplification at lower denaturation temperature-PCR (COLD-PCR) method to employ two consecutive rounds of COLD-PCR followed by Sanger sequencing. Using this highly sensitive approach we screened 48 nonmicrodissected lung adenocarcinoma samples for TP53 mutations. Twenty-four missense/frameshift TP53 mutations throughout exons 5 to 8 were identified in 23 out of 48 (48%) lung adenocarcinoma samples examined, including eight low-level mutations at an abundance of approximately 1 to 17%, most of which would have been missed using conventional methodologies. The identified alterations include two rare lung adenocarcinoma mutations, one of which is a "disruptive" mutation currently undocumented in the lung cancer mutation databases. A sample harboring a low-level mutation ( approximately 2% abundance) concurrently with a clonal mutation (80% abundance) revealed intratumoral TP53 mutation heterogeneity. The ability to identify and sequence low-level mutations in the absence of elaborate microdissection, via COLD-PCR-based Sanger sequencing, provides a platform for accurate mutation profiling in clinical specimens and the use of TP53 as a prognostic/predictive biomarker, evaluation of cancer risk, recurrence, and further understanding of cancer biology.