Despite the widespread use of liposome-mediated gene transfer in cancer therapy protocols, little is known about the tissue distribution of intralesionally administered DNA. We have previously shown that antisense gene therapy targeting the epidermal growth factor receptor (EGFR) inhibited tumor growth in a human head and neck squamous cell carcinoma (HNSCC) xenograft model. Further investigation demonstrated lack of systemic toxicity with intramuscular or intratumoral administration of this liposomal-DNA complex. In the present study, we compared two approaches to determine the presence of exogenous DNA in the plasma and tissues of mice treated with intramuscular injection of EGFR antisense gene therapy. PCR analysis using genomic DNA plus plasmid DNA as template was 83-fold more sensitive than PCR using a mixture of total RNA and plasmid DNA as template. With the more sensitive method (able to detect fewer than 500 molecules of EGFR antisense DNA in 1 mg of genomic DNA), foreign DNA was detected in all organs up to 1 month following a single injection. In contrast, using RNA plus plasmid DNA as template, exogenous DNA was only detected at the injection site at 1 week, and was undetectable at 1 month. Optical imaging studies demonstrated plasmid DNA only at the injection site. Although less sensitive than PCCR, Southern blot hybridization showed no evidence of integration of foreign DNA into the host genome in vitro or in vivo. These results emphasize the importance of defining the assays used to detect foreign DNA and suggest that the ability to detect intralesionally administered liposomal gene therapy, in organs distant from the injection site, is directly correlated with the sensitivity of the method employed.