Cancer is a genetic disease involving the sequential accumulation of somatic mutations. Genetic damage leading to these mutations can occur at the level of the gene, (e.g. point mutations) or the chromosome (e.g. aneuploidy, translocations). Numerous biomarkers have been developed to detect early mutational and chromosomal effects of carcinogenic exposure in humans. Historically, biomarkers have tended to measure mutations in surrogate genes, such as HPRT, or use cytogenetics to assess chromosomal aberrations (CA). These biomarkers are elevated by a wide range of carcinogenic exposures and CA have been shown to be predictive of future cancer risk in prospective epidemiology studies. They do, however, have several drawbacks in that surrogate genes are not on the causal pathway of disease and cytogenetic markers require metaphase spreads to be prepared and their analysis is time consuming and expensive. New biomarkers of early effect for cancer are therefore urgently needed. We propose to develop biomarkers of early effect for blood cancers by generating high-throughput single cell PCR assays for mutation in key genes or regions linked to leukemia and lymphoma. This approach will take advantage of recent advances in microfluidics and lab-on-a-chip technologies that make single cell genetic analysis (SCGA) feasible on a high-throughput scale. Specifically we plan to develop methods for performing high throughput single template copy PCR amplification in nanoliter emulsion droplets and then apply these methods to develop high-throughput SCGA methods to detect low frequencies of mutations of importance in blood cancers, such as translocation t(14;18), deletion of 5q31 and mutations in NFtAS and NPM1. We will test if the new SCGA methods can detect mutations in cells exposed to alkylating agents and benzene metabolites and humans exposed to benzene. The studies proposed will be a first step towards providing revolutionary new assays for the detection of genetic mutations of relevance to blood cancer risk in healthy individuals. Such biomarkers will be useful in epidemiological studies of leukemia and lymphoma, which have long latency periods, as well as providing tools for early detection for those individuals at risk. The high-throughput detection methods we propose to develop should be applicable to large human population studies and will work with existing biobanked specimens