Epidemiological studies indicate that benzene exposure is associated with an increased risk of human leukemia and lymphoma ( 1 , 2 ). Chromosomal translocations are thought to be initiating events in leukemia and lymphoma, and specific translocations are used as markers of diagnosis, disease progression, and relapse. As benzene is thought to act by producing chromosomal aberrations and altered cell differentiation ( 3 ), we investigated the possibility of using circulating levels of two acute myeloid leukemia (AML) -associated translocations, t(8;21) ( 4 ) and t(15;17) ( 5 ), and a follicular lymphoma -associated translocation, t(14;18) ( 6 ), as biomarkers of early effect for benzene. Evidence for the occurrence of these translocations in normal individuals is limited, with t(8;21) detected in 1 in 496 cord blood samples ( 7 ) and t(15;17) detected in healthy volunteers ( 8 ). As recently reviewed ( 9 ), the prevalence of t(14;18) in peripheral blood lymphocytes of healthy donors is relatively high at ~ 50% (range 8% -80% depending on the polymerase chain reaction (PCR) methodology employed and target cell population) ( 9 -13 ) and rises with smoking ( 14 ), but the correlation between frequency and age remains unclear ( 13 , 15 ). The biological consequences of these aberrations in healthy individuals are yet to be understood. One possibility is that the frequency of translocationpositive cells represents a biomarker of early effect for hematopoietic carcinogens which could correlate with the cumulative risk of developing t(14;18)-related follicular lymphoma and t(8;21)-and t(15;17)-related AML, and perhaps other cancers.The two principal technologies for detection of specifi c chromosome translocations are fl uorescence in situ hybridization (FISH) and real-time PCR. FISH distinguishes individual chromosomes by hybridization with sequence-specifi c, chromosomepainting probes labeled with spectrally nonoverlapping fl uorophores. In real-time quantitative PCR (qPCR), a fl uorescently labeled probe is hybridized to the amplifi cation target, and fl uorescent signal proportional to the input DNA/cDNA is generated during amplifi cation, thereby allowing quantifi cation. qPCR assays the whole peripheral blood mononuclear cell (PBMC) population in G 0 , whereas FISH examines a culture-modifi ed popula tion [consisting of mainly mature T lymphocytes stimulated by the mitogen phytohemagglutinin that have gone through two cell divi sions ( 16 )]. The two methods also generate different estimates of t(14;18) translocation levels with FISH detecting all possible t(14;18) breakpoints and the real-time assay described here only detecting breakpoints occurring in the major breakpoint region (MBR) of the BCL-2 gene. Although it was previously thought that the majority of BCL-2 breakpoints occur in the MBR ( 17 ), it has been shown more recently that many breakpoints occur outside of the MBR and the minor cluster region ( 18 ). The ability of FISH to detect more translocation breakpoints is offset by its lower relative sensit...