Abstract BACKGROUND: The identification of breakpoints involved in chromosomal damage could help to detect genes involved in genetic disorders, most notably cancer. Until now, only one published study, carried out by our group, has identified chromosome bands affected by exposure to oil from an oil spill. In that study, which was performed two years after the initial oil exposure in individuals who had participated in clean-up tasks following the wreck of the Prestige, three chromosomal bands (2q21, 3q27, 5q31) were found to be especially prone to breakage. A recent follow-up study, performed on the same individuals, revealed that the genotoxic damage had persisted six years after oil exposure. OBJECTIVES: To determine whether there exist chromosome bands which are especially prone to breakages and to know if there is some correlation with those detected in the previous study. In addition, to investigate if the DNA repair problems detected previously persist in the present study. DESIGN: Follow-up study performed six years after the Prestige oil spill. SETTING: Fishermen cooperatives in coastal villages. PARTICIPANTS: Fishermen highly exposed to oil spill who participated in previous genotoxic study six years after the oil. MEASUREMENTS: Chromosome damage in peripheral lymphocytes. For accurate identification of the breakpoints involved in chromosome damage of circulating lymphocytes, a sequential stain/G-banding technique was employed. To determine the most break-prone chromosome bands, two statistical methods, the Fragile Site Multinomial and the chi-square tests (where the bands were corrected by their length) were used. To compare the chromosome lesions, structural chromosome alterations and gaps/breaks between two groups of individuals we used the GEE test which takes into account a possible within-individual correlation. Dysfunctions in DNA repair mechanisms, expressed as chromosome damage, were assessed in cultures with aphidicolin by the GEE test. RESULTS: Cytogenetic analyses were performed in 47 exposed individuals. A total of 251 breakpoints in exposed individuals) were identified, showing a non-uniform distribution in the human ideogram. Ten chromosome bands were found to be especially prone to breakage through both statistical methods. By comparing these bands with those observed in certain exposed individuals who had already participated the previous study, it was found in both studies that four bands (2q21, 3q27, 5q31 and 17p11.2) are particularly sensitive to breakage. Additionally, the dysfunction in DNA repair mechanisms was not significantly higher in oil-exposed individuals than in non-exposed individuals. LIMITATIONS: The sample size and the possibility of some kind of selection bias should be considered. Genotoxic results cannot be extrapolated to the high number of individuals who participated occasionally in clean-up tasks. CONCLUSION: Our findings show the existence of at least four target bands (2q21, 3q27, 5q31 and 17p11.2) with a greater propensity to break over time after an acute exposure to oil. The breaks in these bands, which are commonly involved in hematological cancer, may explain the increase of cancer risk reported in chronically benzene-exposed individuals. In addition, a more efficiency of the DNA repair mechanisms has been detected six years after in fishermen who were highly exposed to the oil spill. To date, only this study, performed by our group on the previous and present genotoxic effects, has analyzed the chromosomal regions affected by breakage after an acute oil exposure.