DNA ploidy and S-phase fraction by image and flow cytometry in breast cancer fine-needle cytopunctures

The aim of this study was to achieve a better definition of the respective indications of flow cytometry (FCM) and image cytometry (ICM) in clinical practice by comparing their efficiency in a series of 104 primary breast carcinomas that were sampled by means of fine-needle cytopuncture and analyzed by both methods. The comparison involved the DNA content and the estimation of S-phase fraction (SPF). For this purpose, a manual rectangular model was used for ICM, and Modfit software was used for FCM. With respect to DNA ploidy, the concordance rate between the two methods was 87% (DNA diploid vs. DNA nondiploid) and 81% when subclasses of DNA ploidy were used (hypodiploid, diploid, hyperdiploid, tetraploid, or multiploid). True discordance was observed in 19 cases (18%). Eleven of these discordances were the result of an underestimation of multiploidy by FCM. In the 91 cases evaluable by ICM, the SPF median value was 3.5 for DNA-diploid tumors and 8 for DNA-nondiploid tumors. In the 67 cases evaluable by FCM, the SPF median value was 2.31 for DNA-diploid tumors and 5.25 for DNA-nondiploid tumors. The concordance between ICM and FCM in the 46 uniploid tumors was 0.90. Our results led us to obtain samples for the two techniques systematically, to perform FCM routinely because it is more rapid and adequate in a large number of cases, to confirm by ICM any DNA-diploid or DNA near-diploid peak revealed by FCM, and to use ICM when FCM is inadequate, i.e., in case of low cellularity (fewer than 5000 cells for analysis of the cell cycle), suspected tetraploidy, samples not exclusively composed of malignant cells, the identification of several morphologically different malignant populations, or populations with large coefficients of variation and high background.