A morphometric expression of differentiation in fine‐needle biopsies of breast cancer

Abstract

An automated method of particle sizing in fine-needle biopsies of human breast carcinomas is described. The average sue of clusters of cells aspirated from breast carcinomas was strongly correlated with cytologic differentiation of the tumors and also with the recurrence rate 3 years after mastectomy so that smaller cluster size w a s associated with poor differentiation and a poor prognostic outcome. It thus seems possible to express in quantitative terms the degree of differentiation of breast carcinomas. Key terms: Breast cancer, differentiation, morphometryIn a leading article in Lancet 1960 (3) the author concludes that "pathetically little is known about the biophysical bases of (cell) adhesion." Although knowledge has increased the last 20 years, there are still many unclear factors involved in elucidating how cells adhere to one another. Local invasion of cancer has been claimed to depend on the cohesiveness of tumor cells (5). Not only are cancer cells less cohesive than normal cells but also there seems to be variability in the tion of the tumor border to the surrounding fat (9). It is confusing that differentiation is considered together with these factors since it is not known to what extent they contribute independently to the prognostic outcome. This is further underlined in a recent truncated component analysis (17) showing that they do not seem to describe the same biologic phenomenon. In addition, in one study (4) only one expression of differentiation-tubule formation is evaluated, cohesiveness of cell populations of different tumors obtained from the same organ. In fine-needle biopsies from breast cancer this is reflected in varying degrees of dissociation of tumor cells. The smaller the clusters, the less differentiated the tumor is thought to be. In addition, in a study of the prognostic value of the aspiration biopsy smears using AIDanalysis, Wallgren and Zajicek (19) noted that the cluster size (more than or less than 10 cells/cluster) was significant in predicting 10-year survival.In some studies of prognostic factors in breast cancer various histopathologic expressions of degree of differentiation are correlated with 5-year survival or 3-year recurrence rate (4, 9, 15)' although differentiation in most of these studies is evaluated in combination with other factors-nuclear atypia, frequency of mitoses or hyperchromatic nuclei (4, 14) or degree of lymphocytic infiltration in the tumor and the rela-I Supported by NIH/NCI Contract NOlCB53968. 'Stenkvist B, Bengtsson E, Eriksson 0, Nordin B, Jarkrans T, Westman-Naeser S: Submitted for publication.an unexplained restriction, since acinus, lobule or duct formation also determine differentiation.We have recently published a study of computerized nuclear morphometry of cell nuclei from breast cancer (16). In looking for a method to quantitate morphometrically the degree of differentiation of cancer we tried a particle suing experiment described in this study. Material and MethodsThe material analyzed in this study was part of a study of 179 wome...

Biological meaning of stage and grade in human breast cancer: Review and hypothesis

“…Reproducibility among several observers was extremely high. Other investigators have reported similar results using morphometric methods (107,108). We believe that such quantitative methods of grading tumors will assume increasing importance in the future because of their precision and reproducibility.…”

Reduced expression of Claudin‐7 in fine needle aspirates from breast carcinomas correlate with grading and metastatic disease

“…In clinical cytology, cellular dissociation, with intact, single lying neoplastic cells is one of a number of features associated with malignancy. Previous studies have demonstrated that regional and distant metastases as well as local recurrence correlate with decreased cellular cohesion 15–18 . A dysfunctional, intercellular adhesion system may be responsible for the tumour cell dispersion pattern seen on fine needle aspiration cytology (FNAC).…”

Image analysis of irregularity of cluster shape in cytological diagnosis of breast tumors: Cluster analysis with 2D‐fractal dimension

“…11 The overall size of the cluster has previously been suggested to be a factor influencing prognosis. 14 In that study, malignancy correlated with the complexity in cyto configuration, in which the overall cell shape was related to the strength of intercellular cohesion. Important factors indicative of malignancy, such as cell overlap, cell adhesion, and irregular nucleus shape, are clearly expressed in this present study, according to the observed cluster shape.…”