Correlated flow-cytometric (FCM) and microspectrophotometric (MSP) techniques were applied to investigating whether intratumoral variations in the DNA distribution patterns of 21 primary mammary adenocarcinomas can occur. Although neoplastic cell populations with both diploid and tetraploid (i.e., euploid) distribution patterns could be found in varying proportions in some of the tumors, there was no evidence in any tumor nodule for the presence of euploid populations in one part and aneuploid populations in another. This statement was based on the results of the MSP technique, where the assessments were made on cytodiagnostically identified neoplastic cells. Also, when applying the FCM technique the statement was found to be essentially valid; only one of the tumor nodules showed a DNA distribution pat--tern that, by means of the criteria used in this procedure, was defined as being both euploid and aneuploid. Here, however, the technique consists of assessments made on a great number of microscopically non-identified cells.It was concluded that when conflicting reports are given from different laboratories on the prognostic value of the cytochemically assessed DNA distribution patterns in breast carcinomas, they are not likely to be attributed to intratumoral DNA heterogeneity but, rather, to differences in the methods used and in the criteria applied for the so-called ploidy assessments.Key terms: Mammary carcinoma, cytochemical DNA ploidy, intratumoral heterogeneity, flow-cytometric DNA assessment, microspectrophotometric DNA Several comprehensive investigations of cytochemically assessed nuclear DNA contents in mammary adenocarcinomas indicate that patients with tumors exhibiting euploid DNA distribution patterns (diploid and/or tetraploid) have a better prognosis than those with aneuploid patterns (1,2,9,12). These observations can have therapeutic implications. However, conflicting data also have been presented, and the prognostic value of DNA determination in breast cancer is not yet fully established (14,16,23).Of utmost importance for the validity of the results of such cytochemical malignancy gradings in clinical tumor material is whether the cell samples analyzed are actually representative for the whole tumor. In addition, the method used to obtain DNA histograms has to be taken into account, to adequately compare and interpret the results of different studies (11).The aim of the present study was to investigate whether intratumoral variations occur in the DNA distribution patterns in mammary carcinomas. This was done by comparing DNA histograms obtained in multiple biopsy specimens of the same tumor. In addition, the representativity of a randomly taken excisional surgical biopsy specimen versus a fine-needle aspirate (FNA) of the intact tumor was assessed using both flow-cytometric (FCM) and microspectrophotometric (MSP) procedures for the cytochemical DNA analysis.

Chronic lymphocytic leukemic B cells (B-CLL) were found to produce an IL 1-like growth factor spontaneously in vitro for mouse thymocytes. This factor was comitogenic with concanavalin A (Con A) and nonmitogenic combinations of phorbol ester and calcium ionophore but not with phyto-hemagglutinin (PHA). Growth factor production was dose- related to the number of in vitro cultured cells and detectable at 6 hours using high cell concentrations. A small number of admixed normal T cells was not important for factor production. No growth of autologous B-CLL or allogeneic thymocytes was induced by the factor. A chromatographic high-performance liquid chromatography (HPLC) analysis and inhibition experiments with a polyclonal rabbit anti interleukin 1 (IL 1) antiserum indicated that the B-CLL-derived growth factor belonged to the IL 1 family. This was supported by the direct demonstration of IL 1 beta in supernatants from B-CLL by radioimmunoassay. Possible biologic implications for B-CLL-derived IL 1 are discussed in relation to tumor cell growth in different clinical stages.