At least five regions of imbalance on 6q in breast tumors, combining losses and gains.

The long arm of chromosome 6 is frequently rearranged in human cancer. In breast cancer, allelotyping studies have indicated the existence of three to four distinct regions of allelic imbalance. Chromosome transfer studies have shown the presence of several growth inhibiting or senescence promoting genes in the segment between 6q13 and 6q27. Moreover, results from comparative genomic hybridization (CGH) analyses have indicated that 6q was indeed a site of chromosomal losses, but that it was also involved in a substantial number of gains. In the present work, we allelotyped 178 pairs of breast tumor and normal tissue DNAs using 30 CA repeat markers from the Genethon collection. Seventy-six percent of the tumors in our panel displayed allelic imbalance (AI) of at least one locus, but patterns of AI could be complex. Whereas 11 tumors showed AI at all markers tested, 57 presented zebra profiles, and 28 showed AI at one site only. We characterized five distinct domains of AI defined, from centromere to telomere, by D6S300 (domain 1), D6S434 (domain 2), D6S261 (domain 3), D6S314-D6S409 (domain 4), and D6S441-D6S415 (domain 5). Some of the domains could be narrowed down to intervals of 1cM or less. We performed CGH analysis on a subset of 34 tumors presenting AI of variable extent at 6q. In 10/34 tumors, CGH did not reveal any anomaly on 6q. Most of these presented AI on short intervals, thus being below the detection threshold by CGH. Of the remaining 24 tumors presenting anomalies by CGH, 11 presented gains involving all or portions of 6q and 15 losses (2 presented combined losses and gains). By CGH, the 6q21-22 region was most commonly involved in gains, whereas 6q13-14 and 6q25-27 were frequently lost. Thus, allelic imbalances on 6q can either represent a gain or a loss depending on the region involved. Genes Chromosomes Cancer 27:76-84, 2000.

Cytokine gene expression in human multiple myeloma.

In the present study the gene expression of cytokines promoting in vitro myeloma-cell growth was investigated by Northern blot analysis using total RNA of 36 tumour samples of patients with multiple myeloma (MM) or plasma cell leukaemia and poly(A)+ RNA of 10 human myeloma cell lines (HMCL). These cytokines included interleukin (IL)-1 alpha, IL-1 beta, IL-3, IL-6, granulocyte-macrophage (GM)-colony-stimulating factor (CSF) and granulocyte (G)-CSF. IL-1 beta, IL-6 and G-CSF genes were coexpressed in most patients, although at variable levels. IL-1 alpha transcripts were detected in 32% of patients in whom coexpression of IL-1 beta gene was found. IL-3 gene was not expressed in patients' cells and GM-CSF mRNA was detected in only 1/32 patients. No detectable transcripts for the above cytokines were present in HMCL, whereas IL-6 gene was expressed in 2/10 HMCL. We also looked for the presence of transcripts for IL-2, leukaemia inhibitory factor (LIF) and transforming growth factor (TGF)beta in cells of tumour samples from the same patients and in HMCL. IL-2 gene was not expressed in MM patients and HMCL. Weak expression of LIF gene was detected in three patients (9%), and transforming growth factor beta (TGF beta) mRNA was observed in 12/12 tumour samples analysed and all HMCL. These results suggest that, among cytokines shown to control myeloma-cell growth in vitro, IL-1, IL-6 and G-CSF could play a role in the development of myeloma disease in vivo.

Characterization of proliferative cells in malignant melanomas and their inflammatory infiltrates.

Distinct changes in the antigenic phenotypes of mononuclear cells infiltrating primary and metastatic malignant melanomas (MM) have been shown to characterize distinct steps of melanoma progression. The purpose of our study is to establish whether the growth fraction of malignant melanoma cells is related to the mononuclear cell subtypes. Using monoclonal antibody Ki67, the presence of a nuclear antigen in proliferating cells of both tumor and inflammatory infiltrate cells was established in 20 primary recurrent and metastatic cutaneous melanomas. Monoclonal antibodies against lymphocyte and macrophage subsets were also applied in situ. Numerous CD8 and CD4 positive cells and natural killer (NK) cells were detected in all the infiltrates. A low CD4+/CD8+ ratio was observed in most tumors with a high proliferative activity. The presence of positive CD1+ cells seemed also to be correlated with a high activity. Our data suggest a correlation between inflammatory cell subsets and proliferative activity of MM.

In situ evaluation of growth fraction determined by monoclonal antibody Ki-67 and ploidy in surgically resected non-small cell lung cancers.

Ploidy and growth fraction were analyzed by means of a computer-assisted image processor in surgically resected non-small cell lung cancer (NSCLC). This study was done in order (a) to evaluate the distribution of anti-Ki-67 immunostaining and (b) to correlate this distribution to ploidy status and pTNM stage of NSCLC. Thirty-two patients underwent a surgical resection for primary NSCLC following complete staging. Indirect immunoperoxidase reactions of monoclonal antibody Ki-67 were done on frozen tissue sections. Integrated optical density and index of stained nuclear surface were calculated by means of a computer-assisted image processor in 120 fields of each preparation in order to quantify the Ki-67 immunostaining. DNA content was determined by means of cytometry of Feulgen-stained cytological prints. The ploidy status was defined for each tumor by DNA index, percentage of hypodiploid cells, and type of DNA content histogram (near diploid, hyperdiploid, hypodiploid, and multiploid). Reproducibility of immunostaining quantitative analysis was demonstrated by iterative measurements of the same slide. Intratumoral heterogeneity of Ki-67 immunostaining induced integrated optical density variation assessed on six nonconsecutive tissue sections from at least two regions of the same tumor. This intratumoral variability was 15 times lower than integrated optical density variability between tumors. The Ki-67 immunostaining varied significantly according to the DNA content histogram type (P less than 0.05, Kruskal-Wallis test); most of the specimens with high Ki-67 immunostaining were multiploid or hypodiploid. Moreover, Ki-67 immunostaining correlated to the percentage of hypodiploid cells. Ki-67 immunostaining and ploidy status did not vary significantly according to the tumor-nodes-metastasis stage. We conclude that (a) quantitative analysis of Ki-67 immunostaining is a reliable evaluation of growth fraction in NSCLC if a large number of fields are analyzed to take into account intratumoral variability, (b) hypodiploidy and multiploidy are frequent abnormalities of DNA content, (c) Ki-67 immunostaining is significantly higher in hypodiploid and multiploid tumors. Thus, determination of growth fraction and ploidy in surgically resected NSCLC specimens may be considered as complementary prognostic parameters independent of the stage of the disease.