Activation of oestrogen receptor has a promoting effect in early steps of mammary carcinogenesis. Loss of oestrogen receptor, or of its functionality is associated to more aggressive and hormone unresponsive breast cancers. Anti-oestrogens are antiproliferative agents inhibiting oestrogens and growth factors action by binding to the oestrogen receptor. Assay of oestrogen and progesterone receptors, in the cytosol of primary breast cancer provides information on the differentiation and prognosis of the tumour and on the efficacy of endocrine treatment. Several oestrogen receptor variants have been detected in breast cancer, but their significance remains hypothetical. They could act in a positive constitutive fashion (e.g. anomaly of the hormone binding site with conservation of the DNA binding site) or be nonfunctional (e.g. anomaly of the DNA binding domain and or transactivation). They are often associated in the tumor with normal wild type receptors and could decrease their activity by forming nonfunctional heterodimers.

Among the 185 retinoblastoma patients seen at the Lausanne Retinoblastoma Clinic from 1963-1993, 24 (14%) first presented with another sign than classical leukocoria (60.5%) or strabismus (21.5%). Most of these atypical signs were related to inflammatory complications of unrecognized retinoblastoma; they consisted of low vision (1.5%), hypopyon (2%), ocular redness and pain (1.5%), ocular redness and buphtalmia (1.5%), as well as photophobia and headaches (1.5%). The presence of unexplained chronic ocular signs during childhood should always raise the possibility of an underlying retinal malignancy.

Mutations of Gs are found in 30 to 40% of the pituitary tumours responsible for acromegaly (somatotrope tumor), 10% of secreting thyroid tumors, and in the rare McCune-Albright syndrome. Gs is a member of the G protein family, which plays a major role in the mechanism of action of many hormones by coupling their membrane receptors to molecules called effectors, which general intracellular second messengers. Gs is responsible for the coupling of many receptors to adenylate cyclase which synthetizes cyclic AMP. The effect of the mutations of Gs is to permanently activate adenylate cyclase by a mechanism similar to the one exerted by cholera toxin. In the somatotrope cells adenylate cyclase is normally under the control of the hypothalamic hormone GH-RH, which stimulates both growth hormone secretion and cell proliferation. The mutations of Gs allow the somatotrope to escape from the control by GH-RH and to secrete and proliferate in an autonomous way, which leads to the development of a tumor responsible for acromegaly. The same mechanism explains the presence of Gs mutations in the other types of endocrine tumors. Therefore, the mutations transform the gene of Gs into an oncogene, gsp, which represents the first molecular explanation of an old concept familiar to endocrinologists: the autonomy of the secretion and proliferation of endocrine tumors.

In a case of glioblastoma, the following karyotype was determined: 47, X, - Y, + der(1) t(1;9)(p21;p23), t(1;9)(p21;p23), + 3, + 7, der(9) t(Y;9)(q11;p21), - 13, t(13;16)(p13,p11), del(14)(q11q22). Classical satellite DNAs are mainly located in chromosomes 1, 9, 15, 16 and Y. Because, most of these chromosomes were implicated in the rearrangements, a detailed cytogenetic study was undertaken. This study included in situ hybridization of the satellite and alphoid DNAs of chromosomes 1, 9, 16 and Y combined with various chromosome banding methods (DA-DAPI, quinacrine mustard and R-banding). The data obtained, demonstrated that the breakpoints were always located outside the areas containing the satellite and alphoid DNAs. The situation observed here differs from that reported in breast cancers for which a high proportion of the breakpoints occur within these areas. These findings suggest that in glioblastoma, chromosome rearrangements result from different mechanisms than those implicated in breast cancers. Thus, in cancers, chromosomal instabilities may result from several mechanisms.

While the value of hormone replacement therapy appears now to be accepted, it is above all a prophylactic technique requiring evaluation of its benefit/risk ratio. The possible increased risk of breast cancer linked to this type of treatment remains a preoccupation which probably limits its more widespread use. The natural history of this cancer and inherent biases in the various types of study are such that analysis of the literature is often difficult and any definite conclusion is prevented. Certain trends nevertheless appear to emerge from the principal studies. While there is indeed an increased risk of onset of breast cancer, it is slight, other than with the use of high doses of estrogens for prolonged periods in women with a family history of breast cancer. Decreased overall mortality rate in women on replacement therapy and the improvement in their comfort tend to suggest that it is now legitimate to offer such treatment, provided its contra-indications are strictly observed and regular monitoring is ensured.

The use of prognostic factors to help select breast cancer patients for adjuvant therapy is of considerable concern to the oncology community. This need for selection of prognostically less favorable cases is stimulating investigators to identify new and more powerful prognostic factors. Unfortunately however, this identification process is becoming more confusing because of a lack of guidelines for investigators to use to study new factors and for reviewers and readers to use to evaluate papers on this topic. In this paper, we will describe across our experience the main problems encountered in the study of biological prognostic studies. Considering evaluation criteria to be developed in the future, it appears that only multicentric and multidisciplinary structures are able to define decisional trees based on technically and clinically validated parameters in particular patients subgroups. Such a structure exists at the european level ("Receptor Study Group" of the EORTC) and a similar structure has now been created in France to answer these questions.

In all normal cells, two type of genes, oncogenes and anti-oncogenes, are expressed and control cell proliferation and differentiation. Cell growth is stimulated by oncogenes and inhibited by anti-oncogenes. Cancerization involves loss of control due to defective gene expression either by overexpression of a normal protein (loss of quantitative control) or expression of an abnormal protein (loss of qualitative control). Several oncogenes have been identified. They include three oncogenes, c-myc, N-myc and L-myc, known to be overexpressed in small-cell carcinomas of the lung. Point mutations of the oncogene K-ras is found in 15 to 30% of adenoma carcinomas, especially in smokers. Loss of anti-oncogene function has also been described in processes leading to lung cancer. Chromosome abnormalities, for example the 3p14-23 deletion described in 1982, are found in 100% of small-cell carcinomas and in 50% of non-small-cell carcinomas. This deletion is never found in normal tissue. The gene involved has not yet been cloned. Other mutations or deletions include the RB gene, necessary for neuroendocrine differentiation, and the p53 gene which has undergone mutation in 50% of the non-small-cell carcinomas and 70% of the small-cell carcinomas. These acquired mutations are strongly associated with tobacco smoking. Oncogenes and anti-oncogenes play an important role in the complex step-wise process leading to cancerization. As tumour characterization becomes more precise and precancerous states better controlled, future treatments may relief on inhibiting tumoural growth by using drugs which would substitute for the lost effect of anti-oncogenes or inhibit activation of an oncogene. But at the present time, it is still difficult to define criteria predicting high risk of postoperative relapse or resistance and further studies investigating the correlation between genetic abnormalities and clinical staging and survival curves are required.

A measurement of cell DNA content would be highly useful in determining the malignant nature of thyroid tumours in cases without distinctive features such as metastases, capsule invasion or emboli. Abnormal cell ploidy can be recognized with flow cytometry, but it is not known whether such results have diagnostic value. We therefore compared--in a double blind prospective study--the results of flow cytometry and pathologic diagnosis in fresh tumoural and non-tumoural thyroid cells.

The gene responsible for familial adenomatous polyposis, (APC), has been recently cloned and genetic map with several polymorphic markers has been established.

Alteration of chromosome 1 is the most consistent cytogenetic abnormality found in human breast carcinoma. Cytogenetic studies have shown independent alterations on the two arms of chromosome 1; increased copy number of the long arm and loss of the short arm of chromosome 1. We carried out deletion analysis of the 1p region by using restriction fragment length polymorphism markers mapping to the long (six markers) and short arm (22 markers). Thirty-five of the 74 (47%) human breast tumors tested showed somatic loss of heterozygosity at one or more loci on the short arm. Two commonly deleted regions, 1p13-p21 and 1p32-pter, were identified. Our findings suggest that two tumor suppressor genes involved in the development of human breast carcinoma may occur on the short arm of the chromosome 1.

Cytogenic analysis of leukemic cells has proven to be a mandatory part of the diagnosis of malignant hemopathies. Recurring clonal cytogenetic abnormalities may be divided into those exclusively associated with myeloid disorders, those uniquely observed in lymphoid diseases, and those detected in both myeloid and lymphoid hemopathies. Several of the common defects are characteristic of specific FAB types or subtypes and are associated with specific clinico pathologic syndromes and clinical complications. Cytogenetic abnormalities have served to define relatively homogeneous subsets of malignant hemopathies which are not evident from morphological and other available markers. Cytogenetic findings have been demonstrated to be powerful indicators in predicting clinical course and outcome in patients and in guiding their management. Given the significant progress made in the treatment of malignant hemopathies, it is very important to identify parameters which may be used to predict whether patients will respond favorably to standard therapies or if they are unlikely to do so and require alternative strategies, such as bone marrow transplantation. Cytogenetic studies have also provided important insights into the understanding of malignant transformation processes. In a number of recurring chromosome translocations characteristic of leukemias and lymphomas the genes that are located at the breakpoints have been identified. Molecular analysis has revealed that alteration in expression of these genes or in the properties of the encoded proteins resulting from the rearrangements plays an integral part in malignant transformation. Studies of clonality have suggested that several chromosome abnormalities may arise in pluripotent hemopoietic stem cells, whereas others may originate in cells of more restricted lineage. The author focuses first on the implications of the karyotype in the diagnosis and the prognosis of myeloproliferative syndromes, acute leukemias and myelodysplastic syndromes, then on the interest of describing new clinical-cytogenetic associations. Finally, some of the recent results obtained in a cytogenetic study of myelodysplastic syndromes are discussed.

There has been much progress in the cytogenesis, and molecular biology of bone tumours such as Ewing sarcoma and osteosarcomas, greatly improving diagnostic possibilities and prognosis. Ewing's sarcoma is an indifferentiated sarcoma with round cells which usually occurs in children or adolescents. Ewing's sarcoma corresponds to 6% of all bone tumours. Histologically Ewing's sarcoma belongs to a group of small round cell tumours including neuroblastoma, embryon and alveolar rhabdomyosarcoma and non-Hodgkin's lymphoma. Differential diagnosis is difficult. Cytogenetic examinations can now differentiate Ewing's sarcoma from other small round cell tumours. There is a specific 11:12 translocation (q24; q12) which can be used as a marker.

The eponym Li-Fraumeni syndrome is given to a particular form of cancer-prone family in which the main encountered tumors are sarcomas occurring in childhood and breast cancers affecting young adult females. There is also an increased frequency of cerebral tumors, leukemias and adrenal carcinomas among these families. The transmission of the cancer-proneness is autosomal dominant and related to the loss of function of the p53 tumor suppressor gene located on the short arm of the chromosome 17. The related p53 protein is identified but its precise mechanism of action and its regulation are still unclear. It seems to activate the genes which negatively regulate the multiplication of the cell and to act as a factor of transcription. A germline mutation of the p53 gene must be looked for in every cancer-prone family but also in case of multifocal tumors, particularly osteosarcomas and glioblastomas, and in case of second malignant neoplasm. This is of major scientific as well clinical interest. Indeed, the study of new families will help to better understand the molecular mechanisms underlying the syndrome. For the family itself this allow to identify the cancer prone members and to offer them preventive measures and early detection of cancers, particularly early breast cancer detection.

Turner syndrome is a complex human phenotype most commonly seen in association with a 45,X karyotype and it has been proposed that the phenotype is the result of monosomy for genes common to the X and Y chromosomes. Detection of unrecognized Y derived material is now possible by PCR of the SRY gene. Its presence is correlated with the presence of testicular tissue, known to increase the risk of developing gonadal neoplasia. Study of Y chromosome allowed the localisation of a candidate gene for the development of gonadoblastoma, GBY. Moreover, some groups described genes on the Y chromosome whose defects seem to be involved in the development of Turner stigmata: ZFY and RPS4Y. In conclusion; molecular genetics of the Y chromosome develops new pathophysiological and fundamental perspectives of the molecular genetics of the Turner syndrome.

Breast carcinoma is the commonest cancer in women. Segregation analyses support the existence of a major susceptibility gene with a dominant pattern of inheritance in 4 to 10% of cases. Three genes involved in the genetic predisposition to breast tumors have recently been identified: BRCA-1 associated with early onset familial breast cancer and breast/ovarian carcinoma syndrome, p53 associated with Li-Fraumeni syndrome, and the androgen receptor gene in male breast cancer with Reinfenstein syndrome. A better understanding of inherited contribution may have important implications for screening individuals at high-risk in families.