This review focuses on the genetics of inborn predisposition to the development of cancer. The gene defects responsible for the majority of the most common syndromes of inherited predisposition to cancer have been deciphered during the last five years. These discoveries have shown that the genetics of most of these inherited forms of cancer is remarkably simple: the mechanism underlying the risk of cancer is precisely the one proposed by Knudson to explain the inherited forms of retinoblastoma. Only a small minority of the cancers are imputable to inherited mutations causing a very high risk of cancer. However, low penetrance susceptibility genes, mainly acting as environmental risk modifiers, are probably involved in the development of the majority of cancers. As much as possible, we will consider the immediate and long term possible repercussions on everyday medical practice of these advances in basic understanding of genetic cancer risks.

The authors report the observation of a 35-year-old patient, mother of 5 children 4 of whom are carriers of bilateral retinoblastomas. Two of these children presented lesions suggestive of retinoma retinoblastoma association. The patient had peripheral retinal lesions evoking the diagnosis of regressive bilateral retinoblastoma. At present, the term retinoma or retinocytoma seems to be more adequate. The genetic study of this family was done with D esterase and DNA molecular biology.

Despite numerous publications on the subject, the debate over the merits of the various methods available for screening for colorectal cancer is continuing. Choice of the method depends on the patient being treated. In the present state of our knowledge, mass screening by Hemoccult, which is of low cost but of poor sensitivity and only fairly well accepted, cannot be used routinely but must be further evaluated and developed. In persons willing to undergo regular examinations, an annual or biannual Hemoccult examination associated with rectosigmoidoscopy every 5 years appears appropriate. In subjects having a familial history in first-degree relatives, colonoscopic examination every 5 years is adequate. Lastly, in patients with a personal history of cancer or adenoma, screening consists of colonoscopic follow-up adapted to the individual case.

All cancers result from the accumulation of mutations of proto-oncogenes and tumor suppressor genes. Sporadic and familial colorectal cancers result from the accumulation of the following genes, in a relatively stereotyped chronological order: the tumor suppressor gene apc whose mutations are responsible for the familial adenomatous polyposis; the proto-oncogene K-ras which is mutated in 50% of large adenomas (> 1 cm) and adenocarcinomas; the tumor suppressor gene dcc; and the tumor suppressor gene p53 whose inactivation in a factor of bad prognosis. While some of them are induced by mutagens, others result from an instability of the genome. Two types of instability are observed in both sporadic and familial colorectal cancer. The first type, which is found in 25-50% of cases, appears as cytogenetic abnormalities with aneuploidy and allelic losses. The second type of instability is induced by mutations of the hMSH2 or hMLH1 genes which code for proteins involved in the mechanism of DNA repair.

One must always keep in mind the distinction between constitutional genetic mutation and acquired or somatic genetic mutation. Our knowledge of the syndromes resulting from germ line mutations and the detailed clinical description of the diseased patients have been the key to research into the field of susceptibility to colorectal cancer. Research has now entered a new era due two major advances. First it is now possible to identify persons at risk, carriers of constitutional defects involving genes responsible for colorectal cancer. Secondly, it is now possible to better identify high risk lesions based on histological examination and use of molecular markers. In the future, further progress will allow us to analyse acquired mutations. These advances open the way to new diagnostic approaches and raise a considerable number of questions concerning the ethics of molecular diagnosis and treatment.

Between 1987 and 1992, we operated on 15 patients with tubulopapillary tumours (TPT) of the kidney, corresponding to 10% of all renal cancers operated during this 5-year period, which brings the total number of published cases to 418. TPT differ from non-tubulopapillary tumours in several ways. The medical imaging findings may differ when the tumour is not necrotic (20 to 30% of cases). The histology differs not only in terms of the tumour architecture, but also by the frequent association (9/15 cases) with other ipsilateral or contralateral renal anomalies: 2 bilateral tumours, 3 tumours associated with multiple cysts and 4 multifocal tumours. These characteristics, combined with the finding of 12 familial cases suggest a genetic origin for TPT, especially as the genetic abnormalities detected (tri-or tetrasomy 17) differ from the anomalies usually detected in non-tubulopapillary tumours (anomalies of 3p and 5q). This study is the first to report familial cases of TPT. The prognosis of these tumours, generally better than that of non-tubulopapillary tumours, suggests the possibility of conservative surgery when the diagnosis can be established preoperative or at operation.

The development of biotechnologies offers considerable potential for new therapeutical approaches, targetting nucleic acids as information bearing molecules responsible for various pathologies. Gene therapy has, up to now, mainly aimed at compensating deficiencies of gene expression, through complex molecular and cellular constructions. Recently, an increasing interest has focused on short nucleotidic sequences (thus relatively easy to synthetize), or oligonucleotides, which could be able to specifically block the expression of mutated or overexpressed genes. When the target of these oligonucleotides is the mRNA of such genes, the strategy is called "antisense". This strategy has already led to a number of successful results in experimental models in vitro. Much more rare are actual effects in animal models and clinical trials are just being sketched. A review on the current state of the art will enlighten the physiological potential of these new molecules but also underline the still numerous issues before controlling therapeutical applications.

Expression of mdr1 and p53 was assessed on 119 cases of bronchial carcinomas, and compared with clinical chemoresistance. mdr1 expression was evaluated by immunohistochemistry (IHC), using the monoclonal antibody JSB1. The study of p53 expression was performed by both IHC, using six different antibodies, and Northern blotting. We observed a correlation between the expression of mdr1 and the presence of a mutation of p53 in neuro endocrine (NE) carcinomas (P = 0.02). Correlation was not observed when non NE carcinomas were evaluated, nor was found any correlation between mdr1 expression and clinical chemoresistance in patients with small cell lung carcinoma (SCLC). The frequency of complete response however was significantly higher in patients whose tumor did no express mdr1 (P = 0.02). Chemoresistance correlated well with the phenotype of p53 mutant in SCC (P = 0.015). We conclude that p53 mutation is a better predictive factor of clinical chemoresistance in SCLC than mdr1 IHC detection.

Twenty four squamous cell carcinomas of the head and neck (HNSCC) of stage II to IV were evaluated for the expression of potential markers such as oncogenes and tumor suppressor genes in drug-resistance behavior. We have analysed the c-myc, c-jun, c-raf and N-ras and p53 expression in total RNA preparation from tumor biopsies obtained before treatment. The patients underwent chemotherapy including 5-fluorouracil and cisplatinum. No significant differences in c-raf and N-ras expression were found in responding or resistant patients. However, resistance to chemotherapy was associated with low expression of c-myc (P < 0.025) or high expression of c-jun (P < 0.001). In addition, p53 mRNA pre-therapeutic level was increased in unresponsive patients to chemotherapy (P < 0.05). Therefore, analysis of the expression of c-myc, c-jun oncogenes and p53 tumor suppressor gene in tumor cells before initiation of therapy may define a subset of patients with potentially better prognosis.

The prognosis of colorectal cancer has been based essentially on pathological data for many years. The analysis of genetic anomalies has led to fundamental progress and clinical advances. Genetic anomalies are routinely studied. 1--Flowcytometry evaluates the quantity of DNA in the nucleus during the cell cycle. 2--Cytogentics is the study of karyotype anomalies by loss or gain of chromosome material and structural changes. 3--Molecular biology gives a means of recognizing chromosome losses and especially to study oncogenic or antioncogenic mutations. These analyses allow: 1--an evaluation of their value as a prognosis factor and thus their use for indicating adjuvant medical and/or surgical treatments. 2--an understanding of cancerogenic processes. 3--the development of future therapeutic techniques based on a better understanding of the mechanisms involved. 4--familial counselling in high risk families and an examination of responsible or favouring genes in certain familial cancers. Research into familial forms has recently led led to the demonstration of genetic alterations located on chromosomes 1 and 2. These anomalies called RER correspond to alterations found on tumors. Studying these alterations will allow better prediction of high risk subjects in cancer families without polyposis.

The spontaneous release of a glyconucleoprotein complex in the supernatant of eukaryote cell cultures is a general phenomenon independent of cell lysis. The DNA recovered from this glyconucleoprotein material contains most part of the genome. The SW 480 cell line, originating from a human colon carcinoma, presents a point mutation of the K-ras gene on both alleles. These cells in culture release the mutated K-ras gene. When crude SW 480 cell supernatant is given, without any other adjonction, to NIH/3T3 mouse cells, transformed foci appear as numerous as those occurring after a transfection provoked by a cloned E.J. ras gene administered as a calcium precipitate. The presence of a mutated ras gene in the transfected foci of the 3T3 cells has been checked by hybridization, after PCR, with an oligonucleotide probe specific to the mutation. This result was confirmed by sequencing the PCR product.