Systematic allelotyping of colorectal cancers and liver metastases is realized since 1996 in a close collaboration between the surgery, pathology departments and the molecular biology laboratory. Using amplification of 35 different microsatellites allelic imbalances and microsatellites instabilities were recorded. Beside the well documented increasing amount of genomic rearrangements we identified a subset of early stages tumors presenting a profile of rearrangement corresponding to high grades, and high grades cancers with a molecular profile corresponding to low grade. The comparison of synchrones liver metastases and colorectal tumors allowed us to propose for a more extensive studies the existence of sensitive structural chromosomic regions affecting separately both chromosomes and that it should be possible to identify a limited number of specific genes concerned by most of the advanced stages.

Hereditary Non-Polyposis Colon Cancer (HNPCC) or Lynch's Syndrome remains an underappreciated clinical entity-perhaps due to the restrictiveness of the diagnostic criteria. Even though HNPCC makes up only a small fraction (2-3%) of Colorectal Cancer (CRC) compared with sporadic CRC, it is important to make the diagnosis to allow for targeted management of the patients and their kindreds and thus to reduce morbidity and mortality. Better knowledge of tumor characteristics including RER phenotype may allow for the application of less restrictive diagnostic criteria; but these characteristics are not specific to HNPCC and also occur in 15% of sporadic CRC. Nevertheless, a screening strategy combining a broader clinical selection (less strict than the Amsterdam criteria) with a study of tumoral phenotype may allow the identification of more cases of HNPCC. The physician caring for cases of CRC must be aware of these screening strategies and relevant clinical and biological clues which might indicate a more complete genetic investigation.

Two thousand and three was a particularly dense year for publications and communications on therapy for colon cancer summarizing the real advance performed in this field. The last ten years allowed a rapid evolution for colon chemotherapy with a switch from 5-FU modulated by leucovorin to poly-chemotherapy (fluoropyrimidines with oxaliplatin or irinotecan) integrated into therapeutic strategies, where surgery had a place more and more important in metastatic patients. In correlation with these advances, median survival of patient with metastatic colorectal cancer is between 17 and 22 months. Targeted therapeutics with monoclonal antibody such as EGF inhibitors (cetuximab) or VEGF inhibitors (bevacizumab) had for the first time demonstrated efficacy with encouraging results in randomised trials. In adjuvant situation, LV5FU2 is less toxic than monthly FUFOL and no statistically significant difference could be detected in disease-free or overall survival between the two schedules. Oxaliplatin combined with 5 fluorouracil and leucovorin (FOLFOX4) is the first combination to demonstrate significant superiority over 5 fluorouracil and leucovorin in adjuvant treatment of colorectal cancer. Fluorouracil-based adjuvant chemotherapy benefited to patients with stage II or III colon cancer with microsatellite-stable tumours or tumour exhibiting low-frequency microsatellite instability but may be not those with tumours exhibiting high-frequency microsatellite instability (MSI). These data need to be confirmed by prospective studies before changing our therapeutic references. The number of lymph nodes analyzed for colon cancer staging is itself a prognostic variable on outcome. Laparoscopic surgery of colon cancer is demonstrated as a feasible and safe procedure. Shrinkage of tumours after administration of preoperative chemotherapy and availability of ablative techniques (radiofrequency and cryotherapy) now allow to treat with curative intent metastases initially considered as non-resectable.

Cytogenetics has clearly established the key role chromosomal rearrangements play in neoplastic initiation and progression. Investigation methods have evolved considerably from banding analysis of chromosome morphology to fluorescence in situ hybridization (FISH) and now to comparative genomic hybridization (CGH) on chromosomes or micro-array analysis of DNA. In addition to its contributions to the description, prognosis and understanding of oncogenesis and tumor progression, cytogenomics provides the information required for the rational use of new targeted therapies among which Imatimib is the most achieved example.

Progress in understanding the molecular mechanisms of oncogenesis, together with the sequencing of the human genome and the availability of huge databases, bring new tools for the discovery and the evaluation of new potential targets for cancer therapy and for the individualisation of cancer chemotherapy as a function of the molecular characteristics of tumours. Targeting the genetic alterations of cancer cells appears feasible and the first successes of this approach allow to remain optimistic about the renewal of our therapeutic armamentarium. In addition, seeking for correlations between gene expression profiles and chemosensitivity has been performed on the in vitro models of the National Cancer Institute and may allow crucial improvements in the identification of patients who world best take advantage of a specific chemotherapy. Clinical trials, first on a retrospective basis, then performed prospectively, are implemented to validate this approach.

Liposarcoma of the vulva is a rare entity. This unusual localization with atypical clinical and histological appearance may induce diagnostic and treatment delay. We report the 13th case shown in the literature in a 31-year-old woman initially treated for a vulvar lipoma. Arguments based on clinical short term recurrence, histological infiltrating adipocytes, and cytogenentical findings evoked well-differentiated liposarcoma. Even though cytogenetic abnormalities, involving MDM2 and CDK4 genes, have been found, a certainty in malignity diagnosis could be difficult. In these cases, treatment decision may be uneasy. This case report recalls difficulties encountered in uterine hypercellular leiomyomas.

Polycystic ovary syndrome (PCOS) is a common cause of hyperandrogenism in adolescent girls. In its complete post menarchal expression, the syndrome is characterized by the association of typical clinical, biological, and ultrasonographic findings. Many factors have contributed to our knowledge of different clinical forms of PCOS in adolescent girls. They are helpful for clarifying misleading situations in a period of life when diagnosis of PCOS implies a treatment for many years and may interfere with gynecological outcome. During the last 3 years, we had the opportunity to manage in our unit 45 adolescent girls with ovarian hyperandrogenism: 32 of them had PCOS and the other 13 functional ovarian hyperandrogenism defined by clinical and biological hyperandrogenism without ultrasonographic abnormality. In this review, we report, from our personal experience as well as from recent literature data, the different clinical expressions of PCOS in the pubertal period: the classical post menarchal form, the exceptional pre menarchal form, the post precocious pubarche and the post precocious puberty forms, the familial expression as well as the dominant metabolic expression.

Many epithelial cancers have been found to overexpress the receptor to epidermal growth factor (EGFR) including head and neck, breast, colon, lung, prostate, kidney, ovary, brain, pancreas and bladder cancer. Because of the association of EGFR overexpression with overall poor prognosis in patients with cancer, a number of strategies to block or downregulate EGFR have been developed to inhibit tumor proliferation and improve clinical outcome. These include monoclonal antibodies directed against the EGFR such as IMC-C225 which specifically targets EGFR and ZD 1839 (Iressa) capable of inhibiting EGFR tyrosine-kinase in vitro. This report will focus on antibodies that target EGFR in renal cell carcinoma, prostate cancer and bladder cancer.

Melanoma prognosis is based on histological criteria such as tumor thickness (measured by Breslow index), level of invasion (Clarck's level), presence of ulceration and number of mitoses per mm2. However, these parameters do not provide a precise prognosis in all cases: thin melanomas may develop metastases and thick melanomas may remain focalized for many years. For these reasons, the search for other prognostic factors is still ongoing. Many molecules play a part in the invasiveness and metastatic dissemination of melanoma have now been identified. Expression of these molecules has been studied in primary melanoma and correlated with prognosis. An increase in the number of cells positive for Ki67 (detected by Mib1), cycline A, cycline D, p35, MMp-2, beta1 and beta3 integrins, osteonectin, the presence of an intense inflammatory infiltrate and capillary invasion are considered as factors of poor prognosis as well as the decrease in p16, p27, Melan A and nm23. The significance of CD44 modifications is still controversial. Only a small number of these different proteins has a prognostic value independent of tumor thickness. These results need to be confirmed on larger series of patients. Additional hope is given to new techniques such as the analysis of the genes implied in tumor progression by microarray technique in such a way as to provide a molecular map of each tumor.

Cutaneous leiomyoma is a benign tumor, the discovery of which may suggest a hereditary form. We report a family in which 5 generations developed cutaneous and uterine leiomyomas. The originality of this report lies in the large number of generations developing the disease and the association with chronic myeloid leukemia.

Ultraviolet radiation (UV) is considered as a key environmental risk factor of non-melanoma skin cancer (NMSC), but other factors such as immunological status, genetic predisposition and infection by human papillomavirus (HPV) may also be involved. Although there is overwhelming epidemiological and molecular evidence that indicates a direct role for specific mucosal HPV-types in anogenital cancers, in particular cervical cancer, the pathogenic role of HPV in the development of NMSC remains speculative. The association between HPV and NMSC was first identified in patients with epidermodysplasia verruciformis (EV) and later in recipients of organ transplants. All these patients develop NMSC at sun-exposed sites. Cutaneous and mucosal HPV-DNA have been detected in about 60 to 90 p. 100 of NMSC, but also in benign epithelial lesions, and even in normal skin. However and although at a lower rate (about 40 p. 100), HPV-DNA have also been detected in normal skin, in particular in hair follicles, and in premalignant lesions and in NMSC from non-EV and immunocompetent subjects. Furthermore, no particular HPV type predominates and the viral load in NMSC seems lower than in benign epithelial lesions. Although all these findings argue against a direct involvement of HPV in NMSC, they may suggest a "hit and run" mechanism which no longer requires the viral agent but the activity of HPV oncoproteins. High risk mucosal HPV-types encode two major oncoproteins, E6 and E7, which inactivate two suppressor proteins, p53 and pRb respectively, and are sufficient for host-cell immortalization. A polymorphism resulting in either a proline or an arginine at codon 72 may also be a relevant risk factor for mucosal HPV-types-associated NMSC. By contrast, E6 of skin HPV-types fails to interact with p53, but prevents infected cells from UV-induced apoptosis leading thus to the propagation of deleterious UV-induced mutations. Immunosuppressive activities of HPV E6 and E7 proteins permit persistent HPV infection and the impairment of immunologic removal of UV-damaged cells. These results support a role for HPV infection in skin carcinogenesis as a co-factor in association with UV and immunosuppression.

Little is known about epidemiology of adults soft tissue and visceral sarcomas (ASTS). The frequency of previous cancers and associated genetic diseases has been analyzed out of 493 ASTS, treated between 1997 and 2002 at Oscar Lambret Cancer Center. Median age is 51, sex ratio is close to 1. Liposarcomas and malignant fibrous histiocytofibromas are the two main types (respectively 104 and 86 cases). Upper and lower limbs are the two main locations (respectively 176 and 75 cases). Fifteen patients had associated genetic disease, including 12 cases of Recklinghausen diseases. 7 out of those 15 patients have neurosarcoma. 30 patients have previous cancers, including 7 breast cancers, 3 lymphomas and 3 chronic lymphocytic leukemias. Four out of those 30 patients have two different previous cancers. 13 patients have radiation-induced sarcomas, after an average 10-year-period, and an average dose of 53 Gy. Undifferenciated sarcomas are the main histologic type (8/13), followed by angiosarcomas (2/13). Radiation-induced sarcomas are located in the chest wall (7/13), in pelvis (2/13) and head and neck (2/13). Those sarcomas are high grade (10 grade III tumours). ASTS epidemiology is complex with different risk factors depending on histologic type.

BRCA1 is a breast cancer susceptibility gene. Germline mutations in BRCA1 gene are found in 5 to 10% of breast cancer. The aim of this study is to screen the tunisian women with familial or sporadic breast cancer for BRCA1 gene mutations. The authors used the Protein Truncation Test (PTT) and DNA sequencing to detect BRCA1 gene mutations in 12 tunisian families with breast cancer and the Allele Specific Oligonucleotide-PCR (ASO-PCR) to detect the 185delAG and 1294del40 mutations in 150 tunisian women with sporadic breast cancer. A nonsens mutation was found, by PTT, in exon 11 of BRCA1 gene in one case of familial breast cancer. No mutation in the rest of exons was found by the DNA sequencing. The BRCA1 1294del40 mutation was found only in a patient with non familial breast cancer. The 185delAG mutation was absent in all cases of breast cancer. These data suggest that the germline mutation of BRCA1 is implicated in breast cancer in Tunisia and that the 185delAG mutation is absent in arab tunisian women.