Malignant epithelial tumors (carcinomas) are the most common ovarian cancers and the most lethal gynecological malignancies. Based on their heterogeneous morphology, a dualistic model of carcinogenesis was proposed in 2004. Type I carcinomas, composed of low grade serous, endometrioid, mucinous, clear cell carcinomas and malignant Brenner tumors, were distinct from type II carcinomas (high grade serous, undifferentiated carcinomas and carcinosarcomas). However, clinical studies failed to demonstrate the prognostic value of such a classification. The main reproach to this dualistic model was that it lumped together in type I tumors, heterogeneous lesions such as clear cell and mucinous carcinomas. Recent advances on molecular genetic alterations and precursor lesions favor the classification of ovarian carcinomas as five distinct diseases. The dualistic model of carcinogenesis in type I and II can finally be applied only to serous ovarian carcinomas (low grade and high grade).

In France, women carrying BRCA1/2 mutation, at an identified high risk of breast cancer are recommended to undergo breast MRI screening. That screening does not however prevent the risk of developing a breast cancer. The only alternative to breast cancer screening available in France is surgical prevention by prophylactic mastectomy. An interesting option for women who wish to reduce their breast cancer risk, but are unready for prophylactic mastectomy is a preventive hormonal treatment by aromatase inhibitors, or selective estrogens receptor modulators (SERMs). Reliable clinical trials show the efficiency of tamoxifen, raloxifen, exemestane, and anastrozole especially, in reducing breast cancer incidence by 33%, 34%, 65% and 53% respectively. This article tries to sum up the main published trials of breast cancer prevention with hormonal treatment, and presents the latest American and English clinical guidelines concerning hormonal prevention for women at high risk of breast cancer, and starts thinking about the possibilities of hormonoprevention, especially among women carrying a BRCA1/2 mutation in France.

Epidemiological studies have shown a relationship between endometriosis and clear cell/endometrioid ovarian cancers (named "Endometriosis Associated Ovarian Cancer" or EAOC). The recent discovery of signaling pathways (especially the SWI/SNF and PI3K/AKT/mTOR pathways) that linked endometriosis and EAOC could lead to the development of specific biomarkers as ARID1A to screen benign to premalignant endometriosis and to new targeted treatment. Moreover, the better understanding of the pathogenesis of the epithelial ovarian cancer arising from the Fallopian tube could allow new early prevention strategies that will be described in this review.

If acute leukemia is the most frequent cancer in childhood (33%), it remains a very rare diagnosis in infants less than one year old, e.g. less than 5% of cases. At this age, the frequency of acute lymphoblastic leukemia (ALL) (almost all of B-lineage) is quite similar to the one of myeloblastic forms (AML). Infant leukemia frequently presents with high hyperleucocytosis, major tumoral burden and numerous extra-hematological features, especially in central nervous system and skin. Whatever the lineage, the leukemic cell is often very immature cytologically and immunologically. Rearrangements of the Mixed Lineage Leukemia (MLL) gene, located on band 11q23, are the hallmark of these immature leukemias and confer a particular resistance to conventional approaches, corticosteroids and chemotherapy. The immaturity of infants less than 1-year-old is associated to a decrease of the tolerable dose-intensity of some drugs (anthracyclines, alkylating agents) or asks questions about some procedures like radiotherapy or high dose conditioning regimen, responsible of inacceptable acute and late toxicities. The high level of severe infectious diseases and other high-grade side effects limits also the capacity to cure these infants. The survival of infants less than 1-year-old with AML is only 50% but similar to older children. On the other hand, survival of those with ALL is the same, then quite limited comparing the 80% survival in children over one year. Allogeneic stem cell transplantations are indicated in high-risk subgroups of infant ALL (age below 6 months, high hyperleucocytosis >300.10(9)/L, MLL-rearrangement, initial poor prednisone response). However, morbidity and mortality remain very important and these approaches cannot be extended to all cases. During the neonatal period, the dismal prognosis linked to the high number of primary failures or very early relapses and uncertainties about the late toxicities question physicians about ethics. It is an emergency to propose different strategies (targeted therapies) to these infants with acute leukemia as conventional trials failed to improve outcome.

Molecular screening has become a standard of care for patients with advanced cancers and impacts on how to treat a patient. Advances in genomic technologies with the development of high throughput sequencing methods will certainly improve the possibilities to access a more accurate molecular diagnosis and to go beyond the identification of validated targets as a large number of genes can be screened for actionable changes. Moreover, accurate high throughput testing may help tumor classification in terms of prognosis and drug sensitivity. Finally, it will be possible to assess tumor heterogeneity and changes in molecular profiles during follow-up using ultra-deep sequencing technologies and circulating tumor DNA characterization. The accumulation of somatic ADN alterations is considered as the main contributing factor in carcinogenesis. The alterations can occur at different levels: mutation, copy number variations or gene translocations resulting in altered expression of the corresponding genes or impaired protein functions. Genes involved are mainly tumor suppressors, oncogenes or ADN repair genes whose modifications in tumors will impinge cell fate and proliferation from tumor initiation to metastasis. The entire genome of various tumor types, have now been sequenced. In lung cancer, the average number of mutations is very high with more than 8.9 mutations/Mb (Network TCGAR, 2014) that is to say more than 10,000 mutations/genome. These alterations need to be classified, indeed, some are true drivers that directly impact proliferation and some are passenger mutations linked to genetic instability. The development of targeted therapies relies on the identification of oncogenic drivers. The identification of genotype-phenotype associations as in the case of EGFR-TKI (Epidermal growth factor receptor-tyrosine kinase inhibitor) and EGFR mutations in lung cancer led to the restriction of drugs to patients for which tumor genotype predicts efficacy. Tumor-molecular directed therapy based on validated targets (EGFR, ALK) is in the clinics, rapidly, with the developments of multi-targets or multi-drug assays there will be a need for tumor-molecular-profile directed therapy. Today, there are practical challenges to a successful implementation of NGS technologies for clinical applications. Broadly, some are linked to the tumor (heterogeneity), to the tissue (availability, storage, fixative), to the design of specific assays or set of genes, to the interpretation of non-driver mutations and to a possible access to drugs once a target is identified. Technical challenges are solved, NGS (at least targeted-NGS) plateforms have been validated by INCa labeled laboratories, in this context, we will address different questions: How, for whom, what kind of profiling and what can we expect?

The 2015 WHO classification of tumors of the lung, pleura, thymus and heart has just been published with numerous important changes from the 2004 WHO classification. The most significant changes involve (1) use of immunohistochemistry throughout the classification, (2) integration of molecular testing for personalized strategies for advanced lung cancer patients, (3) a new classification for small biopsies and cytology, (4) a new classification of lung adenocarcinoma as proposed by the 2011 IASLC/ATS/ERS, (5) restriction of the diagnosis of large cell carcinoma only to resected tumors that lack any clear morphologic or immunohistochemical differentiation. Regarding adenocarcinoma, the terms bronchioloalveolar carcinoma (BAC) and mixed subtype adenocarcinoma have been suppressed and replaced for the former by the term adenocarcinoma in situ (AIS) as a preinvasive lesion to join atypical adenomatous hyperplasia (AAH). A new category has been defined, the minimally invasive adenocarcinoma (MIA), and invasive adenocarcinomas are now classified according to the predominant subtype after subtyping by semi-quantitatively percentage of various subtypes present in 5% increments. The term "lepidic" is restricted to a non-invasive component (previously classified as BAC) present as part of an invasive adenocarcinoma. "Invasive mucinous adenocarcinoma" is used for formerly adenocarcinomas classified as mucinous BAC, excluding tumors that meet criteria for AIS or MIA. The subtypes of clear cell and signet ring adenocarcinoma are discontinued, as well the term of mucinous cystadenocarcinoma, included in the category of colloid adenocarcinoma. Thus new classification of lung adenocarcinoma is sustained by genetics and has clinical impact for therapeutic strategies.

Ten years after the discovery of the predictive value of KRAS status for anti-EGFR antibodies, other genes involved in oncogenesis and therapeutic responses were identified and are now systematically sought. Molecular diagnosis often requires invasive procedures, sometimes iatrogenic, and is limited by feasibility problems, quantity and quality of samples. Identifying these mutations from blood biomarkers would reduce costs and diagnostic delay. The circulating tumor DNA (ctDNA) is one of the most promising blood biomarkers. In this review, we report and discuss the latest results obtained with ctDNA in colorectal cancer, non-small cell lung cancer, and adenocarcinoma of the pancreas. If the methods highlighting appear very heterogeneous, the correlation between mutations found in tumor and those identified in the blood exceeds 95 % specificity in numerous studies. The detection sensitivity is in turn strongly related to tumor stage patients. The presence of ctDNA appears as a prognostic factor for progression-free survival and overall survival. Finally, recent studies have shown that the changing rate ctDNA during systemic treatments had a predictive value for therapeutic efficacy. These results allow to consider the use of ctDNA in monitoring patients to identify early recurrence or progression.

Nowadays, the analysis of theranostic molecular markers is central in the management of lung cancer. As those tumors are diagnosed in two third of the cases at an advanced stage, molecular screening is frequently performed on "small samples". The screening strategy starts by an accurate histopathological characterization, including on biopsies or cytological specimens. WHO 2015 provided a new classification for small biopsy and cytology, defining categories such as non-small cell carcinoma (NSCC), favor adenocarcinoma (TTF1 positive), or favor squamous cell carcinoma (p40 positive). Only the NSCC tumors, non-squamous, are eligible to molecular testing. A strategy aiming at tissue sparing for the small biopsies has to be organized. Tests corresponding to available drugs are prioritized. Blank slides will be prepared for immunohistochemistry and in situ hybridization based tests such as ALK. DNA will then be extracted for the other tests, EGFR mutation screening first associated or not to KRAS. Then, the emerging biomarkers (HER2, ROS1, RET, BRAF…) as well as potentially other markers in case of clinical trials, can been tested. The spread of next generation sequencing technologies, with a very sensitive all-in-one approach will allow the identification of minority clones. Eventually, the development of liquid biopsies will provide the opportunity to monitor the apparition of resistance clones during treatment. This non-invasive approach allows patients with a contraindication to perform biopsy or with non-relevant biopsies to access to molecular screening.

Therapies targeting immune checkpoints, in particular programmed death 1 (PD-1) and its ligand programmed death ligand 1 (PD-L1), are major new strategies for the treatment of several malignancies including mestatatic non-small cell lung cancer (NSCLC). The identification of predictive biomarkers of response is required, considering efficacy, cost and potential adverse events. Expression of PD-L1 by immunohistochemistry has been associated with higher response rate and overall survival in several clinical trials evaluating anti-PD-1 and anti-PD-L1 monoclonal antibodies. Thus, PD-L1 immunohistochemical companion assays could be required for treatment with some of these therapies in NSCLC. However, heterogeneity in methodologies of PD-L1 assays in terms of primary antibodies and scoring algorithms, and tumor heterogenity for PD-L1 expression are important issues to be considered. More studies are required to compare the different assays, ensure their harmonization and standardization and identify the optimal conditions for testing. PD-L1 expression is likely an imperfect predictive biomarker for patient selection and association with other markers of the tumor immune microenvironment will be probably necessary in the future.

Experts increasingly recognize the hypothesis of "over-diagnosis" as the main factor of the raising incidence of thyroid cancers (TC). The detection of multiple microtumors, mainly of a papillary type, at a sub-clinical stage, with the use of sensitive detection methods supports this hypothesis. However, the intensive management and monitoring of these cancers failed to reduce mortality. Environmental and other risk factors cannot provide a sufficient explanation, as previously thought. In this context, the use of improved tools is needed, and the most promising perspective lies in molecular biology applied to thyroid cancer for diagnosis, evaluation of prognosis and treatment. The next generation sequencing (NGS) has demonstrated its diagnostic performances in recent clinical trials. Its interest in cases with indeterminate cytology is demonstrated and should help better targeting surgical indications. Its promising prognostic and therapeutic applications must be confirmed by additional studies. The integration of NGS in current practice should have a real medical, economic and scientific impact. Indeed, the exponential increase in our knowledge of molecular mechanisms of thyroid tumorigenesis strengthens the will to "reclassify" these cancers into molecular rather than histological subtypes, in order to offer patients more specific and targeted treatment.

Recently, developments of therapies that target abnormally activated signaling pathways are increasing for patients with non-small cell lung cancer. EGFR mutations are found in about 10% of lung cancers, especially in adenocarcinoma, women and non-smokers. Three EGFR inhibitors (erlotinib, gefitinib and afatinib) received a European marketing authorization for up to first line treatment of EGFR mutated NSCLC. Effectiveness of EGFR inhibitors is higher than conventional chemotherapy. Third generation EGFR inhibitors (rociletinib, AZD9291) are effective for patients who develop a resistance mutation such as T790M resistance mutation; they obtained temporary authorization for use in France in 2015. The EML4-ALK translocation is found in about 5% of NSCLC and more particularly in adenocarcinoma of young non-smoking patients. Crizotinib is a new therapeutic standard in ALK translocated NSCLC in second line. Ceritinib is a 2nd generation ALK inhibitor which received a European marketing authorization for up to treatment of ALK translocated NSCLC after progression with crizotinib. INCA supports ACSé program evaluating the efficacy of crizotinib in NSCLC amplified for MET or translocated for ROS1 and ACSé program evaluating the efficacy of vemurafenib in tumors non melanoma mutated V600E BRAF. The role of other biomarkers such as KRAS, BRAF, HER2 and PI3KCA mutations remains to be defined in NSCLC.

Synovial sarcoma (SS) is a rare high-grade malignant mesenchymal tumor affecting children, adolescents, and young adults. Cytogenetically, more than 90% of SS is characterized by the t(X;18)(p11.2;q11.2), translocation resulting in two chimeric fusion genes SYT-SSX1 and SYT-SSX2, confirming histological diagnosis. Pediatric SS arises most often in soft tissues of the extremities (66% of cases), and is a localized tumor without spreading to regional lymph nodes (96% of cases) nor to metastatic sites (94% of cases). Although clinical and radiologic presentation, histologic analysis and tumor biology appear similar in pediatric and adolescent SS, outcome seems better in children than in adolescents, respectively 84% vs 60% of 5years overall survival (OS). If complete resection is the gold standard in SS, other therapeutic modalities differ between pediatric and adult populations, considering SS as an intermediate chemosensitive tumor more frequently by pediatric oncologists. Prognostic factors evaluation (tumor size, site of primary and IRS group) is necessary to establish optimal treatment strategies, with multimodal therapeutic approach in children and adolescents. Thus, recent results about the European prospective EpSSG NRSTS 05 study for children and adolescent patients with localized SS showed a 5years OS >90%. Moreover, recent somatic genetic data about SS open the debate on an appropriate strategy based and stratified on tumor genomic. Multinational prospective pediatric, adolescent and young adult study is necessary to improve optimal and appropriate approach in this rare tumor.

Gigantism and acromegaly, usually caused by a pituitary adenoma linked inappropriate secretion of growth hormone (GH), are generally considered as very rare diseases, even if, according to some authors, their cumulative prevalence is about 1/5000. Starting from the historical case of a giant from Liège we shall describe the different types of GH pituitary adenomas and their pathophysiology. We shall particularly discuss rare forms of inherited GH secreting pituitary adenomas like the FIPA (familial inherited isolated pituitary adenomas) and the X-LAG (X linked acrogigantism), both described for the first time in Liège, in 2000 and 2014, respectively.

Fine needle aspiration is the gold standard method to differentiate benign thyroid nodules from malignant. However, for 15 to 30% of the cases the cytological diagnosis is indeterminate, leading to surgery. Integration of new molecular markers is opening new perspectives in order to increase the diagnostic precision of thyroid nodules with an indeterminate cytology.

Tuberous sclerosis is an autosomal dominant genetic disorder that is characterized by epilepsy, mental retardation and facial angiofibromas. Usually, the disease is diagnosed in childhood but there are frustrates form of tuberous sclerosis with or without genetic mutation. This clinical case about a man who is diagnosed a colonic polyposis, a rectal adenocarcinoma and a tuberous sclerosis.

Gliomas are the most common primary brain tumors and include different diagnoses associated with a different prognosis. Histology remains the gold standard for the diagnosis of these tumors. However, pathologists may encounter diagnostic difficulties due to tumor heterogeneity or to the small size of the samples. Recently, major advances in discovery of molecular alterations of these cancers have led to the development of new molecular markers, some with a diagnostic role, others with a prognostic impact and / or predictive of therapeutic response. The testing of different molecular alterations such as 1p / 19q codeletion, mutations of IDH genes, p16 deletion, EGFR amplification or MGMT promoter methylation has been included in the daily practice in order to confirm the diagnosis, assess the patient prognosis and guide treatment choices.

Malignant melanoma is a highly aggressive cancer with a propensity for early metastasis. Melanocyte transformation results predominantly from oncogenic mutations in BRAF, NRAS or NF1 leading to constitutive activation of the MAP kinase pathway driving cell proliferation and second site mutations such as loss of CDKN1A, or PTEN or activating mutations in the beta-catenin pathway that allow escape from oncogene induced senescence. Nevertheless, irrespective of the nature of the driver mutations, melanoma cell physiology is strongly regulated by transcription factors and epigenetic mechanisms. MITF (Microphthalmia-associated Transcription Factor) and SOX10 are two major transcription factors that regulate both normal melanocyte and melanoma cell physiology. Using a combination of mouse genetics, biochemistry and high throughput genomics we have identified cofactors for MITF and addressed the mechanisms by which MITF, SOX10 and their cofactors regulate gene expression in melanocytes and melanoma.

Renal cell carcinoma is a rare pediatric malignant tumor of the kidney. Unlike Wilms tumor, the efficacy of chemotherapy and radiation therapy in pediatric renal cell carcinoma remains uncertain. Surgery is the best treatment and prognosis is favorable when the tumor is localized and completely eradicated. We report an exceptional observation in a 7-year-old girl with renal cell carcinoma who had been treated 20 months previously for Ewing sarcoma with chemotherapy and radiotherapy. The renal tumor was revealed by abdominal pain without hematuria. She underwent a radical nephrectomy, and histopathology concluded in renal carcinoma associated with translocation Xp 11.2 grade 3 of Furhrman pT3a N1. No adjuvant therapy was given. After 3 years of follow-up, there is no evidence of local or metastatic recurrence. This observation is significant given the very young age of this patient, the occurrence after Ewing sarcoma with a short disease-free interval. It seems that translocation renal cell carcinoma is associated with previous exposure to chemotherapy, particularly topoisomerase II inhibitors or alkylating agents.