Cancer

Imaging and genetics for new treatments

While the State has earmarked nearly €750 million for Cancer Plan II by 2013, including 15% for studying the related environmental and behavioural risks, CEA is investing in two main areas: molecular imaging and cancer genetics. Keys for improved diagnoses and new treatments.

Every year, cancer causes more than 145, 000 deaths in France, making it the leading cause of mortality according to INCa, the French national cancer institute. Cancer is different from other diseases; it carries a very strong emotional charge and is perceived as far more disturbing than other diseases with a similar prognosis. Cancer covers so many different aspects that we should really refer to it in the plural. Although most cancers obviously exhibit common cell mechanisms, there are many different triggering factors. Genetic, epigenetic, environmental, behavioural and infectious factors can all play a role. CEA's research is particularly concerned with finding out more about the causes of cancer and the associated predisposition factors. “Two particular topics are central to CEA's research effort,” says Paul-Henri Romeo, Director of Institut de Radiobiologie Cellulaire et Moléculaire (IRCM): “imaging and genetics as applied to cancer. Both topics are clearly essential for diagnoses and targeted, personalised treatment”.

Europe's leader in molecular imaging

CEA's Institut d'imagerie biomédicale (I2BM) is the leading centre for in vivo imaging in Europe and possesses a technical platform that is unrivalled in France. The institute is involved in several national and international programmes through Service hospitalier Frédéric Joliot (SHFJ) in Orsay. It concentrates on two particular activities as part of its work to develop molecular imaging: a) testing the effectiveness of new therapeutic approaches using contrast agents that can highlight cancer cells and b) designing labels for drugs to see whether they target organs, tumours or metastases.
In this area, researchers from SHFJ and CNRS have just used Titane, one of the most powerful supercomputers in the world, installed at the Centre de calcul recherche et technologie in Bruyères-le-Châtel, to simulate the treatment of two cancer models (lung and brain) by hadrontherapy using carbon-12 ions, and on-line monitoring of dose delivery by positron emission tomography (PET). During this unique “digital experiment”, the various processes involved were modelled in full, from the production of incident carbon ion beams and their interactions in the “digital patients”, up to PET detection of positron-emitting isotopes by radiation beam, with online and in vivo therapeutic monitoring. This represents a significant step forward towards the optimisation of treatment plans based on hadrontherapy and used in conjunction with effective therapeutic monitoring. Biomarkers are increasingly used in characterising and grading tumours and, therefore, in stratifying patients for individualised therapies. A joint INSERM-CEA team at iRTSV*, which has been carrying out molecular analyses on cells and tumour tissues, has recently revealed that protein kinases are major targets for new cancer drugs. After screening collections of chemical molecules, the team identified several specific inhibitors of the protein kinase CK2, an enzyme that is abnormally active in most human cancers examined (breast, colon, lung, leukaemia, melanoma, glioblastoma). These molecules are currently undergoing evaluation in preclinical tumour regression trials. This protein kinase could also serve as a new clinical marker in various types of cancer. “During a survey of 111 adenocarcinomas in the prostate, for example, a polyclonal antibody, which had first been validated using immunoblot and immunofluorescence techniques, was used to carry out an immunohistochemical study of CK2 expression at the proteinic level,” explains Paul-Henri Romeo. “This led to the definition of a sub-population, consisting of patients in whom CK2 was both overexpressed and abnormally relocated in the nuclear compartment of cancer cells. It is important to note that these two events have been shown to correlate statistically with poor prognostic factors. The study provides the first evidence of the existence of a positive correlation between CK2 expression and location and anatomo-clinical data. We now have a unique antibody that could be put to good use as a tool for diagnosing and predicting prostate cancer.” An industrial partnership has been signed with a company called Covalab to extend the analysis to other types of cancer. The goal is to use a monoclonal antibody to target the same epitome in immunohistochemical studies of histological sections obtained from tumour samples. The properties of this antibody mean that it could be used as a reagent that should be of interest to clinical research laboratories.

Genetics and predisposition to cancer

Another CEA team, whose work is aimed at learning more about the mechanisms involved, has made significant progress in the study of cancer stem cells (CSCs) in cases of T-lineage acute lymphoblastic leukaemia. Studies performed in vitro and on immunodeficient mice have pointed to enhanced CSC activity in some blasts, thus opening up possibilities for new treatments and specific tools. Another laboratory has revealed the existence of human tumour stem cells exhibiting an atypical form of telomere maintenance. This could have implications in the treatment of certain types of brain cancer.
CEA's activities relating to genetics are concentrated particularly in the organisation's Centre national de génotypage (CNG), which has been involved in identifying new molecular mechanisms and discovering genes that could have an impact on the severity of certain diseases.
In this field, teams from CNG, Fondation Jean Dausset and the International Agency for Research on Cancer have identified new markers indicating a predisposition to bronchial carcinoma. Their results are the fruit of the most extensive genetic study ever carried out and represent a significant step forward in our understanding of this disease. Researchers from 19 countries studied DNA variants in more than 10,000 individuals, 5,000 of whom had lung cancer. They detected a region in chromosome 15 where recurrent genetic variations associated with bronchial carcinoma are located. The genes in this region encode the nicotinic acetylcholine receptors, which are found in the bronchial epithelium and which have already been identified as playing a role in nicotine dependence. These results are promising and should help improve our understanding of what causes cancer and lead to new therapeutic targets. “Simultaneous progress is being made in research on kidney cancer, glioblastoma and leukaemia,” says CNG Director, Mark Lathrop. “The idea is to obtain a broad overview of the genetic factors of predisposition to these diseases as soon as possible, and find similar risk factors for other disorders.”
Researchers at IRCM revealed one cause of sporadic (i.e. non hereditary) breast cancers when they demonstrated that the AKT1 protein is overexpressed in 50% of cases studied. As a result, the BRCA1 protein is sequestered in the cytoplasm and is unable to penetrate the nucleus, which prevents it from fulfilling its DNA repair function. It is as if the cell had no active BRCA1 gene, although the gene in question shows no signs of mutation, unlike in heriditary forms, where it does. The researchers also assumed the existence of a predisposition to breast cancer that might be due to hormones used in treating menopausal symptoms. If the existence of this predisposition is confirmed, then no hormonal treatment should be prescribed before studying the activity of the AKT1 protein in the blood. A study involving 100,000 women is planned at the beginning of 2010, in collaboration with Institut Gustave Roussy (IGR), to identify those undergoing hormone treatment in whom AKT1 is indeed activated. The primary objective is to detect any correlation between AKT1 activation and a predisposition to breast cancer. Another objective would be to develop a method for carrying out quick, simple and inexpensive assays of blood samples. The same joint CNRS-CEA team, composed of leading radiotherapy experts, is seeking markers that could help doctors to predict radiosensitivity (for which onset is acute in 5 to 9% of cases). The aim is to prove the existence of proteins that are common to all cases of radiosensitivity and could serve as predictive markers. In this area, too, collaboration with IGR will be put to good use as of next year to carry out tests on patients and seek real-life applications.
* Institut des recherches en technologies et sciences pour le vivant.