Molecular Markers in Cancers and Precancers (MOCA)

It is in this context the prescription of targeted therapies is now conditioned by the identification of specific genetic anomalies in the tumour. As such, kinase inhibitors have been shown to be effective in patients with non-small cell lung cancer or metastatic melanoma when mutations in epidermal growth factor receptor (EGFR) or BRAF, respectively, are identified. Conversely, the identification of KRAS or NRAS mutation in metastatic colon cancers predicts resistance to anti-EGFR antibody-based therapies. Under these conditions, the prescription of these targeted therapies provides a major benefit to patients. The efficacy of these therapies is linked to the fact that the target mutations are "driver" mutations delivering a powerful oncogenic signal. These mutations are also an 'Achilles heel' for the tumour cell, which becomes hypersensitive to certain tyrosine kinase inhibitors. However, tumour escape from targeted therapies is well documented after a few months of treatment. Here again, molecular analysis of progressing tumours has revealed intratumoural heterogeneity with, in particular, the appearance of secondary mutations responsible, at least in part, for the development of resistance. This is why it is important to characterise the molecular profile of tumours both during the natural history of the disease and in treated patients in order to offer them appropriate follow-up.

Among the 170 human papillomaviruses (HPV) described in 2013, a dozen or so so-called high-risk or oncogenic HPVs are responsible for all cervical cancers, almost all anal cancers, half of vulvar and vaginal cancers and certain cancers of the upper aerodigestive tract. While HPV infection is necessary for cancer to develop, it is not sufficient and co-factors that promote persistent infection increase the risk of developing precancerous lesions and then cancers. Thus, the natural history of infection by these viruses is closely linked to that of the cancer they induce. The molecular mechanisms of HPV-related carcinogenesis/transformation are well described. It is the combined action of two viral proteins (E6 and E7) on the two tumour repressors p53 and pRb that initially lifts the intrinsic mechanisms of replicative senescence of the cell (which thus acquires the capacity to divide indefinitely), and then gradually leads to its transformation. However, the determinants that lead an infected cell to immortalise and then transform remain poorly understood and the vast majority of infections are eliminated spontaneously within 10 to 18 months following the development of effective immune responses. It is likely that host (immunosuppression, genetic factors), viral, and environmental (smoking, oral contraception) co-factors will influence the carcinogenesis process. Thus, it is recognised that HPV16 is the most carcinogenic genotype. It is the longest persisting HPV and is associated with the highest risk of developing pre-cancer or cervical cancer. Cervical cancers associated with HPV16 (or HPV18/45) have been shown to have a worse prognosis than those infected with other genotypes. Conversely, in upper aerodigestive tract cancers, those induced by an HPV (this is HPV16 in more than 95% of cases), have a better prognosis than those not induced by a virus. Thus, genotyping a tumour to identify the type of HPV involved could be of clinical interest, particularly depending on the location of the tumour.

The therapeutic management of patients with HPV-associated cancers most often consists of a combination of surgery and/or radiochemotherapy (cis-platinum, 5-fluorouracil) depending on the extent of the tumour. For anal cancers, work in Bisonne showed that the addition of a third chemotherapy molecule (taxane) was very promising as it allowed previously unobserved remissions. The reasons for such efficacy are not clear, nor is there a predictor of response to treatment.

Molecular genetic analyses are carried out using various types of samples such as cells from smears or punctures, biopsies, surgical parts and numerous fluids such as urine, cerebrospinal fluid or blood. While the standard for molecular diagnosis today is to analyse the tumour sample, the use of a "liquid biopsy" from a simple blood sample is widely considered. Indeed, it has now been shown that cancers release DNA that can be detected in the blood of patients, this is circulating tumour DNA. Thus, it is possible to diagnose or perform biological monitoring of cancers (e.g. before/after treatment) from a liquid biopsy. However, the study of circulating tumour DNA still faces some difficulties. Firstly, the concentration of circulating DNA is very low, in the order of a few tens of nanograms per mL of plasma. Furthermore, the vast majority of circulating DNA is composed of DNA released from normal cells and the proportion of circulating tumour DNA is only 1-4% of circulating DNA. Finally, circulating tumour DNA is generally fragmented (<200 bp). In order to overcome these constraints, it is necessary to use highly sensitive techniques, both to measure the concentration of circulating DNA and to search for molecular alterations characteristic of the tumour. In this respect, circulating tumour DNA analysis brings a new dimension to the management of cancer patients. Based on circulating tumour DNA analysis, it is possible to direct treatment towards a targeted therapy in the absence of a tissue biopsy, to assess the effectiveness of a treatment, to follow the evolution of the disease, and even to identify recurrence. Circulating tumour DNA analysis also provides a snapshot of all genetic alterations in the tumour (primary and metastatic) reflecting tumour heterogeneity, whereas biopsy results are only representative of the site from which they were taken. For HPV-associated cancers, viral genome detection from liquid biopsies is also largely feasible. However, studies are still needed today to not only validate the principle of liquid biopsy in cancers, but also to clarify its clinical utility. Recent results have shown that changes in plasma HPV viral load predict response to treatment.

Since the sequencing of the human genome in 2001, DNA analysis techniques have progressed enormously and the new 'next generation sequencing' (NGS) technologies allow the simultaneous analysis of a very large number of genes (several hundred) from several dozen different samples. These sequencing capabilities make it possible to rapidly explore a large number of genetic anomalies at a lower cost. Data on genetic abnormalities in HPV-associated cancers exist and it has been very recently reported that HPVs are also subject to genetic variations during the carcinogenesis process.

Technological advances in molecular biology (high-throughput sequencing, digital PCR, circulating tumour DNA) now make it possible to describe very precisely any genetic or epigenetic modifications that could constitute potential biomarkers. A better description of these genetic modifications in sequential samples during the transition between a normal state, a precancer and a cancer as a function of time will make it possible to develop, on the one hand, models predicting the appearance of cancers and, on the other hand, innovative tools for diagnosis and risk stratification of developing a cancer. It will be possible to propose early medical interventions only for those patients who need them. The study of genetic abnormalities in patients treated for cancer will make it possible to propose innovative tools for monitoring the disease, predicting relapse or cure or identifying new therapeutic or vaccine targets.

This is why the investigators want to be able to collect biological samples of various kinds (smears, biopsies, biological fluids, etc.), whether they have been taken in the context of care or stored in declared collections (tumour libraries, for example). These samples will be collected and stored at the Besançon University Hospital in the form of a collection (MOCA collection) which will be associated with clinical data. The investigators will then be able to build up homogeneous cohorts of patients from this collection from which the investigators can study theranostic biomarkers.