Jean Perrin Center

Breast cancer is diagnosed by imaging at a non-palpable stage in more than half of all cases. Surgical removal requires preoperative guidance. Generally, preoperative guidance is performed using a metal guide under local anaesthetic and radiological control. This type of guidance has several limitations. For the patient, it can be painful and traumatic. The procedure involves two departments: radiology and the operating theatre, which poses logistical constraints. What's more, between 10% and 40% of patients require repeat surgery for unhealthy margins, raising the question of the effectiveness of the tracking procedure. The investigators propose to develop a non-invasive intraoperative guidance system: Augmented Reality, which will provide a 3D vision with virtual transparency of the breast during surgery, thanks to real-time fusion of preoperative imaging with video from a camera located in the operating room. The process is illustrated below.
Illustration of the general principle of the augmented reality system for locating non-palpable breast lesions. The images above represent a preliminary test carried out on the computer outside the operating room. This is an initial research prototype which has not yet been validated and is not suitable for routine use.

Only 20% of familial uveal melanomas are explained by a hereditary predisposition, implying the presence of as yet unknown hereditary predispositions. This hypothesis is reinforced by epidemiological studies revealing an excess risk of prostate cancer, thyroid cancer and leukemia in patients who have developed uveal melanoma, even though these cancers are not part of the tumor spectrum of known hereditary predispositions to uveal melanoma (BAP1, MBD4). The identification of new candidate genes, once validated, would enable us to offer these families appropriate surveillance.

Prostate cancer is the most common cancer in men. Its incidence is rising as the population ages. In the localized stage, the 5-year overall survival rate (OS) is 98%. Metastatic progression and resistance to castration have a negative impact on prognosis. Despite recent advances in management, the 5-year OS is around 30%. Therapeutic advances in this indication have been made mainly by the use of taxanes and second-generation hormone therapy. These treatments have improved OS and progression-free survival (PFS). They are now used as standard therapy.
More recently, the Phase III VISION trial confirmed the improvement in OS and radiological PFS achieved by treatment with the radioligand 177Lutetium-PSMA-617 (Lu-PSMA) in patients with advanced metastatic castration-resistant prostate cancer (mCRPC).
This treatment is currently available in early access in France. Despite encouraging results, 40% of patients will not respond to Lu-PSMA, and there are currently no validated predictive factors. Studies are currently on going, but the identification of biomarkers seems necessary to better stratify risk in these patients.
Numerous tissue prognostic tests based on molecular characteristics or cell proliferation are emerging with this in mind. At present, molecular profiling is not a routine technique for prostate cancer, as it is for other solid cancers. At an early stage, the Decipher® Genomic classification tool has shown prognostic utility independently of therapeutic and clinico-pathological data.
According to recent studies, methylome analysis would enable the subdivision of mCRPCs and could help identify new therapeutic targets.
In the metastatic phase, certain molecular abnormalities involving DNA repair genes are predictive of response to PARP inhibitors.
Molecular analysis (mutations, copy number alterations, gene expression, DNA methylation) could therefore be useful in optimizing the management of mCRPC patients treated with Lu-PSMA.
If reliable molecular abnormalities are identified on tissue, a diagnostic technique based on circulating tumor DNA (ctDNA) analysis will be useful in decision-making for these patients. A biological collection will therefore be created during the course of this study, with a view to using ctDNA analysis in subsequent research.