Clinical application of liquid biopsies
Menée à partir d'échantillons plasmatiques prélevés sur 180 patients atteints d'un cancer du poumon non à petites cellules et non épidermoïde (âge médian : 62 ans ; 62 % de femmes), cette étude évalue la performance d'une technique rapide de génotypage de l'ADN circulant par PCR numérique pour détecter des mutations dans les gènes EGFR et KRAS
Precision cancer medicine refers to treatment planning based on detection of driver mutations. This field is currently in its infancy, with some noted successes, such as BRAF inhibitors for driver BRAF mutations in melanoma and EGFR inhibitors for driver EGFR mutations in lung adenocarcinomas, among a few others. There are increasing numbers of “basket” or “umbrella” clinical trials that rely on molecular analysis for eligibility.1 However, there are some major uncertainties in the clinical application of precision medicine. Research is still needed to obtain an agreement on the level of evidence that is needed for a molecular abnormality to be actionable that is beneficial to clinical decision making for the patient. For example, a BRAF V600E mutation in colon cancer leads to many fewer responses than for melanoma.2 Another potential barrier to the use of precision medicine is tumor heterogeneity. Different mutations may be found in different parts of the tumor mass, either at the primary site or in metastases at different sites.3 In addition, the application of precision medicine requires detection of molecular abnormalities in a tumor sample obtained either from an archived specimen or by biopsy. On the one hand, if the tumor sample was obtained months to years prior to the assay, the cancer may have developed additional mutations, as in the case of resistance to a targeted therapy. On the other hand, patients may have tumors that are relatively inaccessible to safe biopsy (eg, mid lung field, near a major blood vessel, or otherwise anatomically inaccessible to a core biopsy needle, or adverse clinical condition of the patient), and the biopsy itself may be subject to sampling error. It is also not clear whether an assay testing for the presence of a few variants vs hundreds of variants is more useful for clinical decision making.
JAMA Oncology , éditorial en libre accès, 2015