Molecular subtypes of gastrointestinal stromal tumor requiring specific treatments

Abstract

This article reviews and discusses the current literature on how molecular subtyping of gastrointestinal stromal tumor (GIST) impacts decision-making in clinical practice. Genotyping has not yet been used for prognostication of localized GIST. Recent findings indicate that mutations in platelet-derived growth factor receptor alpha (PDGFRA) and stem cell growth factor receptor (KIT) exon 11 duplication mutations are associated with a favorable prognosis. Subgroup analyses of randomized trials provide first evidence on how genotyping predicts clinical benefit in the adjuvant setting. In the palliative setting, genotyping should be performed for dose selection but may also be relevant in patients who are intolerant to imatinib. For patients whose tumors harbor the notoriously resistant D842V PDGFRA mutations, novel inhibitors have entered clinical trials, which may end the therapeutic deadlock. Several novel targets have been identified that are currently being investigated in GIST, including inhibitors against fibroblast growth factor receptor, hepatocyte growth factor receptor, mitogen-activated protein, and phosphoinositide 3-kinase. Preliminary clinical data suggest that predictive markers will be needed to define sensitive molecular subgroups. Genotyping should be an integral part of clinical management of GIST as it aids in prognostication, prediction of efficacy, and selection of drugs and dosing. Genotyping reduces the risk of both unnecessary adjuvant treatment as well as undertreatment and overtreatment in the palliative setting.