As more molecular tests become available that stratify patients into different therapeutic groups it is likely that those tests will have to be performed on small endoscopic biopsies, especially where there is the possibility of neo-adjuvant therapy. As yet there have been few studies which have looked at the feasibility of this – do we need to ask endoscopists to take extra biopsies if they believe the lesion is a tumour? Should samples for potential molecular testing be sent in separate specimen bottles? In this month’s Histopathology Hieke Grabsch and colleagues review all the available published studies in this area. Their conclusions are that there is no hard evidence indicating an optimal number of biopsies to be taken if molecular testing is anticipated and whether these should be processed separately. In my experience of sending duodenal biopsies away for T cell receptor gene rearrangement studies (to detect refractory coeliac disease) the usual two or three standard endoscopic biopsies in formalin-fixed paraffin-embedded blocks are sufficient but obviously this might vary with the nature of the molecular test.