'What do you think we should do ?'

You are called into a neighbouring endoscopy room to advice on this lesion at the recto-sigmoid junction.  
WHAT WOULD YOU DO NEXT?

■ Carefully assess and take samples

Yes but in what way would you 'more carefully' assess this?

■ Attempt to remove it now

Well, you would realise that it doesn't lift !

■ Attempt to remove it at a dedicated session

Noo! You missed it !

explanation

This lesion does look completely different when viewed in retroversion (image below).  In retrovertion its obvious that it's malignant. Clearly the correct way ahead is to take some better targeted samples and refer to the next colorectal cancer meeting. If you hadn't assessed the lesion in retrovertion, samples would have revealed a TA+HGD only (this is actually what happened), delaying the correct management. The moral of the story is: always view all of the lesion before deciding on where samples should be taken from.  

Aha! Are we not told that;  "polyps which will be resected endoscopically shouldn't be sampled as it makes removal difficult"?  The problem is that when patients have to wait long for their resection, it makes sense to take a few samples or at least do a 'test-lift'. Having to wait 3 month for a resection only to be told that the lesion is likely to be malignant is a disaster scenario. It's then better to exclude cancer as far as possible before asking the patient to wait months for the removal. Of course, its true that it can make a resection more difficult but it's hardly ever a showstopper !