If it's a liver cell in the pancreas, it's already ignoring normal cellular signalling. And in terms of replicating, these are derived from stem cells. Cancer is partly defined by cells becoming more stem-cell like and being able to replciate, so by definition they've already been turned into a state where they're predisposed to be cancerous (or basically are cancers by definition), and the coaxed to become a pancreatic cell, but unsuccessfully, so it's not a good place for it to go wrong.
It's a known problem though and not an insurmountable one. There's been a lot of work with encapsulation of these cells in a barrier that will prevent the barely controlled tumor that is an artificial pancreas from breaking out and wrecking havoc. And methods that are getting better at controlling these cells like the one in this thread are promising ways of mitigating this risk moving forward as well.
It's a known problem though and not an insurmountable one. There's been a lot of work with encapsulation of these cells in a barrier that will prevent the barely controlled tumor that is an artificial pancreas from breaking out and wrecking havoc. And methods that are getting better at controlling these cells like the one in this thread are promising ways of mitigating this risk moving forward as well.