Would you extend pembrolizumab for over 2 years in a patient with MSI-H pancreatic ductal adenocarcinoma with now-stable disease?

I usually don't, but admittedly, this is a "data-free zone" in the pancreas. I would just stop and watch carefully with scans and tumor markers and have a low threshold to resume or biopsy if it looks like there is progression.

I agree with Dr. @Dr. First Last that data in this area for pancreatic cancer is limited. The approach is in many times picked through extrapolation from other cancers. Possibly the decision depends on the patient's and the cancer's behavior. If patients struggle with the extended treatment mentally...

Rare incidence in question. I wish there were more data to support decision-making, and ctDNA monitoring appears to be a promising biomarker guidance in this management, akin to many other clinical settings of pembrolizumab use beyond 2 years or not.