I congratulate Tanaka et al. [1] on their tour deforce treatment of unresectable pancreatic cancer, which included arterial embolization to achieve a single blood supply and subsequent arterial infusion of 5-fluorouracil. Pancreatic cancer has an overall 5-year cure rate of 6% [2], which drops to virtually zero when the tumor is unresectable. The authors report an estimated gain in survival of 9.8 months, with this figure somewhat mitigated by increased toxicity and possible bias in the selection of the 20 treated patients.

This article is an important contribution to the literature but I respectfully question the authors' final recommendation that “considering the high response rate and promising survival duration…phase 3 multicenter prospective randomized clinical trials…are warranted” [1]. The cost per quality life year saved for this intervention is difficult to estimate but, in my opinion, would be beyond anything that that most clinicians would consider or health care systems could afford.