Last week, the FDA issued a warning against the use of robotic surgery for mastectomy and other cancer-related procedures.  The warning stated, in relevant part, “To date, the FDA has not granted marketing authorization for any robotically-assisted surgical device for use in the United States for the prevention or treatment of cancer, including breast cancer.” The warning went on to say, “Be aware that the safety and effectiveness of using robotically-assisted surgical devices in mastectomy procedures or in the prevention or treatment of cancer has not been established.”  This action came after the Agency reviewed certain literature suggesting that robotically assisted devices were being used “off-label” in cancer treatment.  Incidents like this raise a very important question: How can a company avoid or minimize liability in the event of off-label use?

As I see it, there are two main options.  One involves training, and the other, product design.  The training-based solution likely strikes the best balance between minimizing potential liability and limiting the possibility of creating friction with healthcare providers.

The training-based approach is relatively straightforward. Both sales and other company representatives should be carefully educated about the specific procedures for which the device received marketing authorization.  They should be explicitly discouraged from promoting the device for any other uses.  And written statements prohibiting the promotion of the device for off-label use should be prominently displayed in all training materials.  Ideally, those materials should be maintained in a readily accessible location so that they may be easily retrieved in the event of litigation.

As part of their training, sales and other company representatives should also be instructed on how to best respond when confronted with off-label use.  If, for instance, a surgeon asks a sales representative about off-label use, that representative should reiterate the device’s authorized uses and emphasize that the company does not condone off-label use.  Such interactions should be clearly documented in writing and stored in a centralized location that can be easily accessed if needed later.

The design-based solution—assuming that it is technologically feasible—is more complex, could create friction with physicians, and in some instances could even backfire and present plaintiffs’ counsel with a powerful argument against the company.

As part of this solution, the surgical robot could be programmed to only perform actions that are on-label.  Though this may prove effective at reducing off-label use, it would likely not be well received by the medical community, as some would surely argue that it interferes with their medical judgment.

Another option is to program the surgical robot to display prominent warnings when asked to perform an off-label action.  But the surgeon could then be allowed to continue if she so chooses.  This alternative, however, may prove problematic in the event of litigation.   A plaintiff’s lawyer could easily argue that by giving physicians the option to continue, the company is condoning a use that it clearly knows to be off-label.

Though there is no quick fix for the risks associated with off-label use, on balance, improved training seems to be the most effective approach. It specifically targets the company’s representations to the outside world, while allowing surgeons the latitude to freely exercise their medical judgment in the care and treatment of their patients.  Thus, the company avoids statements or actions that could be used against it during litigation without getting involved in the practice of medicine.