Hi Ellie, you ask great questions, to which I can offer thoughts but not definitive answers. Forgive me if i don’t try to define ethics and instead stick to discussing non-trial access to investigational medical products!
You have raised two separate, but related, questions. One, what happens when the provider and the patient have different stances on seeking an investigational medical product —who “wins”. And two, when in the development process ought we, and ought we not, think ethics requires us to provide an investigational medical product to patients.
So – re 1, you can look at this in 2 ways. On one hand, the physician has a fiduciary responsibility to her patient, and she is obligated to use her knowledge and training to guide the patient in evaluating the possible risks and benefits of trying an investigational medical product. On the other hand, the physician has an obligation to advocate for her patient. Ideally, helping the patient evaluate risks and benefits will result in a decision that the physician is comfortable advocating for. If, however, after helping the patient evaluate risks and benefits, the patient makes a decision that the physician disagrees with, the physician has two options. One, the physician can decline to assist the patient seek her goal (in this case, access to an investigational medical product) and offer to refer her to another physician if the patient so desires. Or, two, the physician can agree to file a request on behalf of the patient but, in so doing, make it clear that she does not agree with the patient’s decision. While it may seem like passing the buck to contact a company and say, “While I don’t think it is a good decision, my patient would like to seek access to your product,” it may allow a physician and patient to continue their relationship, both secure in the knowledge that they have made their opinions known. [Asking a physician to work to secure the object of her patient’s desires even while opposing the patient’s plan to use an investigational medical product is awkward for all parties and ideally should be avoided. However, given that a patient must use a physician to request access to the investigational medical product and that the patient may be at the end of her life or otherwise in a position where it would be difficult to find a new physician, physicians may sometimes be wiling to assume this role —provided, again, that they can be frank in telling the patient, company, or whomever that this is not something the physician is recommending.]
The 2nd issue you raise is when in the development process ought we, and ought we not, think ethics requires us to provide an investigational medical product to patients outside of trials. I believe it is impossible to develop a uniform guideline that will work across all diseases, all agents, and all populations; rather, I think this decision most likely would need to be made on a case by case basis. My chief concern is not, therefore, in drawing a line in the sand after which patients ought be able to access an investigational product. Rather, my chief concern is fairness: if one person can access the investigational drug after point X (be that the end of Phase I, Phase II, or some other moment during the drug development process), other similarly situated persons ought be considered for access at that point as well. So, for example, allowing a politically connected adult patient with breast cancer to access the drug ahead of other adults with breast cancer is unethical. Allowing an adult patient with breast cancer that expresses a certain protein to access the drug while denying it to another adult breast cancer patient whose disease does not express that certain protein is acceptable if there is reason to believe that the investigational drug works on that protein. To the extent that patients are alike in scientifically or medically relevant factors, they ought receive access to the drug at the same time. Non scientific or non medical factors ought not impact when a patient has access to the investigational product.
I have been involved in a pilot attempt to create an ethics-driven process for recommending which patients get access to investigational drugs that may interest you. It is called CompAC, the Compassionate Use Advisory Committee, and it has been created by New York University School of Medicine to assist Janssen. There are currently some publications out about CompAC and there will be more. My colleagues in this endeavor and I would not claim that that we have developed THE way to decide who gets access and who does not, but we have been working with an independent, expert group to try to make these tough decisions and to do so by relying on principles that will be made public at the end of the pilot.

Alison can you give us your thinking on risks and best practices?