Monday, June 16, 2008

Access to Experimental Drugs for Dying Patients

Before I went to the ASCO meeting, I read a fascinating interview conducted by Dr. Val. She interviewed Dr. Emil J. Freireich, the director of the Adult Leukemia Research Program at M.D. Anderson Cancer Center. The discussion took place at a press conference announcing the introduction of the Access, Compassion, Care and Ethics for Seriously Ill Patients Act, which seeks to increase terminally ill patients’ access to promising investigational drugs. Dr. Freireich made some excellent points about the risk-averse nature of the FDA’s drug approval process, and how this process slows the development of new drugs.

Dr. Freireich’s interview made me think more about the process of experimental drug approvals, and the pros and cons of allowing patients access to investigational drugs before they are proven safe.

One interesting point Dr. Freireich made is that, unlike any other scientific research, research on new drugs requires government approval before it even starts. If I want to research the mechanism by which a cancer-related gene contributes to the growth of a sarcoma, I don’t need anyone’s permission to begin the experiments. I just do them. On the other hand, if I want to test a new cancer drug, I not only need the approval of the hospital’s Institutional Review Board (which is charged with protecting patients from unethical medical research), but I also need the approval of the federal government, in the form of the FDA.

Why is this important? Well, the folks who decide which drugs get to be tested are rarely physicians. So they often lack the expertise to appropriately judge the degree of risk posed by a new drug against the risks faced by the patient.

What do I mean by that? I’ll give you an example from my own career. I have been conducting research using a drug called samarium-153-EDTMP for patients with high risk osteosarcoma. This drug works by delivering radiation through the blood stream to places where the osteosarcoma is involving bone. The FDA approved my research, but did not allow me to treat anyone under the age of 13. Why this restriction? Because the drug also delivers radiation to growth plates, the parts of bones where growth occurs. There is a concern that growth plates might be damaged, and if I treat growing kids, they may stop growing. I argued, quite vigorously, that these are kids who are all destined to die of their disease… they represent the osteosarcoma patients with the worst prognosis, and very few are cured. So, to deny them access to a drug that could help, because it may stunt their growth seems… absurd. I lost the argument, though, and all subjects on the study are 13 or older. M, who came to me from Japan for treatment, was not allowed on the study because he was 12. He died several months later.

So why not let terminally ill patients take whatever drug they want? Why limit access at all? Some might argue exactly this point, saying that if the patient is going to die anyway, why not let him/her take whatever they choose. Well, what if an experimental drug hastens death or increases suffering? If a patient has a good quality of life, and an unscrupulous person offers an experimental drug that is not only ineffective but hastens death, depriving the patient of time that could have been spent with friends and loved ones enjoying life, that’s a problem. Or what if the drug is not only ineffective, but causes horrible and unmanageable side effects without hastening death? Surely that is also a bad outcome. Of course, what if the drugs extend the patient’s life with minimal side effects? Shouldn’t patients be allowed to make that gamble even if there isn’t enough research supporting this possibility?

And aren’t there other benefits from “trying everything” beyond extending life? For example, my patient M enrolled on a Phase I study. In my post about his family’s decision making, I mentioned that only 3% of patients in Phase I studies experience any benefit, but then talked about how benefit can mean so much more than “the tumor got smaller.” Benefit can mean feeling at peace with a decision, and feeling a sense of closure. These are real benefits that patients can get from investigational drugs, even if their tumors don’t shrink.

These are issues that come up in my practice all the time. When standard treatments no longer hold any hope of cure for my patients (who are all children or young adults), they and their families become desperate for something, anything, to try. I know how hard it is to say “No” to someone like that, especially someone I’ve treated for years and to whom I have grown very close. But I also know that I vowed to “Do no harm.” And that means making sure that I’m not just pulling something off the shelf to “give something.” Of course, I suggest as much as I can… as much as makes sense and has some scientific backing. I’ve certainly treated patients “off label” (using an FDA approved drug for a different indication) when it has made sense to do so.

So what’s the answer? This is undoubtedly a controversial subject, and a fertile topic for discussion when it comes to medical ethics, so there is no easy answer.

But certainly the drug approval process has to speed up. Between 1948 and 2002, there were 120 new cancer therapies approved by the FDA, but of these only 15 have any pediatric information in their approved product labeling at all. None were specifically approved for use in children. However, drug development is not without risk, and there are worse things than dying. Hopefully, someone will come up with a good solution, balancing a recognition that drug development (especially to treat cancer) requires risk with appropriate safeguards for these very vulnerable patients.

2 comments:

Dr. Val said...

This is one tough issue, for sure. I was intrigued by what Dr. Freireich said: "Regulators have no incentive to approve drugs for trial because of risk aversion. I could create a cure for cancer, but if one person dies in a trial, then they'd fire the FDA guy who approved it."

The other problem is that clinical trials do not want to enroll terminally ill patients (generally) because they're more likely to skew the results of a potentially helpful drug towards the failure side of the equation. If your kidneys are already shot (for example) then even if you're given a life-saving drug, the chances of you surviving are low - and the chances of people erroneously believing that the drug didn't help are high. And that doesn't help anyone.

Personally, I think that terminally ill patients should be allowed access to drugs that have shown promise in stage II and III clinical trials before they've been approved by the FDA. It wouldn't be fair to require insurance to pay for them, but the drug companies could make them available at cost for compassionate use, and write off the fees.

That's a controversial stand - but if the patient is fully informed of the risks and benefits, then I don't think they should be denied treatment if they can pay for it. If the Novartis exec had the chance to use Gleevec prior to FDA approval, he might still be with us.

Doctor David said...

Hi Val,
You're right... a very tough issue, with no easy answers. I'm not sure I agree with Dr. Freireich's pessimism in the line you quoted, though. There are deaths in cancer trials all the time, but drugs get approved. The safety bar is set lower for cancer drugs because of the dire nature of the illness being treated. But, as with samarium, that doesn't mean there aren't unrealistic safety concerns that slow the process to a frustrating crawl!

You're also right about organ function. That's a particularly tough situation, since when evaluating a drug's safety it's important to distinguish safety in a healthy patient from safety in a sick one. Many of us try to write trials that allow in patients that reasonably reflect the population we intend to treat, so allowing some organ dysfunction, just not severe problems. It's an imperfect solution, but a workable compromise (when the FDA allows it).

I think the biggest problem is not with drugs in Phase II or III testing, but drugs in Phase I. Drugs at that stage of development have no data supporting efficacy and we often don't even know the right dose to give, so giving them to a patient off-study is impossible. For drugs in Phase II or Phase III, I have usually been able to get compassionate exemptions for my patients that don't meet the trial criteria when use of the drug seems sensible.

Anyway, I'm glad you posted that interview. I love reading posts that stir up some controversy, new ideas, and real conversation.