If you are among the people in America who agree with one or other of the following two statements, you need an immediate crash course in what approval of a new drug by the U.S. Food & Drug Administration (FDA) does actually mean:
- FDA only approves drugs without serious side effects.
- FDA only approves “extremely effective” drugs.
According to a recent survey conducted by Schwartz and Woloshin and just published in the Archives of Internal Medicine:
- 25 percent of 2,944 people surveyed did indeed believe that the FDA only approves drugs without any serious side effects.
- 39 percent of 2,944 people surveyed did indeed believe that the FDA only approves “extremely effective” drugs.
In fact, neither of these two things are true at all. So what is the truth?
In reality, FDA approval only implies that — after careful review — the agency has determined that “the benefits [of the approved agent] are judged to be greater than the harms. It doesn’t mean that they’re big and important,” said Woloshin in a statement to Reuters, which has also reported on this study.
Schwartz and Woloshin also assessed people’s perceptions about the relative value of newer and older medicines based on FDA approval. The two drugs used in this test actually had approximately equal efficacy and approximately the same side effects in the management of heartburn, but some 66 percent of those questioned picked the newer medication when asked to make a choice between the two.
In fact, in many cases, FDA approval is not based on direct comparisons of effectiveness and safety between drug X and drug Y, so there may be no good reasons to believe that one is any better or safer than the other. In the case of new cancer drugs, the FDA does try hard to encourage drug developers to carry out trials that compare a new drug to the “standard of care” at the time a new trial is being designed, but it is not always possible to do this, and events may overtake reality.
As an example, it would be nice to know whether MDV3100 is or is not safer or more effective than abiraterone acetate in the treatment of castration-resistant prostate cancer (CRPC). However, abiraterone acetate wasn’t approved at the time the Phase III trials of MDV3100 were being designed. At that time, the standard of care for any man with CRPC — metastatic or otherwise — was still chemotherapy with docetaxel + predisone. Unless either the survival benefit demonstrated by MDV3100 is a lot more or a lot less than 3.9 months compared to placebo, or the side effects of MDV3100 are a lot worse or a lot better than those exhibited by patients on abiraterone acetate compared to placebo, there will be no way to assess whether one drug is any “better” than the other. For very similar reasons, we have no idea of the relative merits of treating men with CRPC with either abiraterone acetate of sipuleucel-T.
The FDA does its best to ensure that important new drugs for conditions like prostate cancer are moved through the regulatory process in a timely manner. In recent years, they have become pretty successful at this. However, we are likely, over time, to discover than some of these new drugs have additional, and potentially significant, side effects that were simply not seen or reported in the relatively small numbers of patients who participate in the clinical trials that customarily lead to drug approval.
Here is a brief list of some very simple things that it is well worth understanding about what an FDA approval of a new drug actually does and does not mean:
- FDA approval means that, on average, the benefits of using a drug outweigh the risks in a well-defined set of patients with a specific disorder.
- FDA approval means that the FDA has worked closely with the drug developer to create detailed prescribing information for the new drug
- The prescribing information for every drug includes
- Information about what types of patients the drug is approved for
- Information about the effectiveness of the drug in this set of patients
- Information about the side effects of the drug when used in this set of patients
- Warnings about serious adverse effects (up to and including death) associated with the use of the drug
- Information about the recommended dose of the drug and how it should be given to or taken by patients
- FDA approval of a drug never implies that
- A drug is completely safe
- A drug has no side effects
- All the side effects of a drug have been discovered
- A drug will necessarily be effective for you as an individual
- On average, the benefits of using a drug will outweigh the risks if the drug is used to treat an unapproved condition
- A drug is “better” than other drugs for the approved condition (unless it has been compared to another drug or drug in a “head-to-head” clinical trial)
Filed under: Drugs in development, Management, Treatment Tagged: | approval, benefit, FDA, risk
Search for new and
ongoing trials on the
CTAG PCa web site
There is another issue about FDA approval that took me some time to understand and that is the use of ‘off list’ applications.
As I understand this now, once a drug has been approved by the FDA (on the basis of studies that demonstrate maximum tolerated dose, not minimum effective dose), it can be used for purposes beyond those for the approval study and indeed for patients other than those for which it was actually intended.
Perhaps you’d care to comment on the correctness or otherwise of this understanding?
Dear Terry:
Once the FDA approves a particular dose of a particular product for any one specific indication, in the USA a physician may (at least in theory) use that product at whatever dose level he or she choses and for whatever clinical condition he or she feels is appropriate based on his or her clinical judgement. An example of such a use is that of high-dose ketoconazole as a second-line hormone therapy in advanced prostate cancer. Ketoconazole has never been approved for this indication, and the dose level at which it is used to treat late-stage prostate cacner is way higher than the “approved” dose.
Now it is fair to say that there are all sorts of constraints on such “off-label” usage of drugs. Are there any data to justify such a use? Will the insurer cover the costs associated with such use? Are the perceived benefits associated with the use of the drugs justifiable in terms of the risks being taken? Is the specific off-label use considered to be “common practice? I could go on. And I am sure lawyers who specialize in such issues could go on and on and on! These decisions are not within the remit of the FDA. They also have nothing to do with the maximum tolerated dose (for the approved indication). After all, until comparatively recently the vast majority of drugs were very rarely tested in clinical trials in children, so the use of drugs to treat things like childhood leukemia and childhood infections were pretty much a matter of using drugs developed for treatment of adults in different ways and at very different dose levels.
Let us be clear that off-label use of drugs approved for other clinical reasons is common and accepted clinical practice all around the world (Australia and South Africa included). Many cancers and rare diseases are commonly treated “off-label” because there is no financial motivation for the developer of a specific drug to seek an approval for use in such indications, or no drug has ever actually been subjected to clinical trials because the numbers of patients in which the drug could be tested is too small.
Here in the USA — and in many other countries — what is illegal is for the manufacturer to actually promote the clinical use of a drug “off-label.” Again as an example, abiraterone acetate has been approved here in the USA for the treatment of metastatic, castration-resistant prostate cancer in men who have already received at least one course of docetaxel-based chemotherapy, and Johnson & Johnson can only promote its use for that indication. However, there is clear evidence that it is already being used off-label by some physicians to treat men who have not had chemotherapy and in men who are castration-resistant but not metastatic. Whether these uses are being covered by insurance companies is less clear to me at this time.
Thanks, Mike. I wasn’t being critical of a practice that is so widely followed, but it does seem to me to be odd within the other generally accepted practices of not supporting certain views because of a lack of appropriate studies.
As you know one of the things that puzzles me is why ADT is not used as a potentially curative therapy for early-stage prostate cancer. One of the stock responses to any such suggestion is that there are no prospective, double-blinded, published studies to justify such an approach.
Fair enough, but … why does the “off-label” approach become unacceptable in these circumstances? Why are doctors who may have tried such an approach ignored or even vilified when they try to publish the results of what are essentially experiments? That’s what doesn’t make sense to me.
And then, of course, there are the drugs that do work, but aren’t approved — like the chemopreventatives.
Terry:
I think the answer to that one is pretty straightforward … and it is not about the “off-label” issue at all.
The consensus among “experts” (whatever an “expert” is) — rightly or wrongly — is that the risks associated with using ADT early in the treatment of localized prostate cancer significantly outweigh the benefits. We do know that the combination of ADT and external beam radiation therapy provides a significant survival benefit compared to radiation alone and compared to hormone therapy alone in the treatment of men with locally/regionally advanced prostate cancer, and I have to say that I personally have yet to be convinced by any of the “experiments” to date that current forms of ADT alone are effective and safe treatments for men with truly localized disease (with the single, arguable exception of drugs like dutasteride). Indeed much of the data on the use of ADT as first-line therapy suggests that it is significantly less effective than many other forms of treatment.
Which “experiments” do you think are compelling?
Mike,
I was not aware of any studies or references to “experiments” with intermittent ADT and/or low-dose ADT (to keep potential side effects to a minimum) for early stage prostate cancer. I am not sure if this is because there have been none or because they have not been accepted for publication in a peer-reviewed journal?
My puzzlement derives from the logic that says ADT can control and even destroy small colonies of prostate cancer cells — so use ultrasensitive PSA tests after surgery to hit them as soon as they show their heads above the trench. So why not hit small colonies detected with current technology early and sooner.
The drugs I take to control my heart condition are never going to “cure” me; they have some potential for poor side effects; they allow me to continue to live where I might not have done so; that’s managing the disease. Why don’t we try to manage early-stage prostate cancer instead of hitting it with maximum force first. One of the “cures” for cardiomyopathy is a heart transplant. That wasn’t the first tool my cardiologist reached for.
Terry:
A major difference between the drugs used to manage your heart disease (and mine too) and ADT as a treatment for prostate cancer is that most of the drugs commonly used to manage such heart disorder do not come with a significant risk for accelerating progression of the disease.
The use of ADT is associated with a significant risk for stimulation of the development of androgen-resistant prostate cancer cells. That risk is initiated as soon as a patient with prostate cancer starts to take ADT. That risk is also one of the reasons why the FDA and other regulatory authorities have been so concerned about approving dutasteride and finasteride for the prevention of prostate cancer (although the evidence that induction of androgen resistance resulting from the use of these drugs is much less well defined).
We really do not know what percentage of men who start getting ADT may have their disease accelerated by this type of therapy. All we know is that — on average — significantly more men with metastatic prostate cancer benefit from ADT than those who do not benefit. However, the use of ADT in men with early stage, low- and intermediate-risk disease who might otherwise be managed safely with active surveillance or almost any type of first-line therapy could make such men hormone-refractory very early on. We also know that there was no benefit observed to the use of neoadjuvant ADT in association with surgery as a first-line therapy for localized prostate cancer. This was tried extensively in the early to mid 1990s, and a good deal of data was published on this subject. It just made the surgery more difficult and there was no signal that it decreased the probability for biochemical recurrence post-surgery.
I acknowledge that I am not aware of any significant, randomized clinical trials of ADT (intermittent or otherwise) in the treatment of early-stage prostate cancer. However, there are extensive data suggesting that the use of ADT alone as a first-line treatment for prostate cancer is associated with a minimal survival benefit and a significant risk for the more serious complications of ADT.
FDA approval requires a significant overall survival time benefit. For docetaxel this was 2.4 months (later on adjusted to 2.9 months).
Nothing with respect to pain relief, PSA-reduction and PFS (progression-free survival). This makes sense only if the treatment concerned is the last treatment available to a patient. But new drugs are being developed, for example for docetaxel-resistant patients.
As a consequence of the FDA approval requirement, all new trials have OS as their primary endpoint.
Dear Arnold:
I am sorry but your statement is somewhat misleading.
The FDA is willing to consider a number of different types of endpoints for approval of drugs in the treatment of prostate cancer. For example, denosumab (Xgeva) was recently approved on the basis that it was able to prevent skeletal-related fractures, and Amgen has also submitted data for an approval of denosumab for delay of onset of evident metastasis in men with non-metastatic, castration-resistant prostate cancer. (Whether the FDA will approve denosumab for this indication is not, however, known yet.) Not so long ago, the FDA also approved degarelix (Firmagon) because it was able to lower PSA in men with hormone-sensitive prostate cancer to undetectable levels faster than any LHRH agonist and with at least a comparable side effect profile. There was no consideration of a survival benefit at all.
The FDA also regularly approves drugs based on end-points such as progression-free survival. It has not done this in prostate cancer for a long time, because the last sponsor company to seek an approval for this endpoint failed, and because sponsor companies have got the message that if a company wants an indication suggesting that a product is capable of extending life the FDA very definitely prefers to be able to approve such drugs based on overall survival data. There is no absolute refusal on the part of the FDA to approve drugs on the basis of progression-free survival. It is a validated endpoint in late stage trials for almost all forms of cancer therapy.
It is also worth noting that many products are getting approved based on relatively small extensions of life, almost regardless of the adverse effects. The FDA most certainly would be willing to approve any product that had the just same effect on survival in men with castration-resistant prostate cancer as (say) docetaxel — if it also showed that it could do this with few to none of the adverse effects of chemotherapy, i.e., comparable effectiveness with greater safety and superior quality of life.