Botox for Migraines? Let"s Read the Package Insert
In December 1991 the FDA cleared Allergan's prescription drug Botox® (botulinum toxin A, "Botox") to treat crossed eyes (strabismus) and involuntary eyelid clenching (blepharospasm).
Since then the label for Botox has been expanded to four or five additional therapeutic indications, most recently to treat chronic, severe migraines.
First identified in 1897, botulinum toxin has become what Allergan's executive VP for R&D calls a "pipeline in a vial.
" Ka-ching! What it really means when the FDA approves a drug or expands the number of uses for it, is that the drug's manufacturer is allowed to market the drug for that use.
However, physicians are free to use any approved drug for any use of their choosing -- the FDA does not regulate the practice of medicine.
In fact, there were hints of Botox's potential benefit to treat migraine over ten years ago and some off-label use for migraine has occurred over the years.
Now, Botox is formally approved for that use, what professionals call an indication.
So, the only question remaining is how well does it work? Fortunately, the FDA.
gov web site contains a treasure chest full of technical literature gems called Package Inserts (PI).
The PI is included with shipments of every approved drug so that physicians (and you and I) can read all about it.
To get your free copy of almost any PI, just click on Link 1 at the end of this report and type in the name of your favorite drug.
Occasionally, the drug's manufacturer will post the PI prior to the FDA.
Then, just go to the company web site and search for the drug under Our Products or something like that.
Such is the case for Botox.
So from Link 2 at the end of this report, at the bottom of the linked page, click on "Please see full BOTOX Product Information," which will enable you to download a PDF of the most recent PI for Botox.
Now, let's read the PI together.
Yikes! This PI contains 25 pages of giant words written in tiny print.
You may ask "Does every physician really read every word of every PI for the drugs she prescribes?" Did you read every word of every article in today's newspaper? The key is to read about what you need to know.
What we need to know is how well Botox works to prevent migraines.
The first page-and-a-half of the PI is a bioquickie version, what is called the "Highlights.
" Skip it.
Go right to the Full Prescribing Information section on page two.
Drugs with a particularly serious risk profile may have a so-called Black Box Warning right up front.
Botox has one and so does the antidepressant Zoloft.
Lipitor, for high cholesterol, does not.
Surprisingly, many biotech drugs, recombinant versions of natural proteins, also have Black Box warnings.
These warnings put physicians on high alert for certain side-effects; but, the understanding is that for the right patients, the drug's benefits trump its risks.
If, over time, the risks seem to be more prevalent or new risks appear, a drug may get a more ardent Black Box warning or it may even be withdrawn from the market.
But many drugs with Black Box warnings have been on the market for many years.
To repeat, the question we want to answer is how well does Botox work for migraines? Well, there are all sorts of fun information in this PI, but to answer our question, we need to find and focus on Clinical Studies.
Please jump to page 14.
First, some background in migraines.
Several drugs, such as GlaxoSmithKline's Imitrex and Merck's Maxalt, are FDA approved to treat a migraine attack.
There are also several drugs, such as Johnson & Johnson's Topamax and Abbott's Depakote, that are FDA approved to prevent migraines.
Botox was approved for prevention.
But wait, there's more.
One could try to prevent a migraine episode or try to decrease the number of headache days per month.
This distinction is exactly where Allergan ran into a problem.
The company sponsored a pivotal clinical trial called PREEMPT I which randomized 679 patients to receive either two rounds of Botox injections (31 shots given at seven specified locations, see p.
4 of the PI) or two rounds of placebo injections.
Each injection cycle was given 12 weeks apart and the primary clinical endpoint (PCE) was the change from baseline in headache episode frequency.
The result was that there were 5.
3 fewer headache episodes per month, which isn't bad.
Unfortunately, that was in the placebo group.
There were 5.
2 fewer episodes per month in the Botox group, essentially a wash.
Given that all patients started the trial with 12-13 episodes a month, a roughly 40 percent reduction in migraine episodes is certainly clinically meaningful, just not Botox dependent.
Sadly, none of the data in this paragraph are in your PI.
The data were published in a professional journal, Cephalalgia, in March of this year.
Here's the rub.
All PIs are composed by the company hoping to sell its drug and edited by the FDA.
A Package Insert is a negotiated document.
Most people, including migraine sufferers, do not read the scientific literature, let alone Cephalalgia.
Before now, I had never even heard of Cephalalgia, but I don't get out much either.
Now, back to page 14 of our PI.
The paragraph at the bottom of the page tells us how the clinical trial was conducted.
On page 15 is Table 6 with the data we need.
Note that two studies are presented, each with almost 700 patients evenly split between Botox and placebo.
And, note that the efficacy endpoint described was the change from baseline in the frequency of headache days per month.
The measurement of headache days was a secondary clinical endpoint (SCE) in PREEMPT I.
In general, a trial that hits a SCE or multiple SCEs but misses its PCE is a bust.
Maybe the company should have run another clinical trial.
Allergan took a different approach.
With the PREEMPT I writing on the wall, Allergan revised its second pivotal trial, PREEMPT II, and made the PCE the change from baseline in headache days.
Voilà, success! Curiously, in this second trial, there was also a statistically significant decrease in migraine episodes which was now a SCE.
However, the improvement in episodes was not due to a better result with Botox -- the number of episodes per month dropped by 5.
3, similar to that in PREEMPT I -- but due to a weaker placebo effect.
In PREEMPT II, the decrease in the number of migraine episodes due to placebo was 4.
6 days per month, compared with 5.
3 in PREEMPT I.
The joke on Wall Street was that PREEMPT I clearly used "extra-strength" placebo.
You gotta love their sense of humor.
Using Table 6 we see that in PREEMPT I, the patients receiving Botox had 1.
4 fewer headache days per month than those receiving placebo (7.
8 minus 6.
4).
Likewise, in PREEMPT II, there was a 2.
3 day reduction in headache days per month favoring Botox over placebo (9.
2 minus 6.
9).
On average, Botox reduced the number of headache days per month by about two days compared with placebo.
As for migraine episodes, the data for Botox are equivocal since the two trials had different results.
It seems to me that Allergan had every reason to choose migraine episodes as its PCE in both studies and was right in doing so initially, as in PREEMPT I.
It also seems to me that changing the endpoints to follow the data is not quite right; a bit like a dog chasing his tail.
But that's my opinion.
One last point.
In both clinical trials, patients had experienced about 20 headache days per month before being randomized to Botox or placebo.
That tidbit is in the two Cephalalgia papers, not in the PI.
That means that patients getting two rounds of Botox to prevent migraines experienced a benefit, net of placebo injections, of about ten percent (two fewer headache days per month).
And that's how well Botox works for migraines.
Plus, now you have been introduced to reading the Package Insert.
Postscript: There is much more information in the PI that I did not cover.
There is information on adverse events (AE), often called side-effects.
And there are AE that may have resulted in hospitalizations, called serious adverse events (SAE).
We also learn how many patients dropped out of the trial due to AE -- it was relatively few.
Lots of dropouts can be a bad sign for future drug sales, depending on the situation.
And, there is often a Medication Guide designed to educate patients about the drug of interest.
Still, how well the drug works remains one of my favorite questions answerable by a PI.
Link 1: http://www.
accessdata.
fda.
gov/scripts/cder/drugsatfda/index.
cfm?fuseaction=Search.
Search_Drug_Name Link 2: http://www.
allergan.
com/products/eye_care/botox.
htm
Since then the label for Botox has been expanded to four or five additional therapeutic indications, most recently to treat chronic, severe migraines.
First identified in 1897, botulinum toxin has become what Allergan's executive VP for R&D calls a "pipeline in a vial.
" Ka-ching! What it really means when the FDA approves a drug or expands the number of uses for it, is that the drug's manufacturer is allowed to market the drug for that use.
However, physicians are free to use any approved drug for any use of their choosing -- the FDA does not regulate the practice of medicine.
In fact, there were hints of Botox's potential benefit to treat migraine over ten years ago and some off-label use for migraine has occurred over the years.
Now, Botox is formally approved for that use, what professionals call an indication.
So, the only question remaining is how well does it work? Fortunately, the FDA.
gov web site contains a treasure chest full of technical literature gems called Package Inserts (PI).
The PI is included with shipments of every approved drug so that physicians (and you and I) can read all about it.
To get your free copy of almost any PI, just click on Link 1 at the end of this report and type in the name of your favorite drug.
Occasionally, the drug's manufacturer will post the PI prior to the FDA.
Then, just go to the company web site and search for the drug under Our Products or something like that.
Such is the case for Botox.
So from Link 2 at the end of this report, at the bottom of the linked page, click on "Please see full BOTOX Product Information," which will enable you to download a PDF of the most recent PI for Botox.
Now, let's read the PI together.
Yikes! This PI contains 25 pages of giant words written in tiny print.
You may ask "Does every physician really read every word of every PI for the drugs she prescribes?" Did you read every word of every article in today's newspaper? The key is to read about what you need to know.
What we need to know is how well Botox works to prevent migraines.
The first page-and-a-half of the PI is a bioquickie version, what is called the "Highlights.
" Skip it.
Go right to the Full Prescribing Information section on page two.
Drugs with a particularly serious risk profile may have a so-called Black Box Warning right up front.
Botox has one and so does the antidepressant Zoloft.
Lipitor, for high cholesterol, does not.
Surprisingly, many biotech drugs, recombinant versions of natural proteins, also have Black Box warnings.
These warnings put physicians on high alert for certain side-effects; but, the understanding is that for the right patients, the drug's benefits trump its risks.
If, over time, the risks seem to be more prevalent or new risks appear, a drug may get a more ardent Black Box warning or it may even be withdrawn from the market.
But many drugs with Black Box warnings have been on the market for many years.
To repeat, the question we want to answer is how well does Botox work for migraines? Well, there are all sorts of fun information in this PI, but to answer our question, we need to find and focus on Clinical Studies.
Please jump to page 14.
First, some background in migraines.
Several drugs, such as GlaxoSmithKline's Imitrex and Merck's Maxalt, are FDA approved to treat a migraine attack.
There are also several drugs, such as Johnson & Johnson's Topamax and Abbott's Depakote, that are FDA approved to prevent migraines.
Botox was approved for prevention.
But wait, there's more.
One could try to prevent a migraine episode or try to decrease the number of headache days per month.
This distinction is exactly where Allergan ran into a problem.
The company sponsored a pivotal clinical trial called PREEMPT I which randomized 679 patients to receive either two rounds of Botox injections (31 shots given at seven specified locations, see p.
4 of the PI) or two rounds of placebo injections.
Each injection cycle was given 12 weeks apart and the primary clinical endpoint (PCE) was the change from baseline in headache episode frequency.
The result was that there were 5.
3 fewer headache episodes per month, which isn't bad.
Unfortunately, that was in the placebo group.
There were 5.
2 fewer episodes per month in the Botox group, essentially a wash.
Given that all patients started the trial with 12-13 episodes a month, a roughly 40 percent reduction in migraine episodes is certainly clinically meaningful, just not Botox dependent.
Sadly, none of the data in this paragraph are in your PI.
The data were published in a professional journal, Cephalalgia, in March of this year.
Here's the rub.
All PIs are composed by the company hoping to sell its drug and edited by the FDA.
A Package Insert is a negotiated document.
Most people, including migraine sufferers, do not read the scientific literature, let alone Cephalalgia.
Before now, I had never even heard of Cephalalgia, but I don't get out much either.
Now, back to page 14 of our PI.
The paragraph at the bottom of the page tells us how the clinical trial was conducted.
On page 15 is Table 6 with the data we need.
Note that two studies are presented, each with almost 700 patients evenly split between Botox and placebo.
And, note that the efficacy endpoint described was the change from baseline in the frequency of headache days per month.
The measurement of headache days was a secondary clinical endpoint (SCE) in PREEMPT I.
In general, a trial that hits a SCE or multiple SCEs but misses its PCE is a bust.
Maybe the company should have run another clinical trial.
Allergan took a different approach.
With the PREEMPT I writing on the wall, Allergan revised its second pivotal trial, PREEMPT II, and made the PCE the change from baseline in headache days.
Voilà, success! Curiously, in this second trial, there was also a statistically significant decrease in migraine episodes which was now a SCE.
However, the improvement in episodes was not due to a better result with Botox -- the number of episodes per month dropped by 5.
3, similar to that in PREEMPT I -- but due to a weaker placebo effect.
In PREEMPT II, the decrease in the number of migraine episodes due to placebo was 4.
6 days per month, compared with 5.
3 in PREEMPT I.
The joke on Wall Street was that PREEMPT I clearly used "extra-strength" placebo.
You gotta love their sense of humor.
Using Table 6 we see that in PREEMPT I, the patients receiving Botox had 1.
4 fewer headache days per month than those receiving placebo (7.
8 minus 6.
4).
Likewise, in PREEMPT II, there was a 2.
3 day reduction in headache days per month favoring Botox over placebo (9.
2 minus 6.
9).
On average, Botox reduced the number of headache days per month by about two days compared with placebo.
As for migraine episodes, the data for Botox are equivocal since the two trials had different results.
It seems to me that Allergan had every reason to choose migraine episodes as its PCE in both studies and was right in doing so initially, as in PREEMPT I.
It also seems to me that changing the endpoints to follow the data is not quite right; a bit like a dog chasing his tail.
But that's my opinion.
One last point.
In both clinical trials, patients had experienced about 20 headache days per month before being randomized to Botox or placebo.
That tidbit is in the two Cephalalgia papers, not in the PI.
That means that patients getting two rounds of Botox to prevent migraines experienced a benefit, net of placebo injections, of about ten percent (two fewer headache days per month).
And that's how well Botox works for migraines.
Plus, now you have been introduced to reading the Package Insert.
Postscript: There is much more information in the PI that I did not cover.
There is information on adverse events (AE), often called side-effects.
And there are AE that may have resulted in hospitalizations, called serious adverse events (SAE).
We also learn how many patients dropped out of the trial due to AE -- it was relatively few.
Lots of dropouts can be a bad sign for future drug sales, depending on the situation.
And, there is often a Medication Guide designed to educate patients about the drug of interest.
Still, how well the drug works remains one of my favorite questions answerable by a PI.
Link 1: http://www.
accessdata.
fda.
gov/scripts/cder/drugsatfda/index.
cfm?fuseaction=Search.
Search_Drug_Name Link 2: http://www.
allergan.
com/products/eye_care/botox.
htm
Source...