Back to my St. John's wort page
My thoughts on the Pfizer St. John's wort study:
This study was sponsored by Pfizer, Inc., who holds the patent on Zoloft. Pfizer also sold St. John's wort, but the company stopped selling herbs last year because it wasn't profitable. I'm sure that St. John's wort cuts into their Zoloft prescriptions and I feel the study was purposefully designed to show that St. John's wort would not work.
First of all, they used a dose of 900 mg to 1200 mg. I've never heard of an antidepressant that was not dose dependent and I feel that St. John's wort is no exception. For example, the range for prescribing Effexor is from 37.5 mg to 450 mg. I would equate 900 mg of St. John's wort to the 37.5 mg dose of Effexor and they definitely would not be prescribing 37.5 mg of Effexor for major depression.
Imipramine has been used in many studies with other antidepressants because, to the best of my knowledge, nothing has been shown better for depression. The reason no one uses it is because of the severe side effects. For example, men may grow breasts. When imipramine is used for major depression, the dose is usually 150 to 200 mg. See http://qualitycounts.com/fp/severedepression.htm. When imipramine was tested against St. John’s wort, they used 150 mg of imipramine against 1800 mg of LI 160 St. John’s wort or 75 mg imipramine against 900 mg SJW.. See http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=9342765&form=6&db=m&Dopt=b and http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=7857502&dopt=Abstract. That would seem to indicate that 25 mg of imipramine is equal to 300 mg of St. John’s wort. Also, if 150 to 200 mg of imipramine is required for major depression, then it would make since that 1800 mg to 2400 mg of St. John’s wort would be required for major depression. St. John’s wort has seen safely tested up to 3600 mg (see http://www.nhir.com/tests/St._Johns_Wort_Module.pdf). But in the Pfizer SJW study they used 900 to 1200 mg of St. John’s wort, and it still showed an effect: "In the intention-to-treat sample, there were significant differences in remission rates, but the rates were extremely low: 14.3 percent for St. John's wort and 4.9 percent for placebo" See http://www.docguide.com/news/content.nsf/news/8525697700573E1885256A300067A09B?Open&id=F79F0A10A8262633852568C6005EBD19&count=10 or http://qualitycounts.com/wortpsl.htm. Maybe that's because the dose was extremely low of an inferior extract. Duh!
I also question the brand of St. John’s wort used for the study. The brand used in the German studies was Kira® with the LI 160 patented extract. I know I’ve switched to that brand and could tell a huge difference. If Pfizer was selling St. John's wort, I'm sure that is what was used in the study and I feel that invalidates the study right from the start.
One of the problems you run in to is that no one is sure what the active ingredient is in SJW. Some think it is the hypericin (should be at least 0.3%). Others theorize that it could be a combination of things. For a discussion, see http://www.nhir.com/tests/St._Johns_Wort_Module.pdf. There is also the question of bioavailability. To use an extreme case as an example, you could have wood that was 0.3% hypericin and flower pedals that were 0.3% hypericin. If you were to grind up the wood and grind up the flower, which do you think would be more bioavailable? With SJW, you don't have wood but you do have flowers, leaves, stems and roots. For a discussion on that, see http://www.consumerlab.com:8080/news/news_041101.asp or better yet, get the subscription version for a more in depth discussion.
The amount of flavonoids in the extract has also been shown to effect. An example: Effects of Hypericum perforatum on levels of 5-hydroxytryptamine, noradrenaline and dopamine in the cortex, diencephalon and brainstem of the rat - J Pharm Pharmacol 1999 Jun;51(6):723-8 - "When the extracts contain a higher concentration of flavonoids the effects are more widespread"
Also, there is a threshold below which some supplements do absolutely nothing. An example: Multivitamin With Folic Acid May Reduce Homocysteine More Than With Folic Acid Alone - Doctor's Guide, 4/20/01 - "the supplement containing 800 micrograms of folic acid reduced homocysteine levels by nearly 50 percent while the supplement containing 160 micrograms of folic acid had no significant effect above placebo"
I feel that the people designing the study were influenced by Pfizer to ensure that St. John’s wort did poorly in the study. They used far to small a dose plus I'm suspect that they used the Pfizer brand of SJW and not the patented LI 160, which has shown to be effective in numerous studies. Plus, it contradicts the German study with the LI 160 (http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=9342765&form=6&db=m&Dopt=b).
Just a couple of the many LI 160 studies:
Comparison of an extract of hypericum (LI 160) and sertraline in the treatment of depression: a double-blind, randomized pilot study - Medline/Clin Ther 2000 Apr;22(4):411-9, 4/00 - "Clinical response (defined as a > or =50% reduction in HAM-D scores) was noted in 47% of patients receiving hypericum and 40% of those receiving sertraline ... The [LI 160] hypericum extract was at least as effective as sertraline in the treatment of mild to moderate depression in a small group of outpatients" Note: Sertraline is the generic name for Pfizer's Zoloft®. - funny how the Pfizer St. John's wort study was plastered all over the news yet the study showing SJW better than Zoloft was never picked up by the media.
Efficacy and tolerability of St. John's wort extract LI 160 versus imipramine in patients with severe depressive episodes according to ICD-10 - Medline, 9/97 - "6-week trial comparing 1800 mg LI 160/die to 150 mg imipramine/die in severely depressed patients according to ICD-10 ... mean values 25.3 to 14.5 in the LI 160 group and 26.1 to 13.6 in the imipramine group ... Regarding adverse events, the nonrejection of the nonequivalence hypothesis denotes a superiority of the herbal antidepressant. These main result indicate that LI 160 might be a treatment alternative to the synthetic tricyclic antidepressant imipramine in the majority of severe forms of depressions"
Consider Undertreatment in Resistant Depression - Clinical Psychiatry News, 4/01, user=benhess, pwd=asdfgh - "Dosage should be pushed to the maximal tolerated level—something many psychiatrists are not prepared to do—and then maintained there for at least 6 weeks before an agent is concluded to be ineffective"
Journal Charges Drug Companies Too Powerful - ABC News, 5/18/00 - "Drug companies that pay for research and clinical tests of new medicines have been suppressing or manipulating the results"
The following is Richard Shelton, M.D.'s response. I still don't buy his argument for using such a low dose. On the one hand he is saying that imipramine is marginally effective below a certain dose but on the other hand, that is not the case for SJW. Also Hypericum LI 160 inhibits uptake of serotonin and norepinephrine in astrocytes - Brain Res 1999 Jan 23;816(2):358-63 - "We found that LI 160 inhibited both serotonin and norepinephrine uptake in a dose-dependent manner" - Note "dose-dependent". I still say to substitute 300 mg of LI 160 for 25 mg of imipramine at any effective dose of imipramine and see what happens. If the effective dose of imipramine is 225 mg, then use 2700 mg of SJW. I'm not going to be satisfied until I see that kind of study.
Mr. Hess:
I was very interested in your recent communications regarding our St. John's wort study. Thank you for the opportunity to respond. I
Have read you webpage carefully and have found a number of serious errors regarding
the facts of this study. Please let me clarify them if I can. I will take your points one at a time:
You said: "This study was sponsored by Pfizer, Inc., who holds the patent on Zoloft. Pfizer also sold St. John's wort, but the company
stopped selling herbs last year because it wasn't profitable. I'm sure that St. John's wort cuts into their Zoloft prescriptions and I feel
the study was purposefully designed to show that St. John's wort would not work."
My response: This paragraph contains several errors. Pfizer does sell Zoloft. However, it also still sells a popular brand of St. John?s
Wort extract under the brand name of Quanterra Emotional Balance via its Warner Lambert Division. Curiously, the popularity of SJW
has had no discernable effect on the market for antidepressants including Zoloft. The most likely
interpretation is that the people who take SJW for depression are not the ones who would take an antidepressant. Also, if our experience is any
indicator, there are many people who have taken SJW without benefit and have ended up on an antidepressant.
Perhaps some history is in order. Our study grew out of a set of observations made by my colleagues and I, that many people were
showing up in our clinics with moderate to severe depression who had failed to have an
adequate response to SJW. I then reviewed the literature and was surprised and disturbed by what I saw there. I contacted a group of excellent
investigators in the field and we came up with an acceptable research protocol. I then set about finding a source of funding for the study. Since
Pfizer, Inc. markets an antidepressant a SJW extract product I approached them for funding for this project. We insisted that this should be an
independent study. After funding was agreed upon, neither I nor anyone else in the project communicated with the company about this project. At
the completion of the project, the data were analyzed by a separate program at the University of Pennsylvania. Neither I nor any other
researcher had control of the data. In fact, the U Penn group wrote the 'Results' section
of the paper. As you may know, the results were pretty much negative: SJW is no different than placebo. However, as I say clearly in the discussion,
this by no means puts the question of SJW to rest. We need further studies to confirm or refute our finding. In fact, a study funded by the NIH is
nearing completion. The design is similar to ours. I believe that we need to wait until that study is completed before making any final
conclusions about SJW for this group of people.
The statements are further undermined by the fact that many of the previous research was funded by companies that make SJW extract.
You said: "First of all, they used a dose of 900 mg to 1200 mg. I've never heard of an antidepressant that was not dose dependent and I
feel that St. John's wort is no exception. For example, the range for prescribing
Effexor is from 37.5 mg to 450 mg. I would equate 900 mg of St. John's wort to the 37.5 mg dose of Effexor and they definitely would not be
prescribing 37.5 mg of Effexor for major depression.?" Etc., etc.
My response: This segment of your argument hinges on finding the "right" dose of SJW, like any other antidepressant. You may be
surprised to discover that there are many recommendations regarding the ?right? dose.
However, we did not rely on our own judgment to determine the "right" dose to employ. You may recall that there were more than 20 published clinical
trials to date before we set out with our study. The dosing on the study was carefully chosen to replicate the range of doses used in the previous
research. Please note that we used the same or higher dose that other studies used when they showed a positive result (i.e., SJW > placebo or SJW
= antidepressant). Although doses of SJW up to 3600 mg./day have been shown to be safe, as far as I am aware, there is no research
evidence suggesting a greater benefit of SJW above 1200 mg./day. Therefore, unless you are
willing to throw out all the previous studies, the dose we chose was appropriate. Also note that almost all of the people in the study received
1200 mg./day (higher than almost all of the doses in prior research).
However, I would also note that the argument about dose is a spurious. Arguing that a higher dose would produce a greater effect can
never be refuted. This is because you can simply say that we used a dose that was
too low, no matter what dose we chose. You do not have the same limitations in time and space that we do. Your position reflects a tautology along the
lines of "Since St. John's wort works, the dose you used must be too low." I will defend our choice as being representative of the doses of SJW
extract used in previous studies that supposedly showed that SJW is effective. Further, it is the dose that was recommended by the manufacturer
of the product.
You said: "In the intention-to-treat sample, there were significant differences in remission rates, but the rates were extremely low: 14.3
percent for St. John's wort and 4.9 percent for placebo"
My response: We acknowledge this issue fully in our paper. Although I believe that it is a statistical artifact, I, nonetheless, suggested
that SJW might have been effective for a tiny fraction of our patients.
You said: "I also question the brand of St. John?s wort used for the study. The brand used in the German studies was Kira® with the
LI 160 patented extract. I know I've switched to that brand and could tell a huge
difference. If Pfizer was selling St. John's wort, I'm sure that is what was used in the study and I feel that invalidates the study right from the
start."
My response: This is a strange argument. "Kira" is the brand name for Lichtwehr Pharma LI 160 in the US, marketed as "Jarsin 300"
in Germany. This is exactly the product we used in our study. It says so clearly in
the Methods section of our paper. I suppose, then, that we weren't invalidated right from the start.?
By the way, it is my view that Lichtwehr Pharma is an excellent company. We found the SJW extract to be remarkably stable,
indicating excellent manufacturing practices.
You said: "One of the problems you run in to is that no one is sure what the active ingredient is in SJW. Some think it is the hypericin
(should be at least 0.3%). Others theorize that it could be a combination of things.
For a discussion, see http://www.nhir.com/tests/St._Johns_Wort_Module.pdf.
There is also the question of bioavailability. To use an extreme case as an example, you could have wood that was 0.3% hypericin and flower pedals
that were 0.3% hypericin. If you were to grind up the wood and grind up the flower, which do you think would be more bioavailable? With SJW, you
don't have wood but you do have flowers, leaves, stems and roots. For a discussion on that, see
http://www.consumerlab.com:8080/news/news_041101.asp or better yet, get the subscription version for a more in depth discussion." Etc., etc.
My response: It seems to me that the argument about active ingredients is effectively dealt with by the fact that we used the very
product that you have already endorsed (Kira, LI 160). However, I would like to make one
comment. The issue about which is the "active ingredient" in SJW is much like the argument about dose. There are at least a dozen chemicals in SJW
in significant quantities, and many more at trace levels. If you don't like the results of the study, you can always argue that we didn't get the right
chemical in our study.
You said: "I feel that the people designing the study were influenced by Pfizer to ensure that St. John's wort did poorly in the study.
They used far to small a dose plus I'm suspect that they used the Pfizer brand of SJW
and not the patented LI 160, which has shown to be effective in numerous studies. Plus, it contradicts the German study with the LI"
My response: I think I have dealt with these issues sufficiently above, although no amount of data will satisfy some people. However,
let me take one more stab at it:
There is a common confusion about the terminology used in the study. We studied people with "Major depression" which some have
argued has never been the group intended by promoters of SJW. Major Depression is a simple
(albeit misnamed) category, sort of like "cancer" or "heart disease". To achieve this distinction someone must have certain number of symptoms for
at least two weeks. However, the category does not imply anything about severity. Within the broad category of major depression people can have
mild, moderate, or severe symptoms. Many commentators have confused "major" with "severe." However, the terms are not synonymous.
We chose to focus on the middle group (i.e., those who are 'moderately' depressed) for several
reasons. The main argument for this group is that we wanted to give SJW any chance to work. Many of the previous studies that suggested a benefit with
SJW studied people with major depression at this level. In fact, even those that studied 'mild to moderate' depression claimed to have included a
large number of people at this level. Any interpretation that we studied a different group from the previous research is incorrect. Our
study replicated the basic design elements of many of those projects. Furthermore, since we could not ethically study people who were
severely depressed, we focused on this group in order to attempt to discern any therapeutic effect with SJW. Since the placebo
response rate is so high in truly mildly depressed persons, SJW is unlikely to fare well in a placebo
controlled study in people who have depression at that level. In fact, we 'enriched' the sample intentionally to favor SJW in order to detect any
benefit: 1. We excluded anyone who had experienced a prior treatment failure with SJW or any antidepressant medication; 2. We included a placebo
run-in week prior to randomization. This elimates people who will have an early placebo response, favoring the experimental treatment. 3. We
included broader symptom measures in case the standard measurements were not effective in detecting meaningful differences. We
still saw no effect.
You quote two studies:
Comparison of an extract of hypericum (LI 160) and sertraline in the treatment of depression: a double-blind, randomized pilot study -
Medline/Clin Ther 2000 Apr;22(4):411-9, 4/00 - "Clinical response (defined as a > or =50% reduction in HAM-D scores) was noted in
47% of patients receiving hypericum and 40% of those receiving sertraline ... The [LI 160]
hypericum extract was at least as effective as sertraline in the treatment of mild to moderate depression in a small group of outpatients" Note:
Sertraline is the generic name for Pfizer's Zoloft®. - funny how the Pfizer St. John's wort study was plastered all over the news yet the study showing
SJW better than Zoloft was never picked up by the media.
Efficacy and tolerability of St. John's wort extract LI 160 versus imipramine in patients with severe depressive episodes according to
ICD-10 - Medline, 9/97 - "6-week trial comparing 1800 mg LI 160/die to 150 mg imipramine/die in severely depressed patients
according to ICD-10 ... mean values 25.3 to 14.5 in the LI 160 group and 26.1 to 13.6 in the imipramine
group ... Regarding adverse events, the nonrejection of the nonequivalence hypothesis denotes a superiority of the herbal antidepressant. These main
result indicate that LI 160 might be a treatment alternative to the synthetic tricyclic antidepressant imipramine in the majority of severe
forms of depressions"
My response: These two studies exemplify the problems with all of the previous literature comparing SJW with an active
antidepressant. The first study by Brenner and colleagues demonstrates a common problem. The dose of
sertraline (50 mg./day) is though by many in the field to be an inadequate one. A proper comparison would be sertraline 100 - 200 mg./day, which is
the dose being used in the NIH trial. Further, the sample size was so small that it could not have possibly detected a meaningful difference. This
was simply a "failed trial."
The second study does employ a marginally effective dose of imipramine (150 mg./day). However, plasma levels have been used for
more than 20 years for imipramine in these kinds of studies. The levels were conspicuous in their
absence suggesting that the authors either were unaware of the practice (and, therefore, unqualified to do the study) or were trying to minimize
the effect of imipramine. Actually, the trial was so poorly done in general that the former explanation is probably the right one. The study had a
number of design errors that would simply invalidate it. The investigators failed to use standardized diagnostic assessments or symptom ratings. The
dose of SJW was quite low. Finally, the duration was much too short to determine effect. For example, although they did used the marginal
dose noted earlier, they would not have achieved that dose until two weeks before the end of the study. The plasma level of imipramine
would have required one more week to achieve steady-state. Imipramine requires three
or more weeks to produce a therapeutic effect. Therefore, the trial was too brief to detect a meaningful difference.
Our conclusions are based on a belief that strong and largely incontrovertible evidence is needed before making any conclusion about a
treatment, including . However, since our research is at odds with the prior research literature, the data stand in an unclear state.
However, as I stated earlier, that this does not put the issue to rest with SJW.
I'll conclude with some quotes from Bertram Russell:
"I think we ought always to entertain our opinions with some measure of doubt. I shouldn't wish people dogmatically to believe any
philosophy, not even mine."
" In all affairs it's a healthy thing now and then to hang a question mark on the things you have long taken for granted."
"Aristotle maintained that women have fewer teeth than men; although he was twice married, it never occurred to him to verify this
statement by examining his wives' mouths."
Richard Shelton, M.D.
|
|