This opinion piece caught my eye in the New England Journal of Medicine this week for a number of reasons. First off, it involves my least favorite antidepressant, so I am always interested in why there may be broadened indications. I stopped prescribing paroxetine well over 20 years ago because of what I considered to be an unfavorable pharmacokinetic and side effect profile. In this case there is a potential interaction that could decrease the efficacy of tamoxifen - a medication used to treat breast cancer. In my capacity as a consultant, I frequently encounter people who are already on paroxetine and will continue it if that is their preference. I do encounter people with side effects and either transition them off of the medication or onto something else depending on the clinical scenario. Second, FDA decision-making is always interesting to follow and that is also true in this case. Like the recent Zohydro decision, the FDA decided to go against the recommendation of its scientific committee and approve the specified name brand product for hot flashes. Third, the remarketing for generics as brand name drugs with slightly different formulations is another interesting aspect of FDA regulation and in this case that is also true. The product in this case is Brisdelle, a 7.5 mg tablet of paroxetine. Paroxetine is currently available as a generic drug as 10, 20 , 30 and 40 mg immediate release tablets and 12.5, 25 and 37.5 mg controlled release preparations. Some of the 10 mg tablets are scored allowing for an approximation of the approved product. Another interesting pint is that this action is apparently over a year old according to this FDA release.
The scientific evidence comes down to two randomized, double-blind, placebo controlled trials of 1184 women. Baseline moderate to severe hot flashes were occurring at a rate of about 10 per day. The effects observed during the trial were described as modest with an overall decrease in these hot flashes by about 0.9 per day by week 12 in one study and 1.7 per day in the other. Despite that slight effect, the women on active drug reported a meaningful difference at the end of the study and again at 6 months.
In balancing the risk rewards of this study and the regulatory decision, the FDA Reproductive Health Drugs Advisory Committee voted 10-4 against approving Brisdelle. The authors speculate that modest efficacy and existing warnings about suicidality swayed the Advisory Committee in this vote. I thought it was interesting that the authors consider the safety profile of paroxetine to be well established. Paroxetine seems to be one of the most complained about medications on the Internet and a favorite of critics who consider themselves as watchdogs of the pharmaceutical industry. In the case of Brisdelle, the formulation had side effects that did not exceed placebo. Anyone used to prescribing selective serotonin reuptake inhibitor (SSRI) type antidepressants is aware of a possible significant discontinuation syndrome. The authors suggest that Brisdelle does not need to be tapered if discontinued but the package insert from the web site suggests potential problems including potential discontinuation symptoms and symptoms in neonates exposed to the medication.
The authors discuss the importance of a non-hormonal agent to treat hot flashes, especially give the impact on lifestyle and risks of stroke, coronary artery disease, thromboembolism, and breast cancer. The only current FDA approved treatment is hormonal therapy in the form of estrogen or estrogen-progestin. The problem is that using SSRIs or SNRIs for hot flashes is already a fairly widespread practice and it is backed up to some degree by double blind controlled trials where approval by the FDA was not the goal of the study. I did a Medline search of controlled clinical trials for hot flashes and came up with this rough collection. If you look up the specific studies you will find some positive and some negative. I did not look for meta-analyses. The most interesting opinion was naturally the one that matched mine and I found it in this study. In this study the authors correctly conclude that potent CYP2D6 inhibitors should be avoided in women taking tamoxifen and the recommendation to avoid paroxetine is being ignored. Using a database from the Netherlands paroxetine remains a widely prescribed antidepressant.
The last reference reflects my opinion on paroxetine in general and certainly for this specific problem. It is also an example of how the FDA could consciously avoid problems rather than sending them out into the free market and waiting for the post marketing surveillance. That is acceptable in the case of serious conditions with no current therapies. But this is a case of alternate therapies with scientific backing being readily available and experts in the field recommending not to use paroxetine. I also have a concern about the impact that remarketing a generic drug with an unusual dose on the generic drug market in general. I am thinking about the precedent of generic doxepin being remarketed as Silenor in unique dosages of 3 and 6 mg. Most pharmacies have generic doxepin 10 mg tablets as the smallest available dose size with a liquid form that can be ordered for additional dose adjustments. My impression is that the prescribing of doxepin in low doses has increased in general with the recognition that low doses can be used for insomnia. Doxepin is primarily a substrate for CYP enzymes, but at some point could probably accumulate with higher doses. If the same thing happens with paroxetine it may result in standard antidepressant doses being prescribed for hot flashes. That could lead to more complications and potentially more drug interactions.
George Dawson, MD, DFAPA
1: Ronald J. Orleans, M.D., Li Li, Ph.D., Myong-Jin Kim, Pharm.D., Jia Guo, Ph.D., Mahboob Sobhan, Ph.D., Lisa Soule, M.D., and Hylton V. Joffe, M.D., M.M.Sc. FDA Approval of Paroxetine for Menopausal Hot Flashes. N Engl J Med 2014; 370:1777-1779.
2: Binkhorst L, Mathijssen RH, van Herk-Sukel MP, Bannink M, Jager A, Wiemer EA,
van Gelder T. Unjustified prescribing of CYP2D6 inhibiting SSRIs in women treated
with tamoxifen. Breast Cancer Res Treat. 2013 Jun;139(3):923-9. doi:
10.1007/s10549-013-2585-z. Epub 2013 Jun 13. PubMed PMID: 23760858.