Return to the Cognitive Enhancement Research Institute Home Page or Parkinson’s Page.
From the June 3rd, 1996 issue of Smart Drug News [v5n1]. Copyright (c) 1996, 1999. All rights reserved.

Smart Drug Update:

Recent Developments
With Deprenyl

by Ward Dean M.D. and Steven Wm. Fowkes

Over the last year, we have been investigating several controversial matters concerning L-deprenyl (selegiline), one of which is creating a mini-controversy on the Internet. The question being discussed on the Internet arose from last year’s publication of a long-term study of Parkinson’s disease which evaluated three therapeutic regimens: 1) L-dopa; 2) L-dopa plus deprenyl; and 3) Parlodel (bromocriptine). The results of the study raised some serious concerns about the use of deprenyl/dopa combination therapy for Parkinson’s disease, and some have questioned the value of deprenyl alone for Parkinson’s disease and, of broader concern to many readers of Smart Drug News, it’s use as a potential life-extending, anti-aging and cognition-enhancing substance in normal, healthy people.

This article should be considered an interim report, pending a more comprehensive evaluation that is being prepared. We had intended to incorporate personal communications from the authors of the study in question, as well as from other prominent Parkinson/deprenyl researchers and clinicians. However, because of the furor being generated on the Internet, we believed that this preliminary (and necessarily incomplete) report should be made at this time.

Dopa/Deprenyl Combination Therapy

The focus of the controversy regards the long-term findings of the Parkinson’s Disease Research Group (PDRG) of the United Kingdom [Lees et al., 1995]. The results of this study indicated that the combination of deprenyl plus L-dopa (also known as levodopa, or just dopa for short) resulted in a increased incidence of mortality compared with the groups on L-dopa alone or bromocriptine alone. These findings were a surprise to the authors of the study, who fully expected that their results would confirm an earlier but smaller study that indicated that deprenyl-dopa combination therapy resulted in increased survival in Parkinson’s patients [Birkmayer, 1985]. The earlier retrospective study found substantial life-extending benefits from the use of the deprenyl-dopa combination, while the larger and more recent prospective study found that dopa plus deprenyl “increased the number of early adverse effects” (compared with dopa as a single agent) and increased overall mortality by half. This finding was statistically significant.

In this study, more than 500 Parkinson patients were randomized into either a dopa-plus-deprenyl group or a dopa-only group. A third group was treated with bromocriptine only. The dopa-only group was intended as a control, by which the efficacy of added deprenyl could be ascertained. Inexplicably, no deprenyl-only control group was included, despite the fact that this study was motivated, in part, by a 1989 report from the United States Parkinson’s Study Group which indicated that deprenyl monotherapy delayed the need for dopa therapy by 9 months.

The increased mortality of the deprenyl-plus-dopa group as displayed by a cumulative statistical assessment of death (a Kaplan-Meier estimate) makes it quite clear that there is a significant adverse interaction between deprenyl and dopa that shows up three years into the study. (At the half-way point of the study, the researchers had reported that survival rates for deprenyl-dopa and dopa-only therapy were similar [Lees, 1993]). The size of the study group (520 at the start, decreasing to 201 at six years) makes it unlikely that the finding is an accident of chance. The researchers concluded that there needs to be a reassessment of deprenyl’s use in treating Parkinson’s disease.

Although we agree that there are life-and-death issues that need addressing, we believe that the reassessment should focus primarily on the combined use of L-dopa with deprenyl in Parkinson’s disease. Much evidence has accumulated in recent years that dopa adversely effects several key receptor and enzyme systems and may actually accelerate the progression of Parkinson’s disease. In our judgment, recent suggestions on the Internet that deprenyl therapy for Parkinson’s disease be abandoned is premature and hardly justified by the results of this study. However, we do believe that the simultaneous use of dopa and deprenyl should be discouraged.

Two Facets of the PDRG Study

The data graphed in the PDRG study show a dramatic increase in death rate by year three which is maintained through year six (see Figure on left, bottom two curves). The number of surviving subjects participating in the study at various years (see middle two curves) show roughly parallel plots that differ only slightly. The drop in these data not only reflect deaths, but also patients who dropped out of the study for various reasons and patients who had not been in the study long enough to reach the latter years of treatment. The ratios between the dopa-only and dopa-plus-deprenyl groups (plotted in the top curves) show a notable shift in the ratios between these two groups. The higher dopa-only ratio at three years shifts to a higher dopa-plus-deprenyl ratio at five and six years. This could be due to variations in recruitment during the original randomization process or differential drop-out rates, or it could be due to a decrease in mortality associated with long-term deprenyl use. We need to obtain and evaluate the raw (unpublished) data to know for sure.

A Window of Death?

There is one highly unusual feature visible in the cumulative Kaplan-Meier data presented in the PDRG study. The deprenyl-plus-dopa survival data appear to be decidedly non-linear. For the first two years of the study, the death rate from the dopa-plus-deprenyl group was almost identical to that of the dopa-only group (2% mortality per year). Then, suddenly, at about five months into the third year, the dopa-plus-deprenyl death rate suddenly increases by a factor of four (to 8% mortality per year). It stays high for about six months and then drops down over the next six months to levels similar to the dopa-only group (which have slowly risen to approximately 4% mortality per year), where it stays constant for the next two years.

There are three possible explanations for these observations: 1) it is an accident of chance; 2) it is due to an adverse effect of the dopa-deprenyl combination; or 3) it is due to statistical error.

To see what the deprenyl-plus-dopa survival curve might look like without the data from this one-year period (see figure at right), we replaced the high-death-rate curve within the “death window” (the red rectangle) with an extrapolated curve (see light-blue dashed line) and moved the post-41-month curve (the magenta line to the right of the death window) down to join the extrapolation (see twin purple down-arrows). Then we re-corrected the extrapolation to a smooth curve that better fit the new data points and re-adjusted the position of the post-41-month curve one last time. The result shows a remarkable congruence between the dopa-plus-deprenyl and dopa-only curves. One might speculate that the entire negative result of this study is solely due to data from this one-year period. What was happening during that year?

Although Dr. Lees and colleagues provide extensive documentation of the procedures used in their study, there is no discussion of the suddenness of this increase in mortality or factors that might account for it. Although we might be inclined to assume that changes in treatment protocol (it was an open study) might be expected to influence death rate in a non-linear fashion, this study’s participants were recruited over a five-year period. Thus, treatment options and outcomes were not concurrent in the 475 patients who were in the study near the 3-year mark. Overlapping changes would be expected to “smooth out” any such ripples and make them all but invisible. It seems decidedly more likely that this effect is inherent in the dopa, deprenyl or interaction between the two.

A Brief Review of Therapeutic Approaches to Parkinson’s Disease

Parkinson’s disease is the first documented example of a neurological disease which is consistently correlated with a deficiency in a specific neurotransmitter, i.e., dopamine. This observation provided the rationale for the use of L-dopa, a precursor of dopamine which is intended to restore dopamine levels and correct the neurotransmitter deficiency. This approach raised the hope that L-dopa might not only relieve the symptoms of Parkinson’s disease, but also delay or stop its progression. Unfortunately, this turned out not to be the case.

Although the primary pathology in Parkinson’s disease is the degeneration of dopaminergic cells in a part of the brain known as the substantia nigra, losses of noradrenergic or serotoninergic neurons have also been observed. In addition, many other peptide transmitters have been discovered, and the concept of cotransmission has been developed, greatly complicating the earlier simplistic description of Parkinson’s disease [Trabucchi et al, 1989].

L-dopa was introduced clinically in 1961 by Birkmayer and Hornykiewicz. When it was found that administration of L-dopa to Parkinson’s patients greatly improved their mobility and motility and partially or completely relieved their tremors, dopa was greeted with great expectations. However, these beneficial changes were short-lived due to the short duration of action of L-dopa, and its rapid decarboxylation (breakdown) by the enzyme dopa decarboxylase in peripheral circulation. Combining L-dopa with carbidopa or benserazide (drugs which inhibit dopa decarboxylase activity) greatly increased the amount of L-dopa which crossed the blood-brain barrier, and significantly extended its duration of action. This advance greatly enhanced the effectiveness of L-dopa therapy.

Unfortunately, it was found that the increased concentrations of dopaminergic neurotransmitters resulted in a down-regulation (or increased insensitivity) of the neurotransmitter receptors. Thus, a third approach was to add dopamine receptor agonists such as bromocriptine, lisuride, or terguride. For very complex reasons, these drugs are very effective in some patients, and equally ineffective (with considerable side effects) in others. For these reasons, therapy with dopamine receptor agonists must be highly individualized. Birkmayer and Birkmayer [1989] believe that dopamine receptor agonists can only be effective when used in conjunction with low doses of L-dopa. Rinne [1989] agrees that dopamine agonists alone do not seem to be useful anti-Parkinsonian agents, and recommends that dopamine agonists be combined with low, submaximal doses of L-dopa. He confirmed this concept in a three year study of 4 groups of patients: 1) L-dopa; 2) Lisuride; 3) L-dopa (low dose) plus Lisuride (high dose); and 4) L-dopa (high dose) plus Lisuride (low dose). Group three resulted in greatly reduced side effects, reduced disability, and much greater quality of life than the other three groups.

The most recent addition to the armamentarium of Parkinson’s therapy is L-deprenyl, which is a specific inhibitor of monoamine oxidase type B, an enzyme which degrades dopamine. L-deprenyl was found to be of significant benefit in Parkinson’s disease, in terms of alleviating symptoms, having minimal side-effects, and without the development of tolerance requiring larger and larger doses (as is the case with L-dopa).

Although the PDRG study is the first to demonstrate a negative synergy between deprenyl and L-dopa, Langston [1990] had previously expressed concern that L-dopa, by accentuating dopamine turnover, might actually be harmful to the remaining nigrostriatal neurons when used in conjunction with deprenyl.

Dr. Ward Dean’s Personal Experiences

I take 3 mg per day of Discovery-brand liquid deprenyl citrate. On several occasions I have experimented with low-dose L-dopa (as a growth-hormone stimulant, and as a possible way to restore hypothalamic sensitivity, in accordance with the concepts of Prof. Vladimir Dilman, M.D., Ph.D., D.M.Sc.). Whenever I took the L-dopa along with deprenyl, I experienced low-grade headaches, a sense of “fullness” in my head, and mental inattentiveness. These effects did not occur with either substance taken alone. Since I never appreciated any benefit from the L-dopa, I discontinued its use, and have likewise not recommended it to my patients. The PDRG study appears to confirm my personal experiences.

Although not the subject of this report, I also believe that there is a qualitative difference between Discovery liquid deprenyl citrate, and Eldepryl-brand deprenyl hydrochloride (the FDA-approved brand that is available by prescription in the US). My latest experience of this difference involved one of my patients who had been taking Discovery’s liquid deprenyl for several months, with extremely favorable results (increased energy, increased sense of well-being, and improved alertness and concentration). His wife had recently obtained health insurance through her job which covered all his prescription medications, so he requested a prescription for Eldepryl, which I provided. Several weeks later, he angrily came into my office and threw the Eldepryl at me. “You keep this junk,” he shouted. He claimed that it had caused headaches and depression when taking 1/3 to 1/2 tablet two or three times per week. I suggested that he not take the Eldepryl for a week or so, and then try again. He did, with the same results. He stopped taking Eldepryl, and is now back on the liquid deprenyl. He is again feeling happy and well. This story is not unique.

Steven Fowkes’ Personal Experiences

I continue to take 1 drop of Discovery liquid deprenyl every other day (0.5 mg/day dosage). During one of my “experiments,” I experienced headache and “mental difficulties” after taking 5 mg of a generic European deprenyl. Although I did not experience much dopaminergic stimulation (I am quite sensitive to such effects), I woke up tired and groggy the next morning and remained mentally dulled through the next day, despite my having taken two grams of tryptophan and 1.25 mg of melatonin before bed. A week later, as a control experience, I took 5 mg of Discovery liquid deprenyl (five times my normal dose) with no headache or cognitive difficulties. I experienced intense dopaminergic stimulation (I was uncomfortably wired), yet I woke up surprisingly rested the next morning after taking tryptophan and melatonin. I will definitely not be repeating that experiment.

Interim Recommendations

For individuals dealing with Parkinson’s disease, we think that L-dopa therapy has significant adverse potential and should be postponed as long as possible or avoided altogether. Deprenyl has been found to be useful towards this end [PSG, 1989; Olanow et al., 1995], and the PDRG study does not contradict that finding.

Although antioxidant therapy is an obvious approach to deal with increased oxidative stress and decreased antioxidant levels, scientists and doctors have been slow to apply this technology. Researchers have started investigating the effect of vitamin E towards this end, but antioxidants are much more effective in combinations than they are singly. More importantly, vitamin E is lipid soluble and provides minimal antioxidant protection to the aqueous (watery) metabolic compartments of the brain that are stressed in Parkinson’s disease. It makes much better sense to employ a broad-spectrum antioxidant intervention which emphasizes water-soluble antioxidants like vitamin C, glutathione, N-acetylcysteine, polyphenols, proanthocyanidins, lipoate, NADH, DMSO, etc. This general approach has been pioneered by Annetta Freeman with outstanding results [see SDN v4n6p1].

For individuals not dealing with Parkinson’s disease, there still may be some degree of increased oxidative stress from deprenyl use. Although we generally recommend broad-spectrum antioxidants as part of a general cognitive-enhancement and life-extension program, we wish to re-emphasize this recommendation for programs which include deprenyl and especially L-dopa. Although our concerns about dopa have grown to the point that we feel justified in discouraging its use (especially in light of the PDRG study), there may be circumstances in which its use may be required (such as in long-term Parkinson’s patients already on dopa therapy).

We continue to believe that deprenyl is a valuable drug with potential cognitive and anti-aging applications. Our present opinion favors Discovery-brand liquid deprenyl as the deprenyl product of choice. Our dosage recommendations in Smart Drugs II for deprenyl use by adults remain unchanged.

A subsequent follow-up article will be addressing the purity of commercial deprenyl products in more depth. We will also be discussing new research which suggests that disturbances in mitochondrial function may be a key factor underlying Parkinson’s disease and dopa toxicity, and possible therapeutic approaches to reducing this toxicity and enhancing mitochondrial efficiency.


Birkmayer W and Birkmayer JGD. The L-dopa story. In: Parkinsonism and Aging, by Calne DB, Comi G, Crippa D, Horowski R and Trabucchi M (eds). Raven Press, New York, 1989, pages 1-7.

Birkmayer W and Hornykiewicz O. Klin Wochenschr 73: 787, 1961.

Langston JW. Towards a strategy for preventing Parkinson’s disease. In: Basic, Clinical and Therapeutic Aspects of Alzheimer’s and Parkinson’s Diseases, Volume 2, by Nagatsu T, Fisher A and Yoshida M (eds). Plenum Press, New York, 1990, pages 505-11.

Lees AJ on behalf of the Parkinson’s Disease Research Group of the United Kingdom. Comparison of therapeutic effects and mortality data of levodopa and levodopa combined with selegiline in patients with early, mild Parkinson’s disease. British Medical Journal 311: 1602-7, 1995.

Mena MA, Pardo B, Casarejas MJ, Fahn S and Garcia de Yebenes J. Neurotoxicity of levodopa on catecholamine-rich human neuroblastoma cells. Mov Disord 5: 62-73, 1992.

Olanow CW, Hauser RA, Gauger L, Malapira T, et al. The effect of deprenyl and levodopa on the progression of Parkinson’s disease. Ann Neurol 38(5): 771-7, Nov 1995.

Parkinson’s Study Group. Effect of deprenyl on the progression of disability in early Parkinson’s disease. N Engl J Med 321: 1364-71, 1989.

Przedborski S, Jackson-Lewis V and Fahn S. Antiparkinsonian therapies and brain mitochondrial complex I activity. Mov Disord 10(3): 312-7, May 1995.

Rinne UK. New strategies in the treatment of Parkinson’s disease. In: Parkinsonism and Aging, by Calne DB, Comi G, Crippa D, Horowski R and Trabucchi M (eds). Raven Press, New York, 1989, pages 255-260.

Shigenaga MK, Hagen TM and Ames BN. Oxidative damage and mitochondrial decay in aging. Proc Natl Acad Sci USA 91: 10771-78, Nov 1994. “The decrease in cardiolipin with age is associated with a decrease in state 3/state 4 ratio.” “The sensitivity of cardiolipin to peroxidation increases with age in rodents, an effect that appears to be attributable in large part to the replacement of 18:2 acyl side chains with more readily peroxidizable 22:4 and 22:5 acyl side chains.”

Smith TS, Parker WD and Bennell JP Jr. L-Dopa increases nigral production of hydroxyl radicals in vivo: potential L-dopa toxicity? Neuroreport 5(8): 1009-11, Apr 1994.

Trabucchi M, Appollonio I, Battaini F, Govoni S and Frattola L. Influence of treatment on the natural history of Parkinson’s disease. In: Parkinsonism and Aging, by Calne DB, Comi G, Crippa D, Horowski R and Trabucchi M (eds). Raven Press, New York, 1989, pages 239-254.