The introduction of 3,4-dihydroxyphenylalanine (DOPA) to the treatment of Parkinson’s disease (PD) has been a major scientific and clinical breakthrough in the treatment of this devastating disease. This can be seen from two aspects. The primary one is, of course, the enormous benefit to patients, but in addition there is the realization that understanding the biochemical deficits can provide a clue as to how replacement therapy can be successfully employed even in neurodegenerative diseases, providing significant symptomatic benefit if not cure. This has had a great impact on attempts to treat other neurodegenerative disorders, particularly Alzheimer’s disease. Unfortunately, in spite of miraculous effects on patients with early and advanced PD and the motor benefits afforded to them, it became clear that DOPA does not slow the neurodegenerative process and its effects are purely symptomatic. Consequently, the dose of drug that is needed to control the motor manifestations has to be increased gradually as the disease progresses. It quickly became clear also that of the two DOPA isomers, only the levorotatory stereoisomer, l-DOPA (levodopa), was able to afford therapeutic benefits, and chemical means to separate the isomers were developed. In practice, only levodopa is now used in the treatment of PD, resulting in an improved safety profile. Soon after came the recognition that some of the adverse effects associated with the drug were the result of peripheral, rather than central, conversion of levodopa to dopamine, which, unlike levodopa, has significant autonomic activity [1]. Unlike levodopa, dopamine does not cross the blood–brain barrier, and thus this metabolite does not contribute to the clinical benefits afforded by levodopa, and in fact causes significant adverse effects, particularly autonomic.

The enzyme involved in the transformation of levodopa to dopamine, l-amino acid decarboxylase (l-AAD, also called DOPA decarboxylase), is widespread in the body, with high concentrations present in the liver. Two agents were developed that could inhibit this enzyme, and both are still in use – carbidopa and benserazide. At present, practically all patients who require treatment with levodopa receive it in a fixed-dose combination with one of these inhibitors. Of course, as it is essential that levodopa be converted to dopamine in the brain, the l-AAD inhibitors should not cross the blood–brain barrier.

The inhibition of peripheral l-AAD had another result, at first unappreciated, namely a prolongation of the biological half-life of levodopa (and therefore also of dopamine in the brain). The importance of this effect is manifested in advanced PD. Early on in PD, there is a dramatic beneficial effect of levodopa, known as the “honeymoon period.” As the disease progresses and additional dopaminergic neurons are lost, there is a need to compensate for this by increasing the daily dose of levodopa. This is first manifested by shortening of the duration of action of individual levodopa doses, called the “end-of-dose” effect or “wearing off.” Later on, other manifestations appear including “peak-of-dose” dyskinesias and erratic responses to levodopa (so-called unexpected “on–off”, or yo-yoing). While the exact mechanism responsible for this erratic response is still elusive, it is at least partly dependent on pharmacokinetic factors such as plasma levels of levodopa. In particular, the phenomenon of wearing off, where the initial prolonged response to individual doses of levodopa is no longer maintained [2], limits patients’ independence. Wearing off probably results from a combination of peripheral and central mechanisms. Peripheral mechanisms include delayed gastric emptying and impaired absorption due to dietary protein, and central mechanisms are mainly due to impaired capacity of the nigrostriatal dopaminergic neurons and their terminals to uptake, store and release dopamine. This problem becomes more severe as gastric emptying becomes more sluggish and more dopaminergic terminals degenerate [2]. Blockade of peripheral l-AAD, which prolongs the biological half-life of the drug, can only incompletely compensate for this

An alternative metabolic pathway for levodopa, in the brain and in the periphery, is through the activity of another enzyme, catechol-O-methyltransferase (COMT). Like l-AAD, this enzyme is widespread in the body, with high concentrations occurring particularly in the liver. Notably, when levodopa tablets are used, the chemical reaches the liver at high concentration through the portal system and is there exposed to enzymatic degradation. This so-called “first-pass effect” is of crucial importance in reducing the concentration of levodopa that reaches the brain. Inhibition of COMT results in slightly higher, and particularly more prolonged, plasma levels of levodopa, manifested as elevation of the area under the curve of the time–concentration curve to a given dose of levodopa. Inhibition of these two enzymes, l-AAD and COMT, does not have a direct antiparkinsonian effect, and they are used only as co-medications, or co-drugs. In particular, the prolongation of serum levels of levodopa offers a valuable benefit to patients with the wearing-off phenomenon [3].

In the brain, COMT is an extracellular enzyme that metabolizes the conversion of levodopa to 3-O-methyldopa and of dopamine to 3-methoxytyramine. It is associated with dopaminergic synapses but actually occurs outside the synapse. Its main physiological action is probably to limit diffusion of the biologically active neurotransmitter, dopamine (as well as norepinephrine), outside the desired site of action. Thus, COMT helps to maintain neurochemical stability in the central nervous system. The metabolism of levodopa by COMT results in its conversion to 3-O-methyldopa, which is inactive as a neurotransmitter but has a long biological half-life and can be converted back to levodopa (and consequently to dopamine). Thus, it may function as a depot prolonging the action of levodopa. Moreover, 3-O-methyldopa may compete with levodopa in uptake through the blood–brain barrier, and thus limit the availability of levodopa for conversion of dopamine centrally.

Normally, no or only very small amounts of levodopa exist in the central nervous system, but this situation is altered dramatically when levodopa is given as a drug and gets into the brain, causing huge amounts to accumulate in the extracellular space. Under these conditions levodopa may be degraded by COMT (and other enzymes), thus limiting the amount available for neuronal uptake and transformation into the active metabolite, dopamine.

The normal release of dopamine from nigrostriatal terminals in the striatum is primarily tonic. Conversely, levodopa administration results in fluctuations in serum and brain levels and pulsatile stimulation of brain dopamine receptors. The smoothing of fluctuations in serum levels of levodopa, provided by inhibition of l-AAD and of COMT, may theoretically reduce the motor fluctuations characteristic of advanced PD by providing more continuous dopaminergic stimulation. More importantly, the unnatural fluctuations of dopamine levels in the brain when levodopa is given to PD patients may contribute to receptoral or post-receptoral changes in the striatum or elsewhere in the brain, which may induce fluctuations that are pharmacodynamic, not pharmacokinetic, in nature. If this still unproven assumption is correct, then treatment with levodopa together with its two metabolic inhibitors might delay the appearance of these manifestations of advanced PD. Based on this speculation, treatment with both inhibitors as soon as levodopa is first prescribed was hypothesized to delay the appearance of motor fluctuations.

Four inhibitors of COMT have been developed, entacapone, tolcapone, opicapone and nebicapone. At present, only tolcapone and entacapone are available on the market. Tolcapone, entacapone and opicapone have a short half-life of 1.4–3.6 h. The standard dose of entacapone is 200 mg, and of tolcapone is 100 mg, and no titration is needed. The dose of opicapone is as yet undetermined and will be either 25 or 50 mg. Nebicapone has not yet been put on the market. Of the four COMT inhibitors, tolcapone and nebicapone inhibit brain COMT [4], whereas entacapone and opicapone do not cross the blood–brain barrier and thus are unable to inhibit COMT in the brain. It is unclear what exactly the consequences of this difference are, but they are unlikely to be large. Because dopamine itself is also a COMT substrate, COMT inhibition is expected to prolong even further the duration of action of a given levodopa dose. As stated above, COMT metabolizes dopamine into 3-O-dopamine. As the latter does not contribute directly to the antiparkinsonian effect of levodopa, and may even block dopamine receptors, COMT inhibitors acting both in the periphery and centrally are theoretically advantageous.

To date, entacapone has been the most extensively studied COMT inhibitor. It is a peripherally acting, selective, reversible inhibitor of COMT. It prolongs the action of levodopa and indeed increases “on” time in patients with motor fluctuations. The main pharmacokinetic effect of COMT inhibitors is to prolong the biological half-life (i.e. the duration of action) of levodopa. Therefore, the area under the curve of the serum concentration of levodopa almost doubles. However, it may cause slight elevation of C_max (peak levodopa concentration). This may result in dopaminergic hyperstimulation, manifested as dyskinesias. The latter can usually be treated by reducing the individual levodopa dose [5]. Nausea, a side effect of levodopa, which is partly a central and partly a peripheral action, may also be increased. If this occurs, it can be managed by either dose reduction or the addition of domperidone. Because of the mechanism of action, entacapone (as well as the other COMT inhibitors) is ineffective unless given concomitantly with levodopa.

In repeated dosed, entacapone reduces the peak-to-trough variations of levodopa in the plasma, and thus presumably provides a more evenly sustained dopaminergic stimulation to the brain. The COMT(HH) genotype in PD patients enhances the effect of entacapone on the pharmacodynamics and pharmacokinetics of levodopa. The response to entacapone after repeated administrations and in heterozygous patients remains to be determined [6], as upregulation of the enzyme has been observed in red blood cells after 3 months of administration, suggesting that long-term administration may limit its efficacy [7]. The main indication for the use of COMT inhibitors is in the more advanced stages of the disease, when wearing off appears [8–13]. Based on the considerations mentioned above, the primary indication for COMT inhibitors is advanced PD, when patients have developed “end-of-dose” effects, “peak-of-dose” dyskinesias necessitating reduction of individual doses of the drug or severe response fluctuations. One point is patient selection. Patients recruited to the pivotal studies were mostly those with advanced disease with motor fluctuations. There are many ways to try and reduce the fluctuations, and it is not clear to what extent an attempt had been made to exploit alternatively methods before the patients were included in the studies.

Unfortunately, most of the pivotal studies cannot really be considered as double blind as entacapone causes urinary discoloration. Thus, to minimize the consequences of unblinding, later studies added an independent evaluator, a clinician, who was involved in the patients’ care and who simply measured the relevant endpoints, such as the United Parkinson Disease Rating Scale (UPDRS) or time spent “off.”

Several of the pivotal studies suffered from additional serious shortcomings. For example, the FILOMEN study [14] included patients at different disease stages and on different dosing regimens, ranging from two to ten times daily levodopa and entacapone administration. Obviously improvement cannot be expected for patients who need levodopa only twice daily, or if they take the medication every 1–2 h while awake. The fact that most patients received continuous-release preparations and many were also taking dopamine agonists further masks the potential benefits of entacapone.

The long-term benefits of entacapone are still not completely clear. Even though the clinical effects of COMT inhibitors are clear, data on long-term use are meager. In the NOMESAFE study, only 60% of patients continued to be treated for the full 3 years (including 30 months of open-label treatment) [15–16]. For example, in the NOMESAFE (Nordic Multicenter Study on Entacapone) study [16], while patients initially improved on entacapone, this benefit gradually diminished after 3 years, so that the duration of benefit from the first morning dose of levodopa, 2.1 h at baseline, extended initially to 2.8 h but then declined again to 2.5 h at the last evaluation. This could be the result of a reduced effect of the drug or a reflection of the progression of the disease, as would be expected after 3 years.

These studies, showing a clinically beneficial effect of entacapone, prompted studies that examined the benefits to PD patients with and without fluctuations. For instance, the UK–Irish Entacapone Study Group found entacapone to be beneficial in both fluctuating and nonfluctuating patients with PD in a randomized, placebo-controlled, double-blind, 6-month study [11].

Most patients suffering from PD are elderly and have comorbid conditions. Consequently, they consume a number of medications. The need to add COMT inhibitors is therefore an impediment, and may confuse the patients and decrease compliance. The practical solution to this problem was to design a tablet that contains the three elements: levodopa, carbidopa and entacapone. This combination, under the trade name Stalevo®, is now available in many countries. The short half-life of entacapone (about 2 h) is consistent with this procedure. One problem, however, is that inhibition of the enzyme is not immediate, and the short latency period may allow some levodopa still to be metabolized.

The encouraging results with entacapone, the concept that continuous dopaminergic stimulation may delay the appearance of motor fluctuations if started early in the disease course and the ease of use of Stalevo® were the driving force behind the STRIDE-PD (Stalevo Reduction in Dyskinesia Evaluation in Parkinson’s Disease) study [17]. The study was an international, multicenter, randomized, double-blind, parallel-group, active-controlled study. Surprisingly, the study found that patients taking Stalevo® had a shorter time to onset of dyskinesia (hazards ratio, 1.29; P=0.04) and increased frequency at week 134 (42 vs 32%; P=0.02). This study has discouraged the initiation of Stalevo® as a first-line therapy in PD.

Thus, it appears that entacapone is efficacious in PD patients with motor fluctuations but nonefficacious as a symptomatic adjunct to levodopa in nonfluctuating patients and nonefficacious in the prevention and delay of motor complications [18]. Importantly, STRIDE-PD raised safety issues regarding the long-term use of Stalevo®. Before this study, it was appreciated that diarrhea is a relatively specific adverse effect of entacapone. Its underlying mechanism is unclear. Appearing in up to 14% of treated patients, diarrhea is the most significant adverse event associated with entacapone and may lead to discontinuation of therapy in as many as half of cases. The diarrhea usually appears within the first few weeks of therapy, but not immediately, and typically disappears soon after withdrawal [8]. As PD patients frequently suffer from constipation, the diarrhea, if mild, is not necessarily a problem. Future pharmacogenetic studies might identify patients susceptible to this effect.

In the STRIDE-PD study, 11 males developed prostate cancer, of whom nine had received Stalevo®. The odds ratio for the occurrence of prostate cancer in males taking Stalevo® was 4.19 (95% Confidence Interval: 0.90–19.63). Healthcare professionals were advised to be aware of this possible risk and follow current guidelines for prostate cancer screening. Furthermore, the US Food and Drug Administration stated that meta-analysis of several studies “appeared to show an increase in the risk of heart attack, stroke, and cardiovascular death for people taking” Stalevo® but also stated “findings were not clear.” No recommendations regarding this possible side effect were issued [19]. Other less common side effects of entacapone include sleep attacks.

As stated above, an important pathway for the metabolism of levodopa is methylation by COMT. In this reaction, COMT uses S-adenosylmethionine as a methyl-group donor. Demethylation of S-adenosylmethionine forms S-adenosyl-homocysteine, which is hydrolyzed to homocysteine. Thus, the methylation of levodopa may interfere with homocysteine metabolism, leading to hyperhomocysteinemia. Although inhibition of COMT should theoretically prevent or reduce levodopa-induced hyperhomocysteinemia, results from several prospective studies are conflicting, possibly owing to differences in vitamin status of the study participants. In patients with low or low–normal folate levels, levodopa administration is associated with a greater increase in homocysteine and concomitant entacapone administration is associated with a greater reduction in homocysteine [20].

Interestingly, entacapone is a metal chelator and may therefore deplete the body of several metals, particularly iron. However, to date there are no reports of clinical problems due to this effect.

Tolcapone inhibits COMT in the brain as well as in the periphery. Meta-analysis of studies using COMT inhibitors has suggested that it may be more efficacious than entacapone [21]. This has been shown in studies such as the Entacapone to Tolcapone Switch Study where patients receiving entacapone switched to tolcapone or continued to receive entacapone [22]. However, its efficacy was overshadowed by its side effects. Patients receiving this medication had elevation of serum levels of liver enzymes and in some also overt evidence of hepatotoxicity. Elevation of liver enzymes occurred in 1–3% of patients receiving tolcapone; this may disappear even on continued treatment, and in most patients after drug withdrawal. However, a few cases of fatal acute fulminant hepatic failure have occurred. This resulted in almost total elimination of tolcapone from the market for a few years. It was reintroduced with a black box warning regarding hepatotoxicity [23]. It was deemed necessary that patients provide informed consent prior to being treated with this medication. The medication should not be initiated if the patient exhibits clinical evidence of liver disease or serum liver enzymes greater than the upper limit of normal. Liver function tests are recommended at baseline and periodically (i.e. every 2–4 weeks) for the first 6 months of therapy. After the first 6 months, periodic monitoring is recommended. Patients who fail to show substantial clinical benefit within 3 weeks of initiation of treatment should be withdrawn from tolcapone because of the risk of hepatotoxicity with no apparent improvement. After the medication was reintroduced, the company distributing the medication (Valeant Pharmaceuticals International) claimed that severe hepatocellular damage occurred in only 0.04% of patients, mostly with no clinical signs or symptoms. The mechanism underlying tolcapone-induced hepatotoxicity is not known.

Tolcapone can also cause diarrhea, constipation, headache, abdominal pain and urine discoloration.

Nebicapone, a peripheral and central COMT inhibitor showed promising results in preliminary studies. In 250 PD patients with motor fluctuations treated with levodopa/carbidopa or levodopa/benserazide (four to eight daily doses) given different doses of the medication or placebo, “off” time decreased significantly compared with placebo with 150 mg of nebicapone [24]. However, clinically relevant elevation in liver enzymes was observed in four of 46 patients with the nebicapone 150 mg dose. No further studies with this medication have been reported.

Opicapone, a peripheral COMT inhibitor, has shown no hepatotoxicity. It was evaluated in a phase 3 trial where it was compared with placebo and entacapone in PD patients suffering from end-of-dose motor fluctuations. A clinical benefit of opicapone has been reported in the company’s website but has yet to be published in medical journals.

Based on the considerations mentioned above, the primary indication for COMT inhibitors is advanced PD, when patients have developed “end-of-dose” effects, “peak-of-dose” dyskinesias necessitating reduction of individual doses of the drug or severe response fluctuations. However, the option of using COMT inhibitors must be seen against other possibilities of dealing with these situations.

Several options are available for the treatment of PD patients in the more advanced stages. When wearing off appears, more frequent dosing of levodopa, long-acting levodopa preparations, and the addition of selegiline or rasagiline as well as longer-acting dopamine agonists [25] have all shown efficacy. However, most of these methods have drawbacks. The addition of yet another agent increases the possibility of confusion as well as of interaction with any of the many other drugs the patient may take (for neurological or other indications), and therefore the addition of a drug that will prolong the action of levodopa itself is expected to benefit these patients. This is particularly so as levodopa remains the most potent of all antiparkinsonian drugs.

As mentioned above, several of the theoretical or actual benefits of more stable levodopa delivery can also be achieved through other means, such as duodenal infusion of levodopa. However, this is technically difficult and may cause severe complications and so has not achieved popularity. A more appealing option is to use direct-acting drugs such as cabergoline [26]. However, dopamine agonists are not selective to the brain, and their peripheral actions may cause significant side effects such as nausea and autonomic changes, particularly orthostatic hypotension [27], and those that are ergot derivatives may cause complications such as cardiac valve fibrosis. They are also considerably less potent than levodopa. The most effective agonist, apomorphine, is also not user friendly because of its poor gastrointestinal absorption and the need for repeated daily injections or a pump connected by a subcutaneous needle. The transdermal absorption of drugs such as rotigotine [28] is appealing, with similar efficacy to other dopamine agonists but with application site reactions [29]. Dopamine agonists may increase sleep time, and the reduced “off” time does not necessarily lead to functional improvement.

Several drug studies have so far been conducted in an attempt to answer some of the questions posed above [8–14, 30, 31]. While confirming the relative safety of both agents, they have not fully addressed relevant issues.

There is only limited information concerning comparisons between entacapone and dopamine agonists as no head-to-head study has been performed. In an open-label, multicenter study, tolcapone was compared with bromocriptine and was found to be somewhat more effective in advanced cases, although the differences were not very large and in many cases failed to reach statistical significance [16]. Recent results suggest that rasagiline is active in this situation. Although comparisons with COMT inhibitors still need to be made, available data from the LARGO study [32] suggest that the two drugs have similar efficacy. Since the two agents act through completely different mechanisms, it may be logical and interesting to study the effect of combining these therapies. Recent meta-analysis has in fact suggested that dopamine agonist therapy may be more effective than COMT inhibitor and monoamine oxidase type B (MAO-B) inhibitor therapy, which have comparable efficacy [21].

The relative advantages, if any, of entacapone versus long-acting dopamine agonists such as cabergoline or the transdermal preparations [22] will have to be studied. In addition, the notion that the longer half-life of levodopa afforded by the addition of COMT inhibitors prevents motor deterioration and the development of dyskinesias (particularly unexplained motor fluctuations) needs to be confirmed by future prospective studies with more attention to the dose of levodopa.

While entacapone currently dominates the market (being practically the only player), several pharmaceutical companies are attempting to develop other COMT inhibitors. One modification could be irreversible “suicide”-type inhibitors, which will require only once daily administration.

Levodopa combined with l-AAD inhibitors and COMT inhibitors should not be used, or only with extreme caution, in patients who receive nonspecific monoamine oxidase inhibitors (MAOIs) such as phenelzine, tranylcypromine or nialamid, because dietary tyramine will not be metabolized and could cause a severe sympathetic storm with extreme hypertension (the “cheese reaction”). Because of the long duration of activity of MAOIs, they should be stopped at least 2 weeks before l-AAD inhibitors are given. This precaution still applies to patients taking COMT inhibitors. On the other hand, selective MAO-B inhibitors are not contraindicated in this situation. Although still not studied formally, the combination of inhibitors of the three enzymes involved in dopamine metabolism may be advantageous and may result in an even longer effect of a given dose of levodopa.

If the dose of 150 mg levodopa is to be increased, for example to 200 mg, a combination of different Stalevo® tablets can be used, but this may not be the correct approach. Stalevo® is marketed with fixed-dose entacapone (200 mg), varying doses of levodopa ranging from 50 to 200 mg and varying doses of carbidopa ranging from 12.5 to 50 mg. The dose of 200 mg entacapone that is contained in any Stalevo® tablet causes maximal enzyme inhibition. In such cases, the addition of levodopa/carbidopa or levodopa/benserazide without additional entacaopone is more logical. Moreover, the safety of individual doses of entacapone exceeding 200 mg has not been assured.

Other drugs may also be metabolized by COMT inhibitors. These include all catechlamines (dopamine, norepinephrine, epinephrine, isoprenaline, dobutamine), as well as α-methyldopa, apomorphine, isoetherine and bitolterol. The effect of these agents may well be enhanced when co-administered with COMT inhibitors, and they should be used carefully, starting with lower doses than commonly recommended.

The addition of COMT inhibitors reduces the peripheral side effects of levodopa (nausea, vomiting and orthostatic hypotension) but, because COMT inhibitors allow more levodopa to be delivered to the brain, will not improve central side effects, which might even be worsened. In particular, the occurrence of dyskinesias may demand reduction of individual levodopa doses.

In conclusion, COMT inhibitors, and particularly entacapone, seem to be effective and safe additions to the armamentarium available for the treatment of advanced PD, resulting in a more continuous dopaminergic stimulation, which avoids exposing striatal (and other) dopamine receptors to alternating high and low concentrations of dopamine, thus reducing the effects and minimizing the risk of late motor complications. In particular, they diminish end-of-dose deterioration in the effect of individual doses, thus prolonging “on” time while decreasing “off” periods. Importantly, the reduced “off” time is due to increased “on” time.