REVIEW

Levodopa Therapy for Parkinson’s Disease: Pharmacokinetics and

Pharmacodynamics
Peter A. LeWitt, MD, MMedSc*

Parkinson’s Disease and Movement Disorders Center, Henry Ford Hospital, West Bloomfield, Michigan; and the Department of Neurology,
Wayne State University School of Medicine, Detroit, Michigan, USA

ABSTRACT: For all its imperfections at treating continuing challenge for better understanding of its
Parkinson’s disease (PD), orally-administered levodopa pharmacology. The pharmacokinetics of L-dopa tend to
(L-dopa) can be regarded as the “platinum” standard of predict some of problems that can emerge during chronic
PD therapeutics for its impact on disability and discom- therapy, which can be linked with its irregular uptake and
fort and its cost-effectiveness. The past half-century has marked dose-by-dose variability in plasma concentra-
confirmed that the typical L-dopa–treated patient gains tions. Several new pharmaceutical approaches are tar-
improvement for most Parkinsonian features, presumably geted at the unique physiology of L-dopa uptake and are
by conversion of this amino acid into dopamine in the likely to improve the consistency of its anti-Parkinsonian
striatum. However, fundamental questions remain as to effect. VC 2014 International Parkinson and Movement

its full mechanism of action and how adverse reactions Disorder Society
evolve. Various aspects of clinical phenomenology asso-
ciated with chronic L-dopa use (such as dyskinesias and K e y W o r d s : levodopa; carbidopa; pharmacoki-
the long-duration anti-Parkinsonian response) present a netics; pharmacodynamics; Parkinson’s disease

Neither obsolete nor ideal as a medication, levo-3,4- long-term benefit with the regular use of L-dopa. Past
dihydroxy-phenylalanine (levodopa [L-dopa]) is now concerns about toxicity or other considerations for
celebrating a half-century of service to Parkinson’s dis- avoiding L-dopa have been largely put to rest.3 L-dopa
ease (PD) patients. It also faces an ongoing challenge is commonly regarded as the “gold standard” of PD
to re-invigorate itself by new formulations improving therapy (but probably deserves an upgrade to
on dose-by-dose motor fluctuations and other thera- “platinum”!), reflecting its therapeutic superiority to
peutic limitations. After L-dopa’s arrival into the ther- all other anti-Parkinsonian treatments (including drugs
apeutic arena,1,2 which followed several years of false more potent in dopaminergic properties). Given its
starts, this amino acid radically transformed the image overall impact on disability, quality of life, and lon-
of PD away from a disorder of inevitable discomfort gevity, L-dopa stands out as one of the most cost-
and disability. Today, most patients with PD achieve effective medications ever to have been developed.
Patients averse to medication can also take some com-
------------------------------------------------------------ fort in knowing that L-dopa is a “natural” substance,
*Correspondence to: Peter A. LeWitt, MD, MMedSc, Parkinson’s Dis-
ease and Movement Disorders Center, Henry Ford Hospital—West found as a normal intermediate in human metabolism
Bloomfield, 6777 West Maple Road, West Bloomfield, MI 48322 USA, and present in some foods.
E-mail: [email protected]
L-dopa’s potential for treating Parkinsonism became
Funding agencies: This study was supported by a gift from MAC Valves
Foundation.
known by the rapid reversal of hypokinesia it accom-
plished in the reserpinized rabbit.4 With this information,
Relevant conflicts of interest/financial disclosures: Nothing to report.
Full financial disclosures and author roles may be found in the online ver- a strategy for enhancing striatal dopaminergic neuro-
sion of this article. transmission became an obvious priority.5 This goal
Received: 23 September 2014; Revised: 9 October 2014; Accepted: could not be achieved by dopamine, because this mole-
16 October 2014 cule’s electrical charge prevents its movement across the
Published online 00 Month 2014 in Wiley Online Library
blood–brain barrier. However, an alternative approach
(wileyonlinelibrary.com). DOI: 10.1002/mds.26082 was to use L-dopa as dopamine’s immediate precursor,6,7

Movement Disorders, Vol. 00, No. 00, 2014 1

P E T E R A . L E W I T T

because it could transfer into the brain via a mechanism of intact noradrenergic innervation.10 The implication
of facilitated amino acid transport. Though only a small for PD therapeutics may be that effective L-dopa therapy
fraction of a systemically-administered dose of L-dopa will need an additional restoration of norepinephrine
enters the brain, this quantity proved to be sufficient for innervation for completely reversing motor and other
restoring striatal dopaminergic neurotransmission. deficits.11
L-dopa’s symptomatic control of parkinsonism was soon Another consideration for inquiry into the complex-
joined by another pharmacological milestone, the devel- ity of PD treatments is that clinical responses to L-
opment of peripherally-acting decarboxylase inhibitors. dopa may be intermingled with ongoing compensatory
With these compounds (benserazide and carbidopa changes within the nigrostriatal pathway. At the ear-
[CD]), L-dopa’s systemic diversion to dopamine could be liest stages of PD, for example, the remaining popula-
blocked, thereby enhancing its CNS bioavailability, toler- tion of dopaminergic neurons is thought to be
ability, and clinical effectiveness.8 upregulated in synthesis, vesicular packaging, release,
The therapeutic concepts behind L-dopa treatment and nerve terminal uptake of dopamine, as was
might seem obvious, though managing PD with L-dopa observed in experimental models of Parkinsonism.12
can be, at times, a frustrating exercise. Its pharm- Hence, over time, a changing response to L-dopa
acology has been extensively investigated in over 5 might be attributable to either a decline in the com-
decades of research, but much more remains to be pensatory responses or to further loss of nigrostriatal
learned. In this article, some of the major accomplish- projections (or both).
ments and challenges still facing L-dopa therapeutics A half-century of investigating L-dopa neurochemis-
are reviewed in the context of its pharmacodynamics
try has also expanded the identity of this molecule
and pharmacokinetics.
beyond just a transient intermediate in the pathway of
catecholamine synthesis. Because L-dopa is not repre-
L-dopa Therapy: More Than a sented by a nucleic acid triplet codon, it lacks func-
Dopamine Precursor tional roles related to incorporation into protein.
However, several unique physiologic actions have
In 1970, at a time of great enthusiasm for the revo- been identified. One is that L-dopa serves directly as a
lutionary change in PD therapeutics, one of the pio- Ca11-dependent “classic” neurotransmitter. This
neers of L-dopa therapy, Oleh Hornykiewicz, sounded property has been observed in several brain regions
a note of concern about L-dopa that proved to be (including the CA1 hippocampus, the hypothalamic
extremely prescient: supraoptic and paraventricular nuclei, and the tractus
. . .if L-dopa works in parkinsonism by replenishing solitarii).13 L-dopa also acts at other central nervous
the missing dopamine in the striatum, then L-dopa is system sites and with neuromodulatory properties dif-
quite obviously the most natural substance we can fering from those of dopamine.14 Among these is
have for treating what I should like to call “the stria- dose-related down-regulation of the enzyme that cata-
tal dopamine deficiency syndrome.” However, it is lyzes its conversion to dopamine, L-aromatic amino
quite clear to me as a pharmacologist that, whatever acid decarboxylase (L-AAAD).15
the mode and site of its action, L-dopa is far from Another important aspect of L-dopa pharmacology
being perfect as a drug.”9 has to do with its nonenzymatic conversion to several
His early recognition of L-dopa’s limitations has been biologically-active compounds. Among these are 2,4,5-
borne out in everyday experiences by many PD patients
trihydroxyphenylalanine (TOPA) and TOPA-quinone,
who, within a few years of starting treatment, experi-
each arising spontaneously from L-dopa in physiologi-
ence inconsistencies in its benefits. A starting point for
cal solutions.16 In addition to their excitotoxic proper-
understanding the problems associated with L-dopa ther-
ties (which can pose a threat to the survival of
apy is to shift the focus of inquiry beyond the actions of
dopaminergic neurons), these compounds may actually
dopamine created by L-dopa. The pathophysiology of
PD involves far more than a selective loss of striatal function as an integral component of normal L-dopa
dopaminergic signaling. Several neurotransmitter sys- pharmacology. This possibility has been explored in
tems—among them, the other monoamine neurotrans- brain slice research demonstrating the production of
mitters along with a number of compounds engaged in TOPA and TOPA-quinone immediately after L-dopa
neurotransmission and neuromodulation—also undergo administration.17 These compounds initiated various
substantial depletion and altered functioning in the PD neuronal responses in dopaminergic pathways (among
brain. Combined interplay of these signaling compo- them increased neuronal firing, inward current flux,
nents results in the unique motor and behavioral features and membrane depolarization) that were not attribut-
of PD. For example, besides L-dopa, restoring motor able to and preceded the other effects generated by
function in an experimental model of Parkinsonism (the dopamine synthesis. As an indication of their excito-
dopamine-depleted rat) requires the additional presence toxic mechanism, the actions induced by TOPA and

2 Movement Disorders, Vol. 00, No. 00, 2014

 L E V O D O P A P H A R M A C O K I N E T I C S A N D P H A R M A C O D Y N A M I C S

TOPA-quinone could be blocked by administration of dosage block less than half of systemic COMT activ-
a glutamate antagonist.17 L-dopa also can react non- ity. Drugs inhibiting systemic L-AAAD activity outside
enzymatically to produce a conjugate with cysteine, 5-S- of the CNS are also limited in the extent of their
cysteinyldopa. 5-Cysteinyldopa is a substrate for under- actions. Although co-administration of CD or bensera-
going decarboxylation to form 5-S-cysteinyldopamine, a zide can markedly decrease the systemic conversion of
23
pharmacologically-active compound that imparts oxida- L-dopa to dopamine, neither drug accomplishes full
tive stress in dopaminergic neurons, where it also can inhibition when dosed at the conventional daily intake
promote the aggregation of alpha-synuclein.18 These and of 75 to 100 mg.8 Supplemental CD can enhance or
other studies highlighting the complexity of L-dopa’s accomplish further decarboxylase inhibition and, con-
neuropharmacology indicate that we have a lot more to sequently, increased L-dopa bioavailability.24 Like
learn about the actions of this drug in the PD brain. COMT inhibition, this approach to diminishing the
Understanding clinical phenomena associated with first-pass metabolism of LD can provide some clinical
L-dopa such as dyskinesias and dose-by-dose fluctuations
improvement, although generally not enough to abol-
ish motor fluctuations with immediate-release (IR)
also may benefit from recognition that this amino acid
L-dopa therapy. Human experiments testing an
behaves as more than just a dopamine precursor.
increased dosage of supplemental CD or benserazide
have shown the inability of these drugs to achieve full
inhibition of L-AAAD activity.25,26
The Challenge for L-dopa Transport
to the Brain
Enhancing Striatal Dopamine With
L-dopa has a complicated path to travel in its gastro- L-dopa
intestinal uptake and transfer to its site of action in
the brain. The facilitated absorption of L-neutral The unique role played by exogenously administered
amino acids such as L-dopa occurs in only a relatively L-dopa in generating dopamine arises from its entry
short territory of the duodenum and the proximal jeju- into the synthetic pathway past the physiological regu-
num. L-dopa transportation uses sodium-dependent latory step for dopamine synthesis, tyrosine hydroxy-
L-neutral amino acid carrier system that is situated in lase. Even in advanced disease, when the PD brain has
the gut wall and at the blood–brain barrier. At both lost most of its dopaminergic projections from sub-
locations, L-dopa competes for uptake with other stantia nigra to striatum, most investigations have
L-neutral amino acids derived from the breakdown of found sufficient L-AAAD activity remains for achiev-
protein and other dietary sources of amino acids.19,20 ing rapid and thorough conversion of L-dopa to dopa-
Most of the load of orally-administered L-dopa is mine. Where this enzymatic activity resides has been a
cleared by first-pass hepatic metabolism or by distribu- longstanding question. In the absence of dopaminergic
tion to skeletal muscle. Unless a catecholamine-O- nerve terminals, the generation of striatal dopamine
methyltransferase (COMT) inhibitor has been used, a from L-dopa might occur by means of L-AAAD
large fraction of an administered L-dopa dose under- located in striatal serotonergic nerve terminals27; alter-
goes conversion to 3-O-methyldopa (as well as small natively, enzymatic activity in various striatal inter-
amounts of 4-O-methyldopa). These compounds, neurons or even in glia might be responsible.28
which accumulate in the bloodstream to higher con- The adequacy of remaining striatal L-AAAD activity
centrations than those of typical L-dopa levels, also in advanced PD has been questioned as to a target for
can also compete with the uptake of L-dopa. Although improving L-dopa responsiveness when its clinical
3-O-methyldopa has been generally regarded as meta- effect seems to be on the decline. One line of investi-
bolically inert, some studies have shown it has gation has proposed improving the effectiveness of
pharmacological activity at even neurotoxicity.21 L-dopa by augmenting the synthesis of L-AAAD in
Blocking the metabolism of L-dopa as it travels striatum by means of a gene therapy intervention.
from gut to brain has had practical applications for Application of this approach in a proof of concept
improving clinical responses. Catecholamine-O- study in PD demonstrated the practicality of using
methyltransferase inhibition with entacapone or tolca- viral vector gene delivery to enhance L-AAAD activ-
pone results in an extension of anti-Parkinsonian ity.29 In these studies (which are in ongoing clinical
effect by approximately 20 to 30 minutes for each research), the initial findings showed improved consis-
L-dopa dose. Each of these drugs at conventional dos- tency of L-dopa clinical effect together with the dem-
age results in an approximately one-third increase of onstration of increased enzymatic activity by PET scan
L-dopa area-under the-time-plasma-concentration studies.30 Together with genes also for generating
curve (AUC).22 Although the COMT inhibitors offer tyrosine hydroxylase and its cofactor, introducing a
some useful extension of clinical effect against gene for augmenting L-AAAD synthesis into PD puta-
wearing-off responses, these drugs at conventional men has also been an intervention for a recent proof-

Movement Disorders, Vol. 00, No. 00, 2014 3

P E T E R A . L E W I T T

of-concept study, which showed that endogenous syn- administration of L-dopa and results from oral admin-
thesis of dopamine can be enhanced.31 istration of both IR and sustained-release formula-
The therapeutic concept underlying the use of mono- tions.36 In healthy volunteers, this study demonstrated
amine oxidase type-B (MAO-B) inhibitors is for the rapid plasma clearance of L-dopa. Maximal con-
extending the clinical actions from each dose of L-dopa centration (Cmax) after two CD-L-dopa 25-100 IR tab-
by retarding breakdown of the dopamine it produces. lets was 1.39 6 0.34 mg/mL for L-dopa and 165 6 77
This effect has been substantiated from extensive clini- ng/mL for CD (which was absorbed more slowly than
cal experience with MAO-B inhibitors that are mar- L-dopa). For L-dopa, AUC was 3.69 6 0.74 mg h
21

keted or have been under clinical investigation.32 21

mL . In the first hour after dosing, approximately
Besides MAO-B, other metabolic routes are involved in 90% of orally administered L-dopa was taken up and
the clearance of dopamine, however. The modest gains the process was near complete by 2 hours. Greater
that MAO-B inhibitors offer for the PD patient experi- delay in uptake was found with Sinemet CR, which
encing motor fluctuations have highlighted the need also had reduced bioavailability for both L-dopa and
for other strategies to achieve improved the constancy CD. Other important observations were that CD
of L-dopa effect. absorption was diminished in the presence of food and
that bioavailability of oral L-dopa in healthy young
L-dopa Pharmacokinetics and controls (84% 6 13) was lower than in older PD sub-
jects (99% 6 21). Disposition kinetic parameters with
Pharmacodynamics intravenous L-dopa showed an AUC (per mg L-dopa) of
21.7 ng hr21 mL21, an initial volume of distribution of
One of the major pharmacodynamic challenges for
optimizing PD therapeutics is dealing with motor fluc- 28.5 L, renal clearance of 72 mL/min, and plasma
tuations that commonly develop with orally-dosed clearance half-life of 1.8 hours (which was almost
L-dopa. The repeated emergence of “off” states, which
doubled by the addition of CD). Renal excretion of L-
for many patients with advanced PD can occupy up to dopa accounted for approximately 10% of total clear-
one-third or more of a typical waking day, is a prob- ance. With oral administration of carbidopa-L-dopa
lem of inadequate L-dopa delivery to the brain. Clini- 25/100 mg IR tablets, its time to maximal drug concen-
cal investigations have determined that providing tration (Tmax) was 0.8 6 0.4 hours in this study
sufficient L-dopa to overcome Parkinsonism requires (Fig. 1).36 Co-administration of L-dopa with either ben-
exceeding a minimal threshold of circulating L-dopa serazide or CD results in extension of its Tmax by
concentration to achieve reversal of signs and symp- approximately 30 minutes.37 Other investigations of
toms. Whereas “off” states can have other causes Tmax with oral dosing of IR CD-L-dopa have reported
besides insufficient L-dopa delivery (such as the phe- this to be 1.1 6 0.7 hours38 and 1.0 6 0.1 hour.39 The
nomenon of transient gait freezing), the most common latter study found that the coefficient of variation for L-
explanation for most motor fluctuations is nothing dopa concentration over a dose cycle (standard devia-
more than inconsistency of L-dopa arriving at the tion/mean L-dopa concentration) was 68.8 6 3.7%.
brain. In these circumstances, “off” episodes can be A recent study of L-dopa population pharmacokinetics
eliminated during continuous intravenous or enteral compared findings using the CD/L-dopa 25/100 mg IR
infusion of L-dopa.33 However, with current orally formulation with those from administering an intesti-
administered L-dopa products, achieving regularity in nally-infused suspension of the drugs.40 Previously, both
drug uptake to the brain is problematic. Even formula- one-compartment41,42 and two-compartment43 struc-
tions designed for achieving more sustained release of tural models have been used in carrying out L-dopa
L-dopa or concomitant inhibition of COMT have not pharmacokinetic analysis. In this population pharmaco-
been particularly successful.34 The essence of the prob- kinetic study, a two-compartment structure provided
lem is that the human gastrointestinal tract is not well the best data fit for reporting on pharmacokinetic
suited for optimizing the uptake of orally administered parameter estimates and measures of their accuracy.40
35
L-dopa. Episodic failure of L-dopa to provide benefit, The pharmacokinetics of L-dopa do not change sys-
a common problem for many patients, has proved dif- tematically during up to 4 years of chronic treat-
ficult to interpret amidst all the factors that influence ment,44 nor are there alterations of L-dopa disposition
uptake and systemic handling of the drug. in PD patients who experience motor fluctuations as
To explore the systemic disposition of L-dopa and compared with those with consistent L-dopa effects.45
the problems of consistent drug delivery, dozens of Several studies have confirmed that the pharmacoki-
L-dopa pharmacokinetic investigations have appeared netics of L-dopa exhibit considerable inter-subject vari-
in the medical literature since the first clinical trials ability. Postmenopausal women show much greater
with this drug. One of the most extensive investiga- L-dopa bioavailability (AUC from a single dose of IR
tions of L-dopa pharmacokinetics is a 1989 report com- CD-L-dopa, 25-100 mg) than men of similar age. This
paring pharmacokinetic data from intravenous finding was evident even after adjustment for

4 Movement Disorders, Vol. 00, No. 00, 2014

 L E V O D O P A P H A R M A C O K I N E T I C S A N D P H A R M A C O D Y N A M I C S

FIG. 1. Pharmacokinetics of L-dopa (left panel) and carbidopa (right panel) in 16 fasted healthy young volunteers. Data are mean plasma concentra-
tions (61 standard deviation) after oral administration of immediate-release CD-L-dopa 25-100 mg (2 tablets). Data taken from Yeh and colleagues
(Fig. 1).36

sex-related differences attributable to body weight. L-dopa pharmacokinetics include increased values of
The mean L-dopa Tmax did not differ between men Cmax and AUC as correlates of both low body mass
and women, however.46 Some of the major influences and increased age.56,57
on pharmacokinetic parameters are likely related to For most patients experiencing dose-by-dose fluctua-
the stomach’s ability for efficient transfer of the drug tions, monitoring of their plasma L-dopa concentration
to the absorptive surfaces of the duodenum and proxi- provides a close correlate of the observed anti-
mal jejunum. In general, this function is delayed in PD parkinsonian effect. Several studies have used timed
patients as compared with healthy controls.47,48 Gas- motor tasks as surrogates of parkinsonism (such as
tric emptying is further slowed by an effect conferred timed walking and finger tapping) to demonstrate
by L-dopa itself.49 pharmacokinetic–pharmacodynamic relation-
Some investigations have reported that gastric Heli- ships.38,44,45 In these circumstances, a typical L-dopa
cobacter pylori infection is a factor contributing to plasma concentration for initiating reversal of parkin-
motor fluctuations, and that improvement in consis- sonism effect is in the range of 0.8 to 1.1 mg/mL (4.1-
tency of L-dopa effect occurs once the infection is 5.6 mM). With administration of IR L-dopa (together
eradicated.50,51 However, a recent study of L-dopa with adequate dosage of decarboxylase inhibitor), the
pharmacokinetics in PD patients both with and with- duration over which the drug’s plasma concentration
out Helicobacter pylori infection does not confirm exceeds the “on” threshold is close to that of the
that disposition of L-dopa is necessarily changed by clearance half-life of IR CD-L-dopa— typically,
presence of the infection.52 Another factor interfering approximately 3 hours. The onset of clinical effect is
with L-dopa uptake and associated with motor fluctua- not exactly coincident with the timing of changes in
tions is bacterial overgrowth occurring in the small plasma L-dopa concentration; a hysteresis function has
intestine. After antibiotic treatment to restrict bacterial been demonstrated in which anti-parkinsonian actions
growth in the small intestine, motor fluctuations can lag by approximately 15 minutes.38 The prediction of
improve.53 clinical effects from measurements of plasma L-dopa
Gastric “housekeeping” function can be influenced concentration (such as timed hand tapping or walking)
by everyday experiences that can alter stomach empty- fit best to nonlinear modeling of the pharmacokinetic
ing, such as pH change by administration of antacids data.58,59 Correlations between Tmax and motor
or proton pump inhibitors. Intake of iron or other responses can vary greatly, as can the timing of onset
divalent metals (which chelate L-dopa) also can inter- of observed beneficial effects. For example, in one
fere with L-dopa uptake. Dietary factors such as fat study of IR CD-L-dopa (Fig. 2),60 the Tmax for maxi-
and L-neutral amino acid content contribute to mal improvement in timed hand tapping was 2.3 6 0.9
decreased L-dopa bioavailability and lower Cmax while hours, whereas the Tmax for global motor improve-
increasing Tmax.54,55 Other influences on orally-dosed ment was 1.1 6 0.6 hours.60

Movement Disorders, Vol. 00, No. 00, 2014 5

P E T E R A . L E W I T T

commonly is replaced by motor fluctuations that occur

regularly over a cycle corresponding to the half-life of
L-dopa plasma clearance (3 hours). Like L-dopa
plasma concentrations, the clinical effects also can be
capricious from dose to dose. The reason for the tran-
sition of L-dopa’s clinical effects is not well under-
stood. Observations of L-dopa treatment outcomes
have discerned two pharmacodynamic patterns after
L-dopa dosing: the “short-duration” response (SDR)
and the “long-duration” response (LDR). Along with
dyskinesias, these response patterns came to be recog-
nized shortly after the introduction of L-dopa.61 They
can be manifested by the same patient. The SDR,
which is characterized by a cycle of improvement of
parkinsonian control over a period of 3 to 5 hours
during dosing with IR CD/L-dopa, can be shown to
parallel the timing of rise and fall in circulating drug
concentration. This pattern appears to be the direct
translation into clinical effect of L-dopa delivered to
the striatum. In contrast, the LDR does not seem to
follow the plasma concentration of L-dopa and persists
for hours to days after the drug has been stopped.62
Evidence for the LDR also can be discerned during
therapy with dopaminergic agonists, indicating that a
mechanism specific to dopamine synthesis, vesicular
packaging, release, or re-uptake mechanisms is
unlikely to be involved.63 The time-course of LDR
undergoes a much slower evolution than the SDR,
sometimes requiring up to several weeks for its full
expression. Other phenomena can distinguish the two
FIG. 2. Pharmacodynamic modeling of actual and modeled mean patterns of L-dopa response: in patients experiencing
effect measurements versus mean plasma L-dopa concentration for dyskinesias, for example, these emerge during the
eight PD subjects receiving two doses (at 4-hour intervals) of cycle of a SDR but not as a feature of the LDR.64
controlled-release CD-L-dopa 50-200 mg (Sinemet CR 50-200).60
Immediately after blood sampling, tapping rate (the number of times in Close investigation of L-dopa–treated patients has esti-
60 seconds subjects alternately tapped levers placed 25 cm apart), mated that, of the drug’s total anti-parkinsonian
walking rate (time required to arise from a seat, walk 6 m, turn, and effect, the SDR constitutes between one third and one
return to a sitting position), and global parkinsonism rating (using a 25
to 0 scale of parkinsonian features [0 denoting normal movement half, whereas the rest originates from the LDR.63
capability and 25 signifying severe parkinsonism] and a 11 to 15 rat- The mechanism by which the SDR is enacted seems
ing for dyskinesia [with 11 denoting minimal involuntary movements
and 15 indicating severe dyskinesia]). Global scores were treated as easier to conceptualize than the enigmatic LDR. To
continuous variables. Walking and global rating scores were fitted those searching for an explanation of the LDR, the
best to a sigmoid Emax pharmacodynamic model, whereas tapping
sustained improvement of parkinsonian features as a
scores were fitted best to an Emax model. Plasma L-dopa concentra-
tions were modeled with effect measures using extended least- feature of LDR stands in contrast to the known phar-
squares regression analysis. The EC50 (half-maximal effective concen- macology of dopaminergic neurotransmission. The
tration) values for walking and global scores were identical, whereas
the EC50 for tapping score was substantially greater. Reproduced LDR cannot arise from a retained CNS depot of
from Nelson and colleagues (Fig. 1).60 dopamine or its precursor, because L-dopa undergoes
rapid turnover in the mammalian brain.65 Nor is
For some PD patients (especially in the first years there evidence for upregulation of dopamine receptor
after start of L-dopa therapy), relief of parkinsonian sensitivity as a possible explanation. Another possibil-
signs and symptoms persists without fluctuation and ity is that of a downstream effect mediated initially
irrespective of the rise and fall in circulating drug con- by dopaminergic stimulation and resulting in activa-
centration. Symptomatic control of parkinsonism in tion of transcription factors and cascades of
these circumstances—sometimes referred to as the immediate-early gene proteins (such as FosB, Jun-B,
“honeymoon” period—makes optimization of therapy c-Fos, and Zif268).66,67 Possibly LDR might originate
easy to accomplish. However, by 2 or more years after from remodeling of striatal outflow pathways as a
the start of L-dopa therapy, this pattern of response consequence of nonphysiological dopaminergic

6 Movement Disorders, Vol. 00, No. 00, 2014

 L E V O D O P A P H A R M A C O K I N E T I C S A N D P H A R M A C O D Y N A M I C S

replacement therapy, as has been hypothesized for the provides reliable consistency of therapeutic concen-
development of L-dopa–induced dyskinesias.68 trations of L-dopa.33,34 There is a promising future
Another explanation for what might constitute the for improving L-dopa therapy with products currently
LDR arises from the realm of learning theory, with under development. These options include a new
demonstration that L-dopa therapy leads to enhanced sustained-release oral CD–L-dopa formulation,71 two
72,73
capability for integrating “chunks” of information L-dopa pro-drugs, two gastric-retentive slow-
into more sustained sequences.69 release L-dopa formulations,74,75 an inhaled L-dopa
Among the highlights of L-dopa pharmacokinetic- delivery system,76 and a subcutaneous infusion system
pharmacodynamic research is a detailed analysis of using solubilized CD and L-dopa.77 These products
extensive pharmacokinetic data from standardized will join intestinally infused CD–L-dopa gel, a formu-
2-hour intravenous L-dopa infusions (at 1 mg/kg body lation currently marketed in Europe and elsewhere,
weight per hour).43 The goal of this datamining was and currently awaiting US regulatory approval.
to understand within-subject variability in L-dopa Although this L-dopa formulation offers fine control
plasma concentrations. From analysis of data from of plasma L-dopa concentration,33 drug administra-
multiple L-dopa infusions, conducted repeatedly during tion by means of a tube through the stomach wall is
intervals of 4 years,44 the authors found that two- not a practical treatment option for most patients
compartment rather than one-compartment models experiencing motor fluctuations.
tended to fit the pharmacokinetic data better. They The unmet needs of L-dopa therapeutics call for a
concluded that a large fraction of between-subject var- more detailed understanding of how this drug works.
iability in plasma L-dopa concentration could be The relative contributions of daily L-dopa intake, dura-
explained by body weight. Also, the variability in pre- tion of drug exposure, and severity of parkinsonian
dicted peak plasma L-dopa concentrations could be features has been difficult to tease out in explaining
largely explained by variation in L-dopa’s central com- the development of dose-by-dose fluctuations (i.e., loss
partment volume in the two-compartment model. The of the LDR) and dyskinesias. A recent study that com-
authors summarized their analysis of L-dopa pharma- pared outcomes from starting L-dopa in Ghanaian and
cokinetic parameters and within-subject variability Italian PD patients observed that motor fluctuations
(using data from as many as 10 pharmacokinetic stud- and dyskinesias did not develop as correlates of L-dopa
ies conducted in the same individuals) with the conclu- use duration but, rather, in association with the extent
sion that these results could not be explained by either of disease progression.78 The message from this study
a systematic (predictable) change in association with (as from several other investigations3) is that, in mild
the increasing years or parkinsonism, or else random PD, delaying the start of L-dopa is unwarranted as a
occurrence. A prior study of these pharmacokinetic means for better long-term benefits from this drug.
findings showed no change over the first 4 years of Some of the most pressing questions for improving PD
L-dopa treatment as an explanation for the evolution
outcomes include how the initial “honeymoon period”
of motor fluctuations.44 of L-dopa therapy can be sustained, and how to avoid
the development of dyskinesias. Fortunately, there is
no shortage of curiosity among researchers for advanc-
Conclusions ing today’s knowledge about L-dopa’s pharmacology
into future treatment for PD.
For all its imperfections as a pharmaceutical, orally
administered L-dopa continues to be the major treat-
ment option for managing PD.70 With conventional References
formulations of L-dopa, irregular absorption and
1. Cotzias GC, Van Woert MH, Schiffer LM. Aromatic amino acids
rapid catabolism are the basis for many of the prob- and modification of parkinsonism. N Engl J Med 1967;276:374-379.
lems associated with its chronic use. Extensive study 2. Cotzias GC, Papavasiliou PS, Gellene R. Modification of Parkin-
of L-dopa pharmacology has provided evidence that sonism: chronic treatment with l-dopa. N Engl J Med 1969;280:
337-345.
this drug behaves as more than just a transient meta-
3. Olanow CW, Agid Y, Mizuno Y, et al. Levodopa in the treatment
bolic intermediate in the pathway of catecholamine of Parkinson’s disease: the second consensus meeting. Mov Disord
synthesis. Optimal dosing regimens can vary greatly 2004;19:997-1005.
among PD patients, as can the interplay of SDR and 4. Carlsson A, Lindqvist M, Magnusson T. 3,4-Dihydroxyphenylala-
nine and 5-hydroxytryptophan as reserpine antagonists. Nature
LDR. Several marketed L-dopa products purport to 1957:180:1200.
offer more continuous dopaminergic stimulation than 5. Hornykiewicz O. A brief history of levodopa. J Neurol 2010;
the IR formulation. However, pharmacokinetic analy- 257(Suppl 2):S249-S252.
sis of Sinemet CR (Merck and Co., Inc., Whitehouse 6. Oster KA, Sorkin SZ. The effects of intravenous injection of L-
dopa upon blood pressure. Proc Soc Exp Biol Med 1942;51:67-71.
Station, NJ, USA), Madopar HBS (F. Hoffmann-La
7. €
Holtz P, Credner K, Kroneberg G. Uber das sympathetikomimetis-
Roche AG, Basel, Switzerland), and Stalevo (Novartis che pressorische Prinzip des Hirns (“Urosympathin”). Arch Exp
AG, Basel, Switzerland) has shown that none of them Path Pharmak 1947;204:28-243.

Movement Disorders, Vol. 00, No. 00, 2014 7

P E T E R A . L E W I T T

8. LeWitt PA. The pharmacology of levodopa in treatment of Parkin- 30. Mittermeyer G, Christine CW, Rosenbluth KH, et al. Long-term
son disease: an update. In: Calne DB, ed. Drugs for the Treatment evaluation of a phase 1 study of AADC gene therapy for Parkin-
of Parkinson’s Disease. Handbook of Experimental Pharmacology, son’s disease. Hum Gene Ther 2012;23:377-381.
vol 88. Berlin: Springer-Verlag; 1989:325-384.
31. Palfi S, Gurruchaga JM, Ralph GS, et al. Long-term safety and tol-
9. Hornykiewicz O. How does L-dopa work? In Barbeau A, McDo- erability of ProSavin, a lentiviral vector-based gene therapy for
well FH, eds. L-dopa and Parkinsonism. Philadelphia: FA Davis; Parkinson’s disease: a dose escalation, open-label, phase 1/2 trial.
1970:393-399. Lancet Neurol 2014;383:1138-1146.
10. Ostock CY, Lindenbach D, Goldenberg AA, Kampton E, Bishop 32. LeWitt PA. MAO-B inhibitor know-how: back to the pharm. Neu-
C. Effects of noradrenergic denervation by anti-DBH-saporin on rology 2009;72:152-157.
behavioral responsivity to L-DOPA in the hemi-parkinsonian rat.
33. Nyholm D. Pharmacokinetic optimisation in the treatment of
Behav Brain Res 2014;270:75-85.
Parkinson’s disease: an update. Clin Pharmacokinet 2006;45:109-
11. Rommelfanger KS, Weinshenker D. Norepinephrine: the red- 136.
headed stepchild of Parkinson’s disease. Biochem Pharmacol 34. LeWitt PA, Jennings D, Lyons KE, et al. Pharmacokinetic-pharma-
2007;74:177-190. codynamic crossover comparison of two levodopa extension strat-
12. Hefti F, Enz A, Melamed E. Partial lesions of the nigrostriatal egies. Mov Disord 2009;24:1319-1324.
pathway in the rat: acceleration of transmitter synthesis and release 35. Melamed E, Bitton V, Zelig O. Episodic unresponsiveness to single
of surviving dopaminergic neurones by drugs. Neuropharmacology doses of L-dopa in parkinsonian fluctuators. Neurology 1986;36:
1985;24:19-23. 100-103.
13. Misu Y, Kitahama K, Goshima U. L-3,4-Dihydroxyphenylalanine 36. Yeh YC, August TF, Bush DF, et al. Pharmacokinetics and bioa-
as a neurotransmitter candidate in the central nervous system. vailability of Sinemet CR: a summary of human studies. Neurology
Pharmacol Ther 2003;97:117-137. 1989;39(11 Suppl 2):25-38.
14. Misu Y, Goshima Y. DOPA as a neurotransmitter candidate. In: 37. Bredberg E, Tedroff J, Aquilonius SM, et al. Pharmokinetics and
Misu Y, Goshima Y, eds. Neurobiology of DOPA as a Neuro- effects of levodopa in advanced Parkinson’s disease. Eur J Clin
transmitter. Boca Raton, FL: CRC Press; 2006:23-34. Pharmacol 1990;39:385-389.
15. King JM, Muthian G, Mackey V, Smith M, Charlton C. L-Dihy- 38. LeWitt PA, Nelson MV, Berchou RC, et al. Controlled release car-
droxyphenylalanine modulates the steady-state expression of bidopa/levodopa (Sinemet 50/200 CR-4). Clinical, pharmacoki-
mouse striatal tyrosine hydroxylase, aromatic L-amino acid decar- netic, and pharmacodynamic studies. Neurology 1989;39(11 Suppl
boxylase, dopamine and its metabolites in an MPTP mouse model 2):45-53.
of Parkinson’s disease. Life Sci 2011;89:638-643.
39. Cedarbaum JM, Kutt H, McDowell FH. A pharmacokinetic and
16. Newcomer TA, Rosenberg PA, Aizenman E. Iron-mediated oxida- pharmacodynamic comparison of Sinemet CR (50/200) and stand-
tion of 3,4-dihydroxyphenylalanine to an excitotoxin. J Neurochem ard Sinemet (25/100). Neurology 1989;39(11 Suppl 2):38-44.
1995;64:1742-1748.
40. Othman AA, Dutta S. Population pharmacokinetics of levodopa in
17. Guatteo E, Yee A, McKearney J, et al. Dual effects of subjects with advanced Parkinson’s disease: levodopa-carbidopa
L-DOPA on nigral dopaminergic neurons. Exp Neurol 2013;247: intestinal gel infusion vs. oral tablets. Br J Clin Pharmacol 2014;
582-594. 78:94-105.
18. Aureli C, Cassano T, Masci A, et al. 5-S-cysteinyldopamine neuro- 41. Triggs EJ, Charles BG, Contin M, et al. Population pharmacoki-
toxicity: influence on the expression of a-synuclein and ERp57 in netics of oral levodopa in parkinsonian patients. Eur J Clin Phar-
cellular and animal models of Parkinson’s disease. J Neurosci Res macol 1996;51:59-67.
2014;92:347-358.
42. Jorga K, Banken L, Fotteler B, Snell JL, Steimer JL. Population
19. Nutt JG, Fellman JH. Pharmacokinetics of levodopa. Clin Neuro- pharmacokinetics of levodopa in patients with Parkinson’s disease
pharmacol 1984;7:35-49. treated with tolcapone. Clin Pharmacol Ther 2000;67:610-620.
20. Leenders KL, Poewe WH, Palmer AJ, Brenton DP, Frackowiak RS. 43. Chan PLS, Nutt JG, Holford NHG. Importance of within subject
Inhibition of L-[18F]fluorodopa uptake into human brain by amino variation in levodopa pharmacokinetics: a 4 year cohort study in
acids demonstrated by positron emission tomography. Ann Neurol Parkinson’s disease. J Pharmacokin Pharmacodynam 2005;32:
1986;20:258-262. 307-331.
21. Lee ES, Chen H, King J, Charlton C. The role of 3-O-methyldopa 44. Nutt JG, Carter JH, Lea ES, Sexton GJ. Evolution of the response
in the side effects of L-DOPA. Neurochem Res 2008;33:401-411. to levodopa during the first four years of therapy. Ann Neurol
22. Kaakkola S. Problems with the present inhibitors and a relevance 2002;51:686-693.
of new and improved COMT inhibitors in Parkinson’s disease. Int 45. Contin M, Riva R, Martinelli P, et al. Pharmacodynamic modeling
Rev Neurobiol 2010;95:207-225. of oral levodopa: clinical application in Parkinson’s disease. Neu-
23. Nutt JG, Woodward WR, Anderson JL. Effect of carbidopa on rology 1993;43:367-371.
pharmacokinetics of intravenously administered levodopa: implica- 46. Kompoliti K, Adler CH, Raman R, et al. Gender and pramipexole
tions for mechanism of action of carbidopa in the treatment of effects on levodopa pharmacokinetics and pharmacodynamics.
Parkinsonism. Ann Neurol 1985;13:537-544. Neurology 2002;58:1418-1422.
24. Cedarbaum JM, Kutt H, Dhar AK, Watkins S, McDowell FH. 47. Edwards LL, Quigley EM, Pfeiffer RF. Gastrointestinal dysfunction
Effect of supplemental carbidopa on bioavailability of L-dopa. in Parkinson’s disease: frequency and pathophysiology. Neurology
Clin Neuropharmacol 1986;9:153-159. 1992;42:726-732.
25. Ward CD, Trombley IK, Calne DB, Kopin IJ. L-dopa decar- 48. Hardoff R, Sula M, Tamir A, et al. Gastric emptying time and gas-
boxylation in chronically treated patients. Neurology 1984;34: tric motility in patients with Parkinson’s disease. Mov Disord
198-201. 2001;16:1041-1047.
26. Dingemanse J, Kleinbloesem CH, Z€ urcher G, et al. Pharmacody- 49. Robertson DR, Renwick AG, Wood ND, et al. The influence of
namics of benserazide assessed by its effects on endogenous and levodopa on gastric emptying in man. Br J Clin Pharmacol 1990;
exogenous levodopa pharmacokinetics. Br J Clin Pharmacol 1997; 29:47-53.
44:41-48.
50. Rees K, Stowe R, Patel S, et al. Helicobacter pylori eradication for
27. Tanaka H, Kannari K, Maeda T, et al. Role of serotonergic neu- Parkinson’s disease. Cochrane Database Syst Rev 2011;11:
rons in L-DOPA-derived extracellular dopamine in the striatum of CD008453.
6-OHDA-lesioned rats. Neuroreport 1999;10:631-634.
51. Rahne KE, Tagesson C, Nyholm D. Motor fluctuations and Heli-
28. Hefti F, Melamed E, Wurtman RJ. The site of dopamine formation cobacter pylori in Parkinson’s disease. J Neurol 2013;260:2974-
in rat striatum after L-dopa administration. J Pharmacol Exp Ther 2980.
1981;217:189-197.
52. Naro_zanska E, Białecka M, Adamiak-Giera U, et al. Pharmacoki-
29. Eberling JL, Jagust WJ, Christine CW, et al. Results from a phase netics of levodopa in patients with Parkinson disease and motor
I safety trial of hAADC gene therapy for Parkinson disease. Neu- fluctuations depending on the presence of Helicobacter pylori
rology 2008;70:1980-1983. infection. Clin Neuropharmacol 2014;37:96-99.

8 Movement Disorders, Vol. 00, No. 00, 2014

 L E V O D O P A P H A R M A C O K I N E T I C S A N D P H A R M A C O D Y N A M I C S

53. Fasano A, Bove F, Gabrielli M, et al. The role of small intestinal striatum by acute or chronic L-dopa treatment. Neuroscience
bacterial overgrowth in Parkinson’s disease. Mov Disord 2013;28: 1999;94:515-527.
1241-1249.
67. Ebihara K, Ishida Y, Takeda R, et al. Differential expression of
54. Nyholm D, Lennernas H, Gomes-Trolin C, et al. Levodopa phar- FosB, c-Fos, and Zif268 in forebrain regions after acute or chronic
macokinetics and motor performance during activities of daily liv- L-DOPA treatment in a rat model of Parkinson’s disease. Neurosci
ing in patients with Parkinson’s disease on individual drug Lett 2011;496:90-94.
combinations. Clin Neuropharmacol 2002;25:89-96.
68. Riverol M, Ord on
~ez C, Collantes M, et al. Levodopa induces long-
55. Nutt JG, Woodward WR, Hammerstad JP, et al. The “on-off” lasting modification in the functional activity of the nigrostriatal
phenomenon in Parkinson’s disease: relation to levodopa absorp- pathway. Neurobiol Dis 2014;62:250-259.
tion and transport. N Engl J Med 1984;310:483-488. 69. Tremblay PL, Bedard MA, Langlois D et al. Movement chunking
56. Robertson DR, Wood ND, Everest H, et al. The effect of age on during sequence learning. Is a dopamine-dependent process: a
the pharmacokinetics of levodopa in patients with Parkinson’s dis- study conducted in Parkinson’s disease. Exp Brain Res 2010; 205:
ease. Eur J Clin Pharmacol 1989;28:61-69. 375-385.
57. Contin M, Riva R, Martinelli P, et al. Effect of age on the pharma- 70. LeWitt PA. Levodopa for the treatment of Parkinson’s disease. N
cokinetics of oral levodopa in patients with Parkinson’s disease. Engl J Med 2008;359:2468-2476.
Eur J Clin Pharmacol 1991;41:463-466. 71. Hauser RA, Hsu A, Kell S, et al. Extended-release carbidopa-
58. Nelson MV, Berchou RC, LeWitt PA, et al. Pharmacokinetic and levodopa (IPX066) compared with immediate-release carbidopa-
pharmacodynamic modeling of levodopa plasma concentrations levodopa in patients with Parkinson’s disease and motor fluctua-
and clinical effects in Parkinson’s disease after Sinemet Clin Neu- tions: a phase 3 randomised, double-blind trial. Lancet Neurol
ropharmacol 1989;12:91-97. 2013;12:346-356.
59. Contin M, Riva R, Albani F, et al. Pharmacokinetic optimization 72. LeWitt PA, Huff FJ, Hauser R, et al. Double-blind study of an
in the treatment of Parkinson’s disease. Clin Pharmacokinet 1996; actively-transported levodopa prodrug, XP21279, in Parkinson dis-
30:463-481. ease. Mov Disord 2014;29:75-82.
60. Nelson MV, Berchou RC, LeWitt PA, et al. Pharmacodynamic 73. Stocchi F, Marconi S. Factors associated with motor fluctuations
modeling of concentration-effect relationships after controlled- and dyskinesia in Parkinson disease: potential role of a new mele-
release carbidopa /levodopa (Sinemet CR4) in Parkinson’s disease. vodopa plus carbidopa formulation (Sirio). Clin Neuropharmacol
Neurology 1990;40:70-74. 2010;33:198-203.
61. Muenter MD, Tyce GM. L-dopa therapy of Parkinson’s disease: 74. LeWitt PA, Giladi N, Gurevich T, et al. Accordion Pill carbidopa/
plasma L-dopa concentration, therapeutic response, and side levodopa (AP CD/LD) for treatment of advanced Parkinson’s dis-
effects. Mayo Clin Proc 1971;46:231-239. ease (PD). Mov Disord. 2014;29(suppl 1): S248.
62. Zappia M, Oliveri RL, Montesanti R, et al. Loss of long-duration 75. Chen C, Cowles VE, Sweeney M, Stolyarov ID, Illarioshkin SN.
response to levodopa over time in PD: implications for wearing- Pharmacokinetics and pharmacodynamics of gastroretentive deliv-
off. Neurology 1999;52:763-767. ery of levodopa/carbidopa in patients with Parkinson disease. Clin
Neuropharmacol. 012;35:67-72.
63. Anderson E, Nutt J. The long-duration response to levodopa: phe-
nomenology, potential mechanisms and clinical implications. Par- 76. Freed MI, Moore JA, Batycky R, Tia D-F, CVT-301-002 Study
kinsonism Relat Disord. 2011;17:587-592. Investigators. Pharmacokinetics (PK) following inhaled levodopa
delivery with CVT-301: Rapid augmentation of systemic levodopa
64. Quattrone A, Zappia M, Agualia U, et al. The subacute levodopa (LD) levels and improvement in motor function in PD patients
test for evaluating long-duration response in Parkinson’s disease. with motor fluctuations. Mov Disord. 2014;29(suppl 1):S239.
Ann Neurol 1995;38:389-395.
77. Caraco Y, Oren S, LeWitt P. Constant therapeutic levodopa (LD)
65. Doller HJ, Connor JD. Changes in neocortical dopamine concen- plasma concentrations maintained by continuous subcutaneous
trations in response to levodopa infusions. J Neurochem 1980;34: (SC) administration of ND-0612, a novel formulation of LD/carbi-
1264-1269. dopa (CD). Mov Disord 2013;28 (suppl 1):S162.
66. Cenci MA, Tranberg A, Andersson M, Hilbertson A. Changes in 78. Cilia R, Akpalu A, Sarfo FS, et al. The modern pre-levodopa era
the regional and compartmental distribution of FosB- and JunB- of Parkinson’s disease: insights into motor complications from sub-
like immunoreactivity induced in the dopamine-denervated rat Saharan Africa. Brain 2014; 137; 2731-2742.

Movement Disorders, Vol. 00, No. 00, 2014 9