39, No. 4, 1993. 629. Measuring L-Dopa in Plasma and Urine to Monitor Therapy of Elderly Patients with. Parkinson Disease Treated with L-Dopa and a Dopa ...
CHEM. 39/4,
CLIN.
Measuring Parkinson J. Dutton,’ have
We
plasma
629-634
(1993)
L-Dopa in Plasma and Urine to Monitor Therapy of Elderly Patients Disease Treated with L-Dopa and a Dopa Decarboxylase Inhibitor L. G. Copeland,2
established
and urine,
and L-dopac,
J. R. Ptayfer,2
a method
including
and
for measuring
using separation
by ion-pair
reversed-phase
with an electrochemical
The
to the therapeutic
was
applied
in
dopamine
HPLC and quantification assay
N. B. Roberts”3
L-dopa
the metabolites
detector.
monitoring
of
elderly patients with established Parkinson disease being treated with L-dopa plus a dopa decarboxylase inhibitor. Plasma L-dopa was evaluated in relation to dosage and postdose sampling time in 71 outpatients with Parkinson disease. L-DOpa concentrations were greatest in the patients taking the highest dosages prescribed and decreased significantly with increasing time after postdose sampling. Comparison of plasma L-dopa concentrations with a published therapeutic range established by intravenous
administration
of L-dopa
was helpful
in assessing
the suitability of each patient’s drug dosage, assessing patients’ compliance, and avoiding overdosage but was not useful in the overall clinical assessment of progression of disease or of the long-term therapeutic response. Urine measurements confirmed the plasma concentrations but showed no further advantage. The recommended time for sample collection is between 1.5 and 3 h after the first morning dose. Plasma is the preferred matrix but if blood sampling is difficult, particularly from elderly! infirm used.
individuals,
Indexing phase
.
L-Dopa
an
untimed
Terms: genatric electrochemical
detection
is the
mainstay
urine
chemistr y
collection
could
be
reversedcompliance
chromatography,
monitoring
of treatment
for patients
with
Parkinson disease (1); indeed, a positive response to L-dopa therapy may help confirm the diagnosis (2). Pharmacokinetic studies indicate that the maximum clinical response is related to the blood concentration of L-dopa (3), and drug infusion studies have shown that maintenance 1.6 mg/L
of L-dopa concentrations between 0.3 and the most consistent response (4). However, many tissues in the body have the ability to enzymatically decarboxylate L-dopa to produce dopamine (5). The introduction of a selective L-dopa decarboxylase inhibitor, either carbidopa or benzerazide, has been included in L-dopa therapy to inhibit this process and thereby reduce the associated peripheral side effects of increased dopamine production, such as postural hypotension and nausea, and also to increase the bioavailability of the parent drug to the brain (6). Because
gives
L-dopa
is
a pro-drug
Departments of’ Clinical Chemistry oyal Liverpool University Hospital, 3Author for correspondence. Received January 6, 1992; accepted
that
must
enter
the
and 2 Geriatric Medicine, Liverpool L7 8XW, UK. November
10, 1992.
with
central within effects,
nervous system and undergo decarboxylation the striatum before it produces pharmacological routine monitoring of drug and metabolite
centrations in plasma has We, however, believe that important cause of variation larly in elderly patients
therapy.
can often take less than and a few increase their dosage inappropriately to toxic concentrations. We have therefore investigated the development of a procedure for quantifying L-dopa and its major metabolites dopainine and L-dopac in plasma. We applied the assay to samples from patients with Parkinson disease seen in a routine outpatient clinic and assessed the usefulness of the drug measurement by examining individual responses to therapy. We also measured L-dopa and metabolites in urine to determine whether this would add any further information to assist in the interpretation and management of therapeutic drug control. Various studies have recently demonstrated that the renal decarboxylase enzyme converting L-dopa to dopemine (dopa decarboxylase; aromatic-L-amino-acid detheir
Many
prescribed
such
con-
not been considered useful. drug compliance may be an in drug response, particureceiving long-term L-dopa
patients
dose,
carboxylase, EC 4.1.1.28) is inhibited by carbidopa and is associated with significant decreases in urinary sodium, in both an experimental rat model (7) and healthy humans (8, 9). It is of considerable interest, therefore, to know whether long-term therapy with L-dopa, in combination with a dopa decarboxylase inhibitor, would continue to block the renal conversion of L-dopa to dopaniine and what effect this might have on the subsequent urinary excretion of sodium and water.
MaterIals
and Methods
The chemicals used were ANALAR-grade (Sigma Chemical Co., Poole, Dorset, UK), and the solvents for chromatography were HPLC-grade (Rathburn, Walkerburn, UK). De-ionized water (Spectrum System, Elga, High Wycombe, UK) was used throughout. The catechol compounds were separated on a 25 X 0.45 cm column of Ultratechsphere ODS or Spheriaorb ODS (5-j.tm particles; HPLC Technology, Macclesfield, UK) by ion-pair chromatography. The mobile phase contained, per liter, 75 mmol of citric acid, 58.5 mmol of sodium dihydrogen orthophosphate, 0.2 mmol of disodium EDTA, and 4.4 mmol of heptanesulfonic acid, carefully adjusted to pH 3.4 and made to a final volume of 2.0 L with de-ionized water. To this we added 200 mL of methanol, then filtered the solution through a 0.45-tim (pore-size) filter and degassed it before use. The HPLC system included a Kratos pump (Model 770; Applied
Biosystems CLINICAL
Inc.,
Warrington,
CHEMISTRY,
UK)
operated
Vol. 39, No. 4, 1993
at a 629
flow rate of 1.0 mL/min and fitted with a pulse-dampening device. The catechol compounds were detected electrochemically with a Coulochem electrochemical detector (ESA Analytical Ltd., Huntingdon, UK). The instrumental settings adopted were as follows: conditioning cell oxidation potential + 0.35 V, first electrode oxidation potential +0.05 V, and the second electrode reduction potential -0.35 V, with gain settings at 100 x 2 for patient monitoring (plasma and urine) and 100 x 9 for untreated controls (plasma). The individual catecholamine compounds were identified by comparing their retention time (we used a computing integrator CI 10; LDC, Stone, UK) with that of standards. The computing integrator settings were 10 mV full-scale, attenuation 3. The unconjugated (free) catechol compounds from plasma or urine were selectively extracted onto alumina (10) before quantification from 0.2 mL of plasma (undiluted) and 0.2 mL of urine (diluted 50-fold with sodium chloride, 9 g/L) for patients taking L-dopa and from 2.0 mL of plasma and 1.0 mL of urine for the healthy control subjects not receiving therapy. Extraction losses were corrected for by use of the internal standard, dihydroxybenzylamine (10). Calibration of the assay for each compound was carried out with standards prepared in human plasma or urine. The assay was calibrated between 0 and 4 gfL for the estimation of plasma concentrations (to include norepinephrine and epinephrine) in nontreated control subjects and between 0.01 and 2.0 mgfL in plasma and urine for the patients receiving L-dopa therapy and for the assay of urine samples from the untreated group (Figure 1). Aliquots of urine were analyzed for sodium and potassium by flame emission spectrometry and for urea and creatinine by standard procedures (Technicon RAXT; Technicon, Tarrytown, NY). The urine data were expressed as mass of catechol per mole of creatinine. We studied 71 patients with Parkinson disease; their clinical status and drug regimen are outlined in Table 1. Each patient received a full clinical examination; muscular rigidity, tremor, and bradykinesia were each assessed on a five-point rating scale. Patients were scored on the Hoehm and Yahr symptom severity scale (11). All observations were made without knowledge of the plasma L-dopa concentration. An overall impression of whether the patient was responding to his or her medication was then made and the response was reported as very good to poor. The patients’ drug dosage and the time of the morning dose was noted, and a timed 10-mL sample of blood taken, usually after an overnight fast, into lithium heparin tubes that were kept in ice-cold water. The plasma was separated within 30-60 mm and stored at -30 to -40 #{176}C. A midmorning urine sample was also collected, the time noted, and maintained in ice-cold water before storage in the laboratory at -30 to -40 #{176}C. The compounds collected and stored under these conditions have been shown to be stable for at least 3 months (10). The group of control patients was 12 elderly patients
630
CLINICAL CHEMISTRY,
Vol. 39, No. 4, 1993
Table
1. ClinIcal
DetaIls of PatIents DIsease
years6
Duration of disease,
Parkinson
Women
Men
Both sixes
37
34
71
Number Age,
wIth
78±6.7
73±7
(58-90)
(60-84)
yearsa
6
±
3.3
(1-15)
Hoehn-Yahr
score”
II
16
11
III
14
15
IV
2
8
L-Dopa
dosage,
mg/day
150-200(22)
()C
300-400 500
(33)
(2)
600-700(10) 800-1250 (4) a
Mean
±
SD (and range).
“Hoehn and Vahr scores (11): II, bilateral midline involvement without impairment of balance; Ill, mild to moderate disability with only slight restriction in activities and able to lead an independent life; IV, severely disabling disease, markedly incapacitated but still able to walk or stand unassisted. c Patients on a specified drug dosage for at least 1 year.
(ages
65-85
years;
7 men,
5 women)
attending
an
outpatient clinic for various conditions, e.g., angina, ischemic heart disease, and diabetes. None was taking L-dopa or had any previous history of Parkinson disease or any other significant central neurological dysfunction. Healthy laboratory control subjects (n = 19, 10 men, 9 women; ages 20-45 years) had no evidence or history of previous disease, medication, e.g., for diabetes normal renal function and Technicon SMAC).
were not on any form of or lipid disease, and had biochemical proffles (by
Results The
simultaneous
detection
of
L-dopa,
L-dopac,
and
dopamine in samples of plasma or urine was established after an alumina cleanup and separation of the ion-pair complexes by reversed-phase HPLC (Figure 1, top). Table 2 shows the retention times for these and other compounds and confirms that the chromatographic procedure can be used for their accurate quantification. Each catechol compound responded differently under the electrochemical conditions used and therefore had to be calibrated separately. Figure 1 (bottom) shows a typical calibration graph adapted for application to L-dopa monitoring in patients taking this drug. The assay performance characteristics (Table 3) indicate acceptable precision with plasma or urine for both withinand between-batch assays. The precision of analysis (CV) for each compound throughout the range of assay varied between 5% and 11% for within-batch analysis and between 7.0% and 12.0% for between-batch analysis.
Analytical
recoveries
of
L-dopa,
dopac,
dopamine from plasma (n = 35) and urine (n = 35), 0.2 mgfL of each compound added, were (mean ± SD, as follows: plasma, 103 ± 6, 92 ± 10, and 108 ± 11; urine, 108 ± 4, 95 ± 12, and 112 ± 8, respectively. 1.0 mgIL added these were: plasma, 96 ± 10, 80 ±
and
for %) an Fo 15
Table 2. Chromatographic Retention Times of Electroactlve Compounds and Metabolltes Related the Catecholamines
4
Ren
d
a
time,
Compound
Dihydroxyphenylglycol 3-Methoxy-4-hydroxymandelic L-Dopa-3,4-dihydroxyphenylalanine Iso-3-methoxy-4-hydroxymandelic
a
d 0
Tb S
is
0
0 m
ren mine
tlm.b
4.23
0.41
4.55
0.43
5.37 5.73
0.51
Norepinephnne
6.00
0.57
Metanephnne
6.95
0.66
3-Methoxy-4-hydroxyphenylglycol
6.98
0.663
Epinephrine 3-Methoxytyrosine
8.18
acid acid
0.54
lso-3-methoxy-4-hydroxyphenylglycol
10.10
0.78 0.88 0.96
Methoxytyrosine
10.52
0.99
Dihydroxybenzylamine Normetanephrine Dihydroxyphenylacetic acid
10.53
1.00
11.62
1.10
9.30
Dopamine
20
to
5-Hydroxyindoleacetic
acid
12.45
1.18
15.89 23.91
1.51 2.27
>30.00 5-Hydroxytryptamine 6Determined with a25 x 0.46 cm Ultratechsphere ODS column. b Relative to the internal standard dihydroxybenzylamine.
ii
250
oo
Table 3. Intra- and Interassay PrecisIon of AnalysIs for L-DOpa and Other Catecholamines In Serum (n = 6)
0
l_
0
50 (0
Mien,
I
cv, s
mL
C
L-Dopa
0.41 0.92
7.6 5.1
Dopamine
0.34
5.8
8.9
6.3
Dopac
1.01 0.25
10.1 10.1
00
50
0.81
Norepinephnne6
0 0.0
0.4
0.8
1.6 Concentration
2.0
mg/L
Example of chromatographic separation of catecholamine compounds on reversed-phase HPLC; (bottom) calibration relationships for L-dopa, L-dopac, and dopamine suitable for theraFig. 1. (Top)
0.3 x iO 0.15 x iO
6.4 5.8
9.6 9.1
9.9
6.1 11.3
7.6 12.4 Epinephnne6 n = 6 each. A serum was supplemented to contain two different concentrations to cover the therapeutic limits expected for patients receiving L.dOpa therapy. Bio-Rad control serum.
peutic drug monitoring of L-dopa Top (a) L-dopa, (b) norepinephrine, (c) epinephnne, ( dopamine, (IS) internal standard (dihydroxybenzylamine) at concentrations of 30,8.5,4.6, and 30 ug/t. (left) and 300, 85, 46, and 300 p.g/L (nght), respectively. L-Dopac elutes at --12.5 mm, between the IS and dopamine. Full-scale deflection, 5 mV. Bottom: the data are expressed as a percentage of the IS, which is added in concentrations sufficient to give a similar electrochemical response to that of the
highest-concentration
dopamine
standard
and
115± 12;andurine, 101 ± 6,98 ± 10, and 108±9. Plasma L-dopa concentrations for 104 analyses of samples from the 71 patients varied from 1.6 mg/L) suggested overdosage. Plasma L-dopa concentrations varied from 0.06 to 2.28 mg/L in nine patients who had had a good response to L-dopa on 375-750 mg/day; however, these values were not different from the concentrations of 0.1-0.9 mg/L observed in three patients, taking 375-825 mg/day, who had poor drug responsiveness. Thus, measurement of L-dopa was helpful in assessing adequate therapeutic concentrations but not in determining the patients’ long-term drug responsiveness. Plasma and urine L-dopa concentrations were signifCLINICAL
CHEMISTRY,
Vol. 39, No. 4, 1993
631
20
2.0
*
*
.1 Ib) 3.2
C
O
2.4
1.2 C
C
2
2
r
05 #{149}
,-0.31
...#{149}
0.8
,
150 DOSE
POST
270 TiME
350
*
12
0.4 150
POST
-
-0.45
270
350
450
DOSE
TIME
n,.
0 2.0
**
0.0
4.0
*
Plasma
0
2.4
E
C 2
*
-0.03
.5
#{149} #{149} 50.001
r-0.42
C
.0 .0
p50.1
50
150 POST
fl.S.
9
o II
:
0.0
*
6-
tI C
270 DOSE
350 liME..
450
0
50
leo POST
270 DOSE
310 liME
*
* * **
450
4
*
C C
-
.4
*
0.0
icantly positively correlated (P