ORIGINAL RESEARCH ARTICLE
Clio. Pharmacokioet. 23 (4): 311-320, 1992 0312-5963/92/00 I 0-0311 / $05.00/0 © Adis International Limited. All rights rese rved . CPK1223
The Effect of Age and Acetylator Phenotype on the Pharmacokinetics of Sulfasalazine in Patients with Rheumatoid Arthritis Allister J. Taggart. Barbara J. McDermott and Stanley D. Roberts Department of Therapeutics and Pharmacology, The Queen's University of Belfast, and Department of Rheumatology, Musgrave Park Hospital, Belfast, Northern Ireland
Summary
The pharmacokinetic disposition of sulfasalazine and its metabolites was studied in 8 young and 12 elderly patients with active rheumatoid arthritis. Equal numbers of slow and fast acetylators were included in each age group. Patients received enteric-coated sulfasalazine 2g daily for 21 days; specimens of serum and urine were collected for 96h after administration on days I and 21. The elimination half-life of sulfasalazine was greater in the elderly patients. Many disposition parameters of sulfapyridine differed in slow and fast acetylators; of greatest significance were the increased values of steady-state serum concentration in the slow acetylators. There was no effect of age on any sulfapyridine disposition parameters. Values for the steady-state serum concentrations of N-acetyl-5-acetylsalicylic acid were greater in elderly than in young patients. The metabolism of sulfapyridine was markedly affected by acetylator phenotype and this was reflected in the composition of sulfapyridine-related material in the urine. Thus, age is a determinant of the steady-state concentrations of salicylate moieties but acetylator phenotype plays a greater role in determining the serum concentration of sulfapyridine, which has greater therapeutic implications in rheumatology.
Sulfasalazine was first used in Sweden over 40 years ago for the treatment of 'rheumatic polyarthritis' and ulcerative colitis (Svartz 1942). An early noncomparative study produced unfavourable results in rheumatoid disease (Sinclair & Duthie 1948) but by the 1960s, the drug was widely accepted as effective therapy for inflammatory bowel disease (Misiewicz et al. 1965). It is only in the past 10 years that the drug has been firmly established as an effective second-line agent in rheumatoid arthritis (Neumann et al. 1983; Pullar et al. 1983). After oral administration, sulfasalazine is split by bacterial azo-reduction in the colon into sulfapyridine and mesalazine (5-aminosalicylic acid). The absorption and disposition of these metabo-
lites occur by complicated pathways: the systemic uptake of sulfapyridine is virtually complete and it is eliminated by polymorphic acetylation, hydroxylation and glucuronidation; mesalazine is subject to both capacity-limited presystemic and systemic acetylation and enterohepatic recirculation occurs (Klotz 1985). There is, therefore, considerable scope for interindividual variation in handling of the drug. Pharmacokinetic studies have been performed in healthy volunteers (Schroder & Campbell 1972) and in patients with inflammatory bowel disease (Azad Khan et al. 1982; Das et al. 1973), but there have been no comparable studies in those with rheumatoid arthritis. In the treatment of the elderly population with rheumatoid arthritis, investigation
312
of the disposition of sulfasalazine in relation to age is important; also of relevance is the influence of acetylator status and the pharmacokinetic profile of sulfapyridine, which may be the active moiety of the compound (Neumann et al. 1986; Pullar et al. 1985a).
Patients and Methods Patients The study, which was approved by the University Ethical Committee, involved patients diagnosed as having classical or definite rheumatoid arthritis according to the American Rheumatism Association diagnosis. Active disease was defined by the presence of an erythrocyte sedimentation rate >28 mm/h or C-reactive protein concentration >5 mg/L or plasma viscosity> 1.72 centipoise, and at least 2 of the following conditions: tenderness of >6 joints, swelling of> 3 joints, morning stiffness of >45 min duration, and Ritchie articular index > 20. Patients were excluded from the study if any of the following criteria held: significant disease or previous surgery of the bowel; treatment with a second-line antirheumatic agent during the 3 months before the study; current therapy with inhibitors or inducers of hepatic enzymes (e.g. corticosteroids, coumarins, oral contraceptives, cimetidine, phenytoin, barbiturates); current therapy with agents which could have a pharmacokinetic interaction with su1fasa1azine (Das & Dubin 1976) such as antibiotics, calcium, iron, oral hypoglycaemic drugs and those which affect gut motility; smoking> 10 cigarettes or drinking >3 pints of beer or its alcoholic equivalent daily; significant cardiovascular, renal or hepatic disease. Drug Administration and Sampling Schedule The patients fasted from 2200h on the night before the start of the study until 2h after administration of the drug on day 1. On this morning, an indwelling cannula was inserted into a forearm vein and the patient received a single 2g oral dose of enteric-coated sulfasalazine. A 10ml specimen of blood was drawn at each of the following times
Clin. Pharmacokinet. 23 (4) 1992
after administration: 0, 1, 3, 6, 9, 12, 24, 30, 36, 48, 60, 72 and 96h. Serum was separated and stored at -20°C. Urine excreted during 4 consecutive periods of 24h after administration was collected, each volume was recorded and a 20ml aliquot of each specimen was retained for analysis of the drug and metabolites. Creatinine clearance over the 24h period of day 1 was determined. Treatment with sulfasalazine 2g daily was continued from days 5 to 21. Patients were reviewed on days 8 and 15 to monitor compliance and adverse effects. The sampling schedule was repeated on days 21 to 25 and creatinine clearance was determined again. Patients were withdrawn from the study if any of the following occurred: withdrawal of consent; noncompliance; significant adverse effects including vomiting on days 1 or 21; urinary incontinence during the sampling periods. Analytical Methods Acetylator phenotype was assessed by the method of Evans (1969). High performance liquid chromatography (HPLC) analysis of sulfasalazine, sulfapyridine and its metabolites, N-acetyl-sulfapyridine and N-acetyl-5-hydroxysulfapyridine, was performed, and sulfapyridine-O-glucuronide and Nacetyl-sulfapyridine-O-glucuronide were codetermined with aglycones by the method of Ahnfelt et al. (1986). Only minor amounts are present in the ag1ycones and so, in calculating the concentrations of N-acetyl-sulfapyridine-O-glucuronide and sulfapyridine-O-glucuronide, it was assumed that only glucuronides were detected. Mesalazine and JII"acetyl-mesalazine were estimated according to Willoughby et al. (1982). The following gives the precision of each assay, with the concentration of the analyte (in mg/L) given in parentheses: 2.9% (1.1), 3.1 % (2.0), 3.4% (2.0), 7.2% (2.0), 13.5% (0.06), 2.1 % (0.07) in serum and 6.6% (1.1), 3.9% (5.0), 3.8% (5.0), 5.4% (5.0), 2.0% (0.9), 2.1 % (2.0) in urine for sulfasalazine, sulfapyridine, N-acetyl-sulfapyridine, N-acetyl-5-hydroxy-sulfapyridine, mesalazine and N-acetyl-mesalazine, respectively. Linear calibration curves were obtained down to 0.5 mg/L for sulfasalazine both in serum and in urine. For the
313
Sulfasalazine Pharmacokinetics
50.0
50.0
Young/fast
Young/slow
c:
~co
Q) ()
c: 0
()
E
2Q)
rJJ
12
24
36
48
50.0
60
72
84
96
12
24
36
50.0
Old/fast
48
60
72
84
96
Old/slow
c:
0
~
E 2l c: 8 E
~
rJJ
0.1 0.0 0
0.1 12
24
36
48 60 Time (h)
72
84
96
0.05 0
12
24
36
48 60 Time (h)
72
84
96
Fig. 1. Serum concentration (± SO) of sulfasalazine (SASP, e) and metabolites, sulfapyridine (SP, 6), N-acetyl-sulfapyridine (AcSP,O), N-acetyl-sulfapyridine-O-glucuronide (AcSPG, X) and N-acetyl-mesalazine (Ac-ME, 0) after oral administration of sulfasalazine 2g on day I of the treatment schedule in the 4 subgroups of patients: youngJfast; young/slow; old/fast; old/ slow.
sulfapyridine compounds, linear calibration curves were obtained down to I mg/L in serum and 2 mg/ L in urine. For mesalazine and N-acetyl-mesalazine, the calibration curves were linear down to 0.06 and 0.07 mg/L in serum, whereas they were linear down to 0.9 and 2.0 mg/L, respectively, in urine. Since pure glucuronides were not available, precision and sensitivity data cannot be stated for these compounds. Pharmacokinetic Analysis Primary parameters of the disposition of sulfasalazine and metabolites were obtained using the calculator spreadsheet of the modelling program, MKMODEL (Elsevier Biosoft, Cambridge, VK) as follows: the elimination rate constant (k) was determined by linear regression of the terminal phase of the log concentration-time curve; the area under the curve (AVC) was calculated using a combin-
ation of the linear and logarithmic trapezoidal rules (Gibaldi & Perrier 1982) and the value of AVC to affinity was estimated by addition of the residual area, given by C*/k, where C* is the final estimated concentration. Calculations using the data obtained during Days 21 to 25 were adjusted as necessary for drug remaining from previous dose(s) using the principle of superposition (Gibaldi & Perrier 1982). Parameters were calculated according to standard noncompartmental methods (Gibaldi & Perrier 1982) noted here. Elimination halflife (t.;,) was calculated as 0.693 divided by k. Renal clearance (CLR) was the amount of drug in urine divided by the AVC (to 96h after administration). Plasma clearance (CL) was the amount of sulfapyridine formed divided by the AVC to infinity, assuming that the urinary excretion of sulfapyridine, 5-hydroxy-sulfapyridine, N-acetyl-sulfapyridine and N-acetyl-5-hydroxy-sulfapyridine accounted for the total elimination of the
Clin. Pharmacokinet. 23 (4) 1992
314
sulfapyridine formed. The volume of distribution (Vd) was calculated as CL divided by k. The average concentration at steady-state (CSS) was predicted from the single dose (AUC to infinity/dosage interval of 96h) and observed on multiple doses (AUC over dosage interval of 24h). Plasma concentration vs time data for sulfapyridine were fitted by least squares estimation with reciprocal variance weighting to a I-compartment model, including a lag time and an absorption phase, using a nonlinear least squares regression program, NONLIN (Metzler 1969). Goodness of fit was estimated using the I-sample independent T-statistic associated with the correlation coefficient, with n (number of data points) - p (number of parameters) degrees offreedom. Using the experimental constants obtained, the following parameters were calculated (Gibaldi & Perrier 1982): maximum concentration in plasma (Cmax ); time taken to reach Cmax (t max ). Statistical Analysis The pharmacokinetic data obtained were assessed using nonparametric methods. The KruskalWallis I-way analysis of variance was used and multiple comparisons were made of the different subgroups of age and acetylator status using a ranking procedure (Conover 1971). Parameters of short
and long term administration were compared using the Wilcoxon signed rank test.
Results The characteristics of the 20 patients who took part in the study are given in table I. Numbers of fast and slow acetylators were equal in the young and old groups. Figure I shows semilogarithmic plots of the serum concentrations of sulfasalazine, sulfapyridine, N-acetyl-sulfapyridine, N-acetyl-sulfapyridine-O-glucuronide and N-acetyl-mesalazine versus time after administration of a single dose of the drug to the 4 subgroups of patients: (a) young/ fast; (b) young/slow; (c) old/fast; (d) old/slow. Trace amounts of mesalazine and sulfapyridine-O-glucuronide were detectable only at 12 and 36h, respectively, after administration. Sulfasalazine Pharmacokinetics In all cases, the disposition of sulfasalazine can be described by a 2-compartment model, in which the distribution equilibrium is reached 12 to 18h after administration. The amount of data describing the distribution phase was insufficient for application of the appropriate fitting procedure. The profiles of the absorption and disposition of the parent drug in the young/slow and the elderly
Table I. Mean (range) patient characteristics Characteristic
Sex Acetylator phenotype Seropositive (OAT >1 : 16) Erosions Age (years) Bodyweight (kg) Duration of rheumatoid arthritis (years) Haemoglobin (g/dl) Erythrocyte sedimentation rate (mm/h)
No. of patients young
old
2 M, 6 F 4 F, 4 S
4 M, 8 F 6 F, 6 S 6 9 74.4 (71-83) 58.3 (44-85) 13.8 (0.5-45) 11.0 (7.1-13.3) 77 (33-104)
8 4 40.5 (35-46) 69.1 (48-95) 10.5 (4-16) 12.5 (9.5-16.4) 36 (2-56)
Abbreviations: M = male; F = female; F = fast; S = slow; OAT = differential agglutination titre (Rose-Waaler test).
315
Sulfasalazine Pharmacokinetics
subgroups (fig. I), were similar in that the Cmax values were 10.35, 10.1 and 10.2 mg/L, respectively, and each was achieved 6h after dosage. In the young/fast subgroup (fig. I), however, the Cmax, 5.6 mg/L, occurred 9h after administration of sulfasalazine and concentrations of the drug were undetectable after 30h. In the young/slow subgroup, concentrations of sulfasalazine were measurable up to 36h, but this increased in the elderly subgroups to 60h after administration. The pharmacokinetic parameters calculated relating to the elimination of sulfasalazine are shown in table II. Mean t'l2 values during short and long term administration in both slow and fast subgroups were significantly increased in elderly compared with young patients. After long term administration, t'l2 values increased and the mean in the slow subgroup of elderly patients was significantly different from that obtained after the single dose. Sulfapyridine Pharmacokinetics The pharmacokinetic parameters estimated for sulfapyridine are shown in table III. The rates of
absorption of sulfapyridine and its formation from sulfasalazine in plasma appear to be sufficiently slow (fig. 1) that the concentration vs time data could be fitted in order to generate values of Cmax and tmax• Slow acetylators had consistently greater Cmax values which occurred later than those in fast acetylators, with the exception of the elderly subgroups during long term administration. The generally large differences, particularly in Cmax values, were not significant, however, because of the large interindividual variations in these estimates. Some sulfapyridine disposition parameters differed in slow compared with fast acetylators, including CL and Vd, which increased after the single dose, and t'l2 which increased only in elderly patients. Of greatest therapeutic significance were the increases in serum concentrations at steady-state in slow by comparison with fast acetylators within both the young and elderly groups during both short and long term administration, which were all statistically different. During long term administration, the mean serum concentration at steady-state increased considerably in the slow elderly subgroup and in both slow and fast elderly subgroups there was a reduction in CL. There was no significant
Table II. Mean (± SO) pharmacokinetic parameters of sulfasalazine Parameter
tv.
p-Valueb
Old
Young fasta
slow"
fast
dosage comparison
slow
dosage comparison
36 ± 1.0t 6.5 ± 2.7
4.9 ± 0.6* 5.8 ± 0.8
9.8 ± 4.2t 13.7 ± 7.5
NS
8.8 ± 2.8* 20.3 ± -11.5
p
1.5 ± 1.4 4.2 ± 4.6
4.7 ± 5.0 4.4 ± 2.3
6.5 ± 5.0 4.6 ± 3.8
NS
5.7 ± 3.5 6.7 ± 5.2
NS
NS NS
0.18 ± 0.21 0.18 ± 0.07
0.051 ± 0.039 0.038 ± 0.009 NS 0.054 ± 0.038 0.13 ± 0.18
0.046 ± 0.040 NS 0.044 ± 0.029
NS NS
(h)
single dose multiple dose C·· (mg/L) single dose multiple dose CLR (ml/min/kg) single dose multiple dose a b
< 0.D1
0.0005 0.011
Only 4 subjects in each subgroup, so no statistical comparison made between single- and multiple-dose studies. Overall significance of Kruskal-Wallis test. Abbreviations and symbols: tv. = elimination half-life; C·s = average plasma drug concentration at steady-state; CLR = renal clearance, which can be converted to L/h/kg by multiplying by 0.06; NS = not significant; t = significant difference between old/fast and young/ fast subgroups; significant difference between old/slow and young/slow subgroups.
*=
Clin. Pharmacokinet. 23 (4) 1992
316
Table III. Mean values (±
SO)
Parameter
Young
tv. (h) single dose multiple dose C·· (mg/L) single dose multiple dose Vd (L/kg) single dose multiple dose CLR (ml/min/kg) single dose multiple dose CL (ml/min/kg) single dose multiple dose Cmsx (mg/L) single dose multiple dose t msx (h) single dose multiple dose
of pharmacokinetic parameters of sulfapyridine p-Value b
Old
fastS
slows
fast
dosage comparison
slow
dosage comparison
9.5 ± 5.9 10.4 ± 5.2
12.5 ± 5.2 14.8 ± 3.7
12.1 ± 8.4 11.9 ± 5.4"
NS
16.7 ± 6.5 27.2 ± 12.5"
NS
9.7 ± 9.7"" 10.8 ± 7.3""
16.6 ± 4.0"' 26.2 ± 9.7""
8.7 ± 2.9" 10.7 ± 4.1"
NS
26.3 ± 12.2" 35.2 ± 26.8"
p
0.57 ± 0.56" 0.49 ± 0.54
0.19 ± 0.16" 0.11 ± 0.07
0.87 ± 0.36 0.70 ± 0.81
NS
0.42 ± 0.18 0.43 ± 0.46
NS
0.083 ± 0.051 0.074 ± 0.056
NS
0.35 ± 0.18' 0.17 ± 0.13
p
NS 0.034
< 0.01
0.012 0.Q18 0.041
NS
0.Q75 ± 0.084 0.042 ± 0.037 0.090 ± 0.046 0.080 ± 0.033 0.052 ± 0.017 0.091 ± 0.078
NS
1.4 ± 1.2"" 0.70 ± 0.78
0.20 ± 0.18'" 0.11 ± 0.08
1.0 ± 0.8" 0.47 ± 0.27
P
13.3 ± 6.3 13.6 ± 9.6
17.5 ± 5.7 30.1 ± 11.0
10.5 ± 4.2 12.3 ± 6.4
NS
15.3 ± 3.4 33.6 ± 34.8
NS
NS NS
11.1 ± 2.8 13.8 ± 5.1
16.2 ± 5.0 16.8 ± 4.0
13.4 ± 3.8 16.5 ± 7.5
NS
15.1 ± 5.7 14.9 ± 6.7
NS
NS NS
< 0.01
< 0.01
NS NS 0.047
NS
a b
Only 4 subjects in each subgroup, so no statistical comparison made between single- and multiple-dose studies. Overall significance of Kruskal-Wallis test. Abbreviations: Vd = volume of distribution; CL = plasma clearance; Cmax = peak plasma drug concentration; t max = time to Cmax ; for other abbreviations, see table II .• = significant difference between old/fast and old/slow subgroups; •• = significant difference between young/fast and young/slow subgroups;
effect of age on any of the parameters of the disposition of sulfapyridine.
ations in this parameter. CLR values increased markedly in all subgroups after long term administration.
N-Acetyl-Mesalazine Pharmacokinetics Table IV shows the pharmacokinetic parameters calculated for N-acetyl-mesalazine and limited data for mesalazine after long term administration. Mean values of steady-state concentrations of Nacetyl-mesalazine were about 3-fold greater in elderly patients than in the young group and were significantly different in the fast acetylators. In all subgroups, steady-state concentrations decreased after long term administration by a factor of 2 to 3, but these changes were not reflected in consistent alterations in t'l2 values, because of large vari-
Urinary Excretion of Sulfasalazine and Metabolites Figure 2 shows the relative contributions of each of the urinary products to the total excretion of the drug, which was 60.0 ± 28.4, 41.2 ± 38.8, 67.2 ± 28.8 and 76.4 ± 32.5 in the young/fast, young/ slow, old/fast and old/slow subgroups, respectively, as shown in figure 2. Renal elimination of sulfapyridine and its metabolites was affected markedly by acetylator status. Mean excretion (as a proportion of the total dose) of sulfapyridine was
Sulfasalazine Pharmacokinetics
317
Table IV. Mean values (± SO) of pharmacokinetic parameters of N-acetyl-mesalazine Parameters
tv:! (h) single dose multiple dose CSs (mg/L) single dose multiple dose 5-ME multiple dose CLR (ml/min/kg) single dose multiple dose
Young
p-Value b
Old
fast 8
slow8
fast
38 ± 25 26 ± 12
46 ± 28 25 ± 12
42 ± 23 47 ± 21
1.2 ± 0.3 0.41 ± 0.31t
2.4 ± 0.7 1.2 ± 0.6t
0.027 ± 0.031
1.7 ± 0.6 0.63 ± 0.23 0.13 ± 0.11
0.16 ± 0.07
0.99 ± 0.85 4.2 ± 3.7
0.99 ± 0.84 2.6 ± 2.1
0.85 ± 0.41 2.5 ± 1.8
dosage comparison
p
< 0.01
slow
dosage comparison
23 ± 16 33 ± 22
NS
2.8 ± 2.4 0.95 ± 0.33
P < 0.01
NS
0.14 ± 0.11 p
< 0.01
1.0 ± 0.5 2.8 ± 1.4
NS 0.039
p
< 0.01
NS NS
a Only 4 subjects in each group. so no statistical comparison made between single- and multiple-dose studies. b Overall significance of Kruskal-Wallis test. Abbreviations: see table II.
greater (p = 0.07) in slow (5.1 ± 4.6 and 13.6 ± 7.0% in the young and elderly groups, respectively) than in fast acetylators (4.1 ± 3.2 and 4.8 ± 1.9% in the young and elderly groups, respectively) and correspondingly, the extent of excretion of the acetylated metabolites (fig. 2) was 12.0 ± 8.5, 32.8 ± 12.3,38.9 ± 21.7 and 42.7 ± 20.0% in these 4 groups, respectively (p = 0.06). These differences did not reach significance, but the amounts of acetylated metabolites, expressed as proportions of the total sulfapyridine excreted (fig. 2), were substantially larger (p = 0.008) in the fast (93.3 ± 21.1 and 77.5 ± 11.0% in the young and elderly groups, respectively) than in the slow acetylators (57.1 ± 14.0 and 60.8 ± 6.8% in the young and elderly groups, respectively). A significant difference was observed between the mean values of the 2 subgroups of the young group. Other comparisons, viz. percentage of total dose excreted and percentage of dose excreted as total sulfapyridine, total mesalazine or as the individual metabolites, were not significant.
Discussion Effect of Age It is acknowledged that the susceptibility of the elderly to drugs may be due partly to altered phar-
macokinetics (Dowling & Crowe 1989). In the present study, age influenced parameters of the disposition of the parent drug, sulfasalazine, and its major salicylate metabolite, N-acetyl-mesalazine, but the elimination of the sulfapyridine moiety was not affected. Following a single dose of sulfasalazine, mean tl/2 values in the young group were similar to the 3 to 5h figures previously reported in healthy volunteers (Ryde & Lima 1981), but increased by approximately 2-fold in the elderly with rheumatoid arthritis. A median value of 10.2h has been observed in patients with ulcerative colitis (Azad Khan et al. 1982). A mean t'l2 of 8h was found after repeated doses of 4g administered to healthy volunteers (Schroder et al. 1973); this value was greater than that observed in our study, yet was found in volunteers receiving twice the dosage. Long term sulfasalazine treatment in the elderly resulted in considerable prolongation in t'l2 to 20h in the slow acetylator subgroup, compared with 6h in the younger group and with the value of 9h obtained after the single dose. It is difficult to assess in pharmacokinetic terms the significance of this increased t'l2 of sulfasalazine. Impairment of CLR, often of greatest consequence for drug handling in the elderly (Dowling & Crowe 1989), was not apparent in this study and, in any case, excretion by
C/in. Pharmacokinet. 23 (4) 1992
318
40 35
~
30
"0
~
Q)
"
25
)(
CD
CD
'"
0 "0
20
'0 c:
.g
15
u
u:'" 10 5
0
~
~ SASP
[J Fast/young
• Slow/young
AcSP
Total ME
~ SPG
AcSPG
~ SP
o
Fast/elderly • Slow/elderly
* 100
100
;?
~ 80
80
.S!
1II
l':'
u )(
'"en
60
OJ
0 "0
~
'0 40 c:
.9 U
u:'"
20 0
~
"0 CD
"0
~
~
~
60
u )( CD
a..
(f)
(ij
40
£ '0 c
0
20 "
Clio. Pharmacokioet. 23 (4): 311-320, 1992 0312-5963/92/00 I 0-0311 / $05.00/0 © Adis International Limited. All rights rese rved . CPK1223
The Effect of Age and Acetylator Phenotype on the Pharmacokinetics of Sulfasalazine in Patients with Rheumatoid Arthritis Allister J. Taggart. Barbara J. McDermott and Stanley D. Roberts Department of Therapeutics and Pharmacology, The Queen's University of Belfast, and Department of Rheumatology, Musgrave Park Hospital, Belfast, Northern Ireland
Summary
The pharmacokinetic disposition of sulfasalazine and its metabolites was studied in 8 young and 12 elderly patients with active rheumatoid arthritis. Equal numbers of slow and fast acetylators were included in each age group. Patients received enteric-coated sulfasalazine 2g daily for 21 days; specimens of serum and urine were collected for 96h after administration on days I and 21. The elimination half-life of sulfasalazine was greater in the elderly patients. Many disposition parameters of sulfapyridine differed in slow and fast acetylators; of greatest significance were the increased values of steady-state serum concentration in the slow acetylators. There was no effect of age on any sulfapyridine disposition parameters. Values for the steady-state serum concentrations of N-acetyl-5-acetylsalicylic acid were greater in elderly than in young patients. The metabolism of sulfapyridine was markedly affected by acetylator phenotype and this was reflected in the composition of sulfapyridine-related material in the urine. Thus, age is a determinant of the steady-state concentrations of salicylate moieties but acetylator phenotype plays a greater role in determining the serum concentration of sulfapyridine, which has greater therapeutic implications in rheumatology.
Sulfasalazine was first used in Sweden over 40 years ago for the treatment of 'rheumatic polyarthritis' and ulcerative colitis (Svartz 1942). An early noncomparative study produced unfavourable results in rheumatoid disease (Sinclair & Duthie 1948) but by the 1960s, the drug was widely accepted as effective therapy for inflammatory bowel disease (Misiewicz et al. 1965). It is only in the past 10 years that the drug has been firmly established as an effective second-line agent in rheumatoid arthritis (Neumann et al. 1983; Pullar et al. 1983). After oral administration, sulfasalazine is split by bacterial azo-reduction in the colon into sulfapyridine and mesalazine (5-aminosalicylic acid). The absorption and disposition of these metabo-
lites occur by complicated pathways: the systemic uptake of sulfapyridine is virtually complete and it is eliminated by polymorphic acetylation, hydroxylation and glucuronidation; mesalazine is subject to both capacity-limited presystemic and systemic acetylation and enterohepatic recirculation occurs (Klotz 1985). There is, therefore, considerable scope for interindividual variation in handling of the drug. Pharmacokinetic studies have been performed in healthy volunteers (Schroder & Campbell 1972) and in patients with inflammatory bowel disease (Azad Khan et al. 1982; Das et al. 1973), but there have been no comparable studies in those with rheumatoid arthritis. In the treatment of the elderly population with rheumatoid arthritis, investigation
312
of the disposition of sulfasalazine in relation to age is important; also of relevance is the influence of acetylator status and the pharmacokinetic profile of sulfapyridine, which may be the active moiety of the compound (Neumann et al. 1986; Pullar et al. 1985a).
Patients and Methods Patients The study, which was approved by the University Ethical Committee, involved patients diagnosed as having classical or definite rheumatoid arthritis according to the American Rheumatism Association diagnosis. Active disease was defined by the presence of an erythrocyte sedimentation rate >28 mm/h or C-reactive protein concentration >5 mg/L or plasma viscosity> 1.72 centipoise, and at least 2 of the following conditions: tenderness of >6 joints, swelling of> 3 joints, morning stiffness of >45 min duration, and Ritchie articular index > 20. Patients were excluded from the study if any of the following criteria held: significant disease or previous surgery of the bowel; treatment with a second-line antirheumatic agent during the 3 months before the study; current therapy with inhibitors or inducers of hepatic enzymes (e.g. corticosteroids, coumarins, oral contraceptives, cimetidine, phenytoin, barbiturates); current therapy with agents which could have a pharmacokinetic interaction with su1fasa1azine (Das & Dubin 1976) such as antibiotics, calcium, iron, oral hypoglycaemic drugs and those which affect gut motility; smoking> 10 cigarettes or drinking >3 pints of beer or its alcoholic equivalent daily; significant cardiovascular, renal or hepatic disease. Drug Administration and Sampling Schedule The patients fasted from 2200h on the night before the start of the study until 2h after administration of the drug on day 1. On this morning, an indwelling cannula was inserted into a forearm vein and the patient received a single 2g oral dose of enteric-coated sulfasalazine. A 10ml specimen of blood was drawn at each of the following times
Clin. Pharmacokinet. 23 (4) 1992
after administration: 0, 1, 3, 6, 9, 12, 24, 30, 36, 48, 60, 72 and 96h. Serum was separated and stored at -20°C. Urine excreted during 4 consecutive periods of 24h after administration was collected, each volume was recorded and a 20ml aliquot of each specimen was retained for analysis of the drug and metabolites. Creatinine clearance over the 24h period of day 1 was determined. Treatment with sulfasalazine 2g daily was continued from days 5 to 21. Patients were reviewed on days 8 and 15 to monitor compliance and adverse effects. The sampling schedule was repeated on days 21 to 25 and creatinine clearance was determined again. Patients were withdrawn from the study if any of the following occurred: withdrawal of consent; noncompliance; significant adverse effects including vomiting on days 1 or 21; urinary incontinence during the sampling periods. Analytical Methods Acetylator phenotype was assessed by the method of Evans (1969). High performance liquid chromatography (HPLC) analysis of sulfasalazine, sulfapyridine and its metabolites, N-acetyl-sulfapyridine and N-acetyl-5-hydroxysulfapyridine, was performed, and sulfapyridine-O-glucuronide and Nacetyl-sulfapyridine-O-glucuronide were codetermined with aglycones by the method of Ahnfelt et al. (1986). Only minor amounts are present in the ag1ycones and so, in calculating the concentrations of N-acetyl-sulfapyridine-O-glucuronide and sulfapyridine-O-glucuronide, it was assumed that only glucuronides were detected. Mesalazine and JII"acetyl-mesalazine were estimated according to Willoughby et al. (1982). The following gives the precision of each assay, with the concentration of the analyte (in mg/L) given in parentheses: 2.9% (1.1), 3.1 % (2.0), 3.4% (2.0), 7.2% (2.0), 13.5% (0.06), 2.1 % (0.07) in serum and 6.6% (1.1), 3.9% (5.0), 3.8% (5.0), 5.4% (5.0), 2.0% (0.9), 2.1 % (2.0) in urine for sulfasalazine, sulfapyridine, N-acetyl-sulfapyridine, N-acetyl-5-hydroxy-sulfapyridine, mesalazine and N-acetyl-mesalazine, respectively. Linear calibration curves were obtained down to 0.5 mg/L for sulfasalazine both in serum and in urine. For the
313
Sulfasalazine Pharmacokinetics
50.0
50.0
Young/fast
Young/slow
c:
~co
Q) ()
c: 0
()
E
2Q)
rJJ
12
24
36
48
50.0
60
72
84
96
12
24
36
50.0
Old/fast
48
60
72
84
96
Old/slow
c:
0
~
E 2l c: 8 E
~
rJJ
0.1 0.0 0
0.1 12
24
36
48 60 Time (h)
72
84
96
0.05 0
12
24
36
48 60 Time (h)
72
84
96
Fig. 1. Serum concentration (± SO) of sulfasalazine (SASP, e) and metabolites, sulfapyridine (SP, 6), N-acetyl-sulfapyridine (AcSP,O), N-acetyl-sulfapyridine-O-glucuronide (AcSPG, X) and N-acetyl-mesalazine (Ac-ME, 0) after oral administration of sulfasalazine 2g on day I of the treatment schedule in the 4 subgroups of patients: youngJfast; young/slow; old/fast; old/ slow.
sulfapyridine compounds, linear calibration curves were obtained down to I mg/L in serum and 2 mg/ L in urine. For mesalazine and N-acetyl-mesalazine, the calibration curves were linear down to 0.06 and 0.07 mg/L in serum, whereas they were linear down to 0.9 and 2.0 mg/L, respectively, in urine. Since pure glucuronides were not available, precision and sensitivity data cannot be stated for these compounds. Pharmacokinetic Analysis Primary parameters of the disposition of sulfasalazine and metabolites were obtained using the calculator spreadsheet of the modelling program, MKMODEL (Elsevier Biosoft, Cambridge, VK) as follows: the elimination rate constant (k) was determined by linear regression of the terminal phase of the log concentration-time curve; the area under the curve (AVC) was calculated using a combin-
ation of the linear and logarithmic trapezoidal rules (Gibaldi & Perrier 1982) and the value of AVC to affinity was estimated by addition of the residual area, given by C*/k, where C* is the final estimated concentration. Calculations using the data obtained during Days 21 to 25 were adjusted as necessary for drug remaining from previous dose(s) using the principle of superposition (Gibaldi & Perrier 1982). Parameters were calculated according to standard noncompartmental methods (Gibaldi & Perrier 1982) noted here. Elimination halflife (t.;,) was calculated as 0.693 divided by k. Renal clearance (CLR) was the amount of drug in urine divided by the AVC (to 96h after administration). Plasma clearance (CL) was the amount of sulfapyridine formed divided by the AVC to infinity, assuming that the urinary excretion of sulfapyridine, 5-hydroxy-sulfapyridine, N-acetyl-sulfapyridine and N-acetyl-5-hydroxy-sulfapyridine accounted for the total elimination of the
Clin. Pharmacokinet. 23 (4) 1992
314
sulfapyridine formed. The volume of distribution (Vd) was calculated as CL divided by k. The average concentration at steady-state (CSS) was predicted from the single dose (AUC to infinity/dosage interval of 96h) and observed on multiple doses (AUC over dosage interval of 24h). Plasma concentration vs time data for sulfapyridine were fitted by least squares estimation with reciprocal variance weighting to a I-compartment model, including a lag time and an absorption phase, using a nonlinear least squares regression program, NONLIN (Metzler 1969). Goodness of fit was estimated using the I-sample independent T-statistic associated with the correlation coefficient, with n (number of data points) - p (number of parameters) degrees offreedom. Using the experimental constants obtained, the following parameters were calculated (Gibaldi & Perrier 1982): maximum concentration in plasma (Cmax ); time taken to reach Cmax (t max ). Statistical Analysis The pharmacokinetic data obtained were assessed using nonparametric methods. The KruskalWallis I-way analysis of variance was used and multiple comparisons were made of the different subgroups of age and acetylator status using a ranking procedure (Conover 1971). Parameters of short
and long term administration were compared using the Wilcoxon signed rank test.
Results The characteristics of the 20 patients who took part in the study are given in table I. Numbers of fast and slow acetylators were equal in the young and old groups. Figure I shows semilogarithmic plots of the serum concentrations of sulfasalazine, sulfapyridine, N-acetyl-sulfapyridine, N-acetyl-sulfapyridine-O-glucuronide and N-acetyl-mesalazine versus time after administration of a single dose of the drug to the 4 subgroups of patients: (a) young/ fast; (b) young/slow; (c) old/fast; (d) old/slow. Trace amounts of mesalazine and sulfapyridine-O-glucuronide were detectable only at 12 and 36h, respectively, after administration. Sulfasalazine Pharmacokinetics In all cases, the disposition of sulfasalazine can be described by a 2-compartment model, in which the distribution equilibrium is reached 12 to 18h after administration. The amount of data describing the distribution phase was insufficient for application of the appropriate fitting procedure. The profiles of the absorption and disposition of the parent drug in the young/slow and the elderly
Table I. Mean (range) patient characteristics Characteristic
Sex Acetylator phenotype Seropositive (OAT >1 : 16) Erosions Age (years) Bodyweight (kg) Duration of rheumatoid arthritis (years) Haemoglobin (g/dl) Erythrocyte sedimentation rate (mm/h)
No. of patients young
old
2 M, 6 F 4 F, 4 S
4 M, 8 F 6 F, 6 S 6 9 74.4 (71-83) 58.3 (44-85) 13.8 (0.5-45) 11.0 (7.1-13.3) 77 (33-104)
8 4 40.5 (35-46) 69.1 (48-95) 10.5 (4-16) 12.5 (9.5-16.4) 36 (2-56)
Abbreviations: M = male; F = female; F = fast; S = slow; OAT = differential agglutination titre (Rose-Waaler test).
315
Sulfasalazine Pharmacokinetics
subgroups (fig. I), were similar in that the Cmax values were 10.35, 10.1 and 10.2 mg/L, respectively, and each was achieved 6h after dosage. In the young/fast subgroup (fig. I), however, the Cmax, 5.6 mg/L, occurred 9h after administration of sulfasalazine and concentrations of the drug were undetectable after 30h. In the young/slow subgroup, concentrations of sulfasalazine were measurable up to 36h, but this increased in the elderly subgroups to 60h after administration. The pharmacokinetic parameters calculated relating to the elimination of sulfasalazine are shown in table II. Mean t'l2 values during short and long term administration in both slow and fast subgroups were significantly increased in elderly compared with young patients. After long term administration, t'l2 values increased and the mean in the slow subgroup of elderly patients was significantly different from that obtained after the single dose. Sulfapyridine Pharmacokinetics The pharmacokinetic parameters estimated for sulfapyridine are shown in table III. The rates of
absorption of sulfapyridine and its formation from sulfasalazine in plasma appear to be sufficiently slow (fig. 1) that the concentration vs time data could be fitted in order to generate values of Cmax and tmax• Slow acetylators had consistently greater Cmax values which occurred later than those in fast acetylators, with the exception of the elderly subgroups during long term administration. The generally large differences, particularly in Cmax values, were not significant, however, because of the large interindividual variations in these estimates. Some sulfapyridine disposition parameters differed in slow compared with fast acetylators, including CL and Vd, which increased after the single dose, and t'l2 which increased only in elderly patients. Of greatest therapeutic significance were the increases in serum concentrations at steady-state in slow by comparison with fast acetylators within both the young and elderly groups during both short and long term administration, which were all statistically different. During long term administration, the mean serum concentration at steady-state increased considerably in the slow elderly subgroup and in both slow and fast elderly subgroups there was a reduction in CL. There was no significant
Table II. Mean (± SO) pharmacokinetic parameters of sulfasalazine Parameter
tv.
p-Valueb
Old
Young fasta
slow"
fast
dosage comparison
slow
dosage comparison
36 ± 1.0t 6.5 ± 2.7
4.9 ± 0.6* 5.8 ± 0.8
9.8 ± 4.2t 13.7 ± 7.5
NS
8.8 ± 2.8* 20.3 ± -11.5
p
1.5 ± 1.4 4.2 ± 4.6
4.7 ± 5.0 4.4 ± 2.3
6.5 ± 5.0 4.6 ± 3.8
NS
5.7 ± 3.5 6.7 ± 5.2
NS
NS NS
0.18 ± 0.21 0.18 ± 0.07
0.051 ± 0.039 0.038 ± 0.009 NS 0.054 ± 0.038 0.13 ± 0.18
0.046 ± 0.040 NS 0.044 ± 0.029
NS NS
(h)
single dose multiple dose C·· (mg/L) single dose multiple dose CLR (ml/min/kg) single dose multiple dose a b
< 0.D1
0.0005 0.011
Only 4 subjects in each subgroup, so no statistical comparison made between single- and multiple-dose studies. Overall significance of Kruskal-Wallis test. Abbreviations and symbols: tv. = elimination half-life; C·s = average plasma drug concentration at steady-state; CLR = renal clearance, which can be converted to L/h/kg by multiplying by 0.06; NS = not significant; t = significant difference between old/fast and young/ fast subgroups; significant difference between old/slow and young/slow subgroups.
*=
Clin. Pharmacokinet. 23 (4) 1992
316
Table III. Mean values (±
SO)
Parameter
Young
tv. (h) single dose multiple dose C·· (mg/L) single dose multiple dose Vd (L/kg) single dose multiple dose CLR (ml/min/kg) single dose multiple dose CL (ml/min/kg) single dose multiple dose Cmsx (mg/L) single dose multiple dose t msx (h) single dose multiple dose
of pharmacokinetic parameters of sulfapyridine p-Value b
Old
fastS
slows
fast
dosage comparison
slow
dosage comparison
9.5 ± 5.9 10.4 ± 5.2
12.5 ± 5.2 14.8 ± 3.7
12.1 ± 8.4 11.9 ± 5.4"
NS
16.7 ± 6.5 27.2 ± 12.5"
NS
9.7 ± 9.7"" 10.8 ± 7.3""
16.6 ± 4.0"' 26.2 ± 9.7""
8.7 ± 2.9" 10.7 ± 4.1"
NS
26.3 ± 12.2" 35.2 ± 26.8"
p
0.57 ± 0.56" 0.49 ± 0.54
0.19 ± 0.16" 0.11 ± 0.07
0.87 ± 0.36 0.70 ± 0.81
NS
0.42 ± 0.18 0.43 ± 0.46
NS
0.083 ± 0.051 0.074 ± 0.056
NS
0.35 ± 0.18' 0.17 ± 0.13
p
NS 0.034
< 0.01
0.012 0.Q18 0.041
NS
0.Q75 ± 0.084 0.042 ± 0.037 0.090 ± 0.046 0.080 ± 0.033 0.052 ± 0.017 0.091 ± 0.078
NS
1.4 ± 1.2"" 0.70 ± 0.78
0.20 ± 0.18'" 0.11 ± 0.08
1.0 ± 0.8" 0.47 ± 0.27
P
13.3 ± 6.3 13.6 ± 9.6
17.5 ± 5.7 30.1 ± 11.0
10.5 ± 4.2 12.3 ± 6.4
NS
15.3 ± 3.4 33.6 ± 34.8
NS
NS NS
11.1 ± 2.8 13.8 ± 5.1
16.2 ± 5.0 16.8 ± 4.0
13.4 ± 3.8 16.5 ± 7.5
NS
15.1 ± 5.7 14.9 ± 6.7
NS
NS NS
< 0.01
< 0.01
NS NS 0.047
NS
a b
Only 4 subjects in each subgroup, so no statistical comparison made between single- and multiple-dose studies. Overall significance of Kruskal-Wallis test. Abbreviations: Vd = volume of distribution; CL = plasma clearance; Cmax = peak plasma drug concentration; t max = time to Cmax ; for other abbreviations, see table II .• = significant difference between old/fast and old/slow subgroups; •• = significant difference between young/fast and young/slow subgroups;
effect of age on any of the parameters of the disposition of sulfapyridine.
ations in this parameter. CLR values increased markedly in all subgroups after long term administration.
N-Acetyl-Mesalazine Pharmacokinetics Table IV shows the pharmacokinetic parameters calculated for N-acetyl-mesalazine and limited data for mesalazine after long term administration. Mean values of steady-state concentrations of Nacetyl-mesalazine were about 3-fold greater in elderly patients than in the young group and were significantly different in the fast acetylators. In all subgroups, steady-state concentrations decreased after long term administration by a factor of 2 to 3, but these changes were not reflected in consistent alterations in t'l2 values, because of large vari-
Urinary Excretion of Sulfasalazine and Metabolites Figure 2 shows the relative contributions of each of the urinary products to the total excretion of the drug, which was 60.0 ± 28.4, 41.2 ± 38.8, 67.2 ± 28.8 and 76.4 ± 32.5 in the young/fast, young/ slow, old/fast and old/slow subgroups, respectively, as shown in figure 2. Renal elimination of sulfapyridine and its metabolites was affected markedly by acetylator status. Mean excretion (as a proportion of the total dose) of sulfapyridine was
Sulfasalazine Pharmacokinetics
317
Table IV. Mean values (± SO) of pharmacokinetic parameters of N-acetyl-mesalazine Parameters
tv:! (h) single dose multiple dose CSs (mg/L) single dose multiple dose 5-ME multiple dose CLR (ml/min/kg) single dose multiple dose
Young
p-Value b
Old
fast 8
slow8
fast
38 ± 25 26 ± 12
46 ± 28 25 ± 12
42 ± 23 47 ± 21
1.2 ± 0.3 0.41 ± 0.31t
2.4 ± 0.7 1.2 ± 0.6t
0.027 ± 0.031
1.7 ± 0.6 0.63 ± 0.23 0.13 ± 0.11
0.16 ± 0.07
0.99 ± 0.85 4.2 ± 3.7
0.99 ± 0.84 2.6 ± 2.1
0.85 ± 0.41 2.5 ± 1.8
dosage comparison
p
< 0.01
slow
dosage comparison
23 ± 16 33 ± 22
NS
2.8 ± 2.4 0.95 ± 0.33
P < 0.01
NS
0.14 ± 0.11 p
< 0.01
1.0 ± 0.5 2.8 ± 1.4
NS 0.039
p
< 0.01
NS NS
a Only 4 subjects in each group. so no statistical comparison made between single- and multiple-dose studies. b Overall significance of Kruskal-Wallis test. Abbreviations: see table II.
greater (p = 0.07) in slow (5.1 ± 4.6 and 13.6 ± 7.0% in the young and elderly groups, respectively) than in fast acetylators (4.1 ± 3.2 and 4.8 ± 1.9% in the young and elderly groups, respectively) and correspondingly, the extent of excretion of the acetylated metabolites (fig. 2) was 12.0 ± 8.5, 32.8 ± 12.3,38.9 ± 21.7 and 42.7 ± 20.0% in these 4 groups, respectively (p = 0.06). These differences did not reach significance, but the amounts of acetylated metabolites, expressed as proportions of the total sulfapyridine excreted (fig. 2), were substantially larger (p = 0.008) in the fast (93.3 ± 21.1 and 77.5 ± 11.0% in the young and elderly groups, respectively) than in the slow acetylators (57.1 ± 14.0 and 60.8 ± 6.8% in the young and elderly groups, respectively). A significant difference was observed between the mean values of the 2 subgroups of the young group. Other comparisons, viz. percentage of total dose excreted and percentage of dose excreted as total sulfapyridine, total mesalazine or as the individual metabolites, were not significant.
Discussion Effect of Age It is acknowledged that the susceptibility of the elderly to drugs may be due partly to altered phar-
macokinetics (Dowling & Crowe 1989). In the present study, age influenced parameters of the disposition of the parent drug, sulfasalazine, and its major salicylate metabolite, N-acetyl-mesalazine, but the elimination of the sulfapyridine moiety was not affected. Following a single dose of sulfasalazine, mean tl/2 values in the young group were similar to the 3 to 5h figures previously reported in healthy volunteers (Ryde & Lima 1981), but increased by approximately 2-fold in the elderly with rheumatoid arthritis. A median value of 10.2h has been observed in patients with ulcerative colitis (Azad Khan et al. 1982). A mean t'l2 of 8h was found after repeated doses of 4g administered to healthy volunteers (Schroder et al. 1973); this value was greater than that observed in our study, yet was found in volunteers receiving twice the dosage. Long term sulfasalazine treatment in the elderly resulted in considerable prolongation in t'l2 to 20h in the slow acetylator subgroup, compared with 6h in the younger group and with the value of 9h obtained after the single dose. It is difficult to assess in pharmacokinetic terms the significance of this increased t'l2 of sulfasalazine. Impairment of CLR, often of greatest consequence for drug handling in the elderly (Dowling & Crowe 1989), was not apparent in this study and, in any case, excretion by
C/in. Pharmacokinet. 23 (4) 1992
318
40 35
~
30
"0
~
Q)
"
25
)(
CD
CD
'"
0 "0
20
'0 c:
.g
15
u
u:'" 10 5
0
~
~ SASP
[J Fast/young
• Slow/young
AcSP
Total ME
~ SPG
AcSPG
~ SP
o
Fast/elderly • Slow/elderly
* 100
100
;?
~ 80
80
.S!
1II
l':'
u )(
'"en
60
OJ
0 "0
~
'0 40 c:
.9 U
u:'"
20 0
~
"0 CD
"0
~
~
~
60
u )( CD
a..
(f)
(ij
40
£ '0 c
0
20 "
