Int. J. Immunopharmac., Vol. 8, No. 1, pp. I - 11, 1986. 0192-0561/86 $3.00+ .

00
Printed in Great Britain. © 1986 International Society for lmmunopharmacology.

S E R U M A Z A T H I O P R I N E A N D 6 - M E R C A P T O P U R I N E LEVELS A N D
I M M U N O S U P P R E S S I V E A C T I V I T Y A F T E R A Z A T H I O P R I N E IN
UREMIC PATIENTS

BO ODLIND*, PER HARTVIGal", BJORN L1NDSTROM§, GUDMAR LONNERHOLM§, GUNNAR TUFVESON~ a n d

NILS GREFBERG*

*Department of Internal Medicine, iHospital Pharmacy and ~Department of Urology, University Hospital, S-75185
Uppsala; §Department of Drugs, National Board of Health and Welfare, S-75125 Uppsala, Sweden.

(Received 11 July 1984 and in final form 1 February 1985)

Abstract - - The pharmacokinetics of azathioprine (AZA) and 6-mercaptopurine (6-MP) was studied in
uremic patients after 100 mg AZA intravenously (fifteen patients) and orally (eight patients). 6-MP was
analysed with gas chromatography mass spectrometry following extractive alkylation. AZA was determined
indirectly assuming quantitative conversion to 6-MP in whole blood. The plasma concentration of AZA fell
rapidly after i.v. administration. The mean half-time of elimination for the first rapid phase (t,/,o) was
6.1 min (S.D. + 4.1) and for the terminal phase (t,/,~) 50 min (_+ 31). The total plasma clearance (C1) was
6.9 l./min (+ 3.0). AZA was rapidly converted to 6-MP in vivo, and maximal plasma concentrations of
6-MP were found as early as 5 rain after i.v. injection of AZA. The mean t~,,~was 4.6 min (_+ 2.2), t,/,~74 min
(_+ 58) and CI 8.0 l./min (_+ 5.8). The plasma levels of both AZA and 6-MP were either low or undetectable
4 - 6 h after dose. In erythrocytes AZA levels were low or undetectable indicating rapid conversion to 6-MP
in these cells. 6-MP concentration - time curve in erythrocytes was similar to that in plasma, except for a
somewhat slower terminal phase of elimination. Oral administration of AZA generated flat plasma curves
for AZA and 6-MP. The area under the concentration-time curve (AUC) was considerably smaller than
after i.v. administration, 18 and 4107o for AZA and 6-MP, respectively. There seems to be little danger of
accumulation of AZA/6-MP in uremia. We also studied inhibition of Leucoagglutin (LA) stimulated
lymphocyte proliferation by patient plasma at different times in six of the patients following AZA i.v. Sera
drawn at 5, 10 and 30 min significantly inhibited the LA-induced proliferation, with an estimated minimum
effective concentration of 6-MP in the cultures of about 0.02-0.04/aM. This suggests the possibility of a
therapeutic effect even of the low plasma levels of 6-MP obtained after AZA orally. The combined use of
sensitive pharmacokinetic and immunological assays as described should be useful in studying the
relationship between plasma levels of AZA/6-MP and their immunosuppressive effect and toxicity.

For m o r e t h a n 20 yr, a z a t h i o p r i n e ( A Z A ) , the reflecting a n increasing interest in the field. A Z A was

l-methyl-4-nitro-5-imidazolyl derivative of
6-mercaptopurine (6-MP) has been a m a i n
i m m u n o s u p p r e s s i v e agent in renal t r a n s p l a n t a t i o n
a n d in a u t o i m m u n e a n d o t h e r diseases. 6 - M P is also
widely used in the c h e m o t h e r a p y of acute
l y m p h o b l a s t i c leukemia. Due to lack o f sensitive a n d
specific analytical m e t h o d s , little i n f o r m a t i o n has
been available c o n c e r n i n g the p h a r m a c o k i n e t i c s o f
these substances (cf. Bach, 1975). Recently,
however, several such m e t h o d s h a v e been developed
( M a d d o c k s , 1979a; Ding & Benet, 1979a; Lin,
Jessup, Floyd, W a n g , v a n Buren, Caprioli & K a h a n ,
1980; Floberg, Hartvig, L i n d s t r 6 m , L 6 n n e r h o l m &
Odlind, 1981; N a r a n g , Yeager & Chatterji, 1982)
originally synthesized as a " s l o w release" f o r m o f
6 - M P , in a hope t h a t the side g r o u p could protect the
thiopurine group of 6-MP from rapid methylation.
A Z A is however, rapidly c o n v e r t e d to 6 - M P , which
is t h o u g h t to be responsible for m o s t o f the in vivo
i m m u n o s u p p r e s s i v e effects o f A Z A ( B e r e n b a u m ,
1971; Bach, 1975), via intraceUular anabolic
t r a n s f o r m a t i o n to thioinosinic acid a n d other
t h i o a n a l o g s o f purine derivatives (Elion & Hitchings,
1975). Ribonucleotides do n o t transverse cell
membranes and ideally their intracellular
c o n c e n t r a t i o n s s h o u l d be d e t e r m i n e d in the target
cells. This is, however, extremely difficult (cf.
Breter, M a i d h o f & Z a h n , 1978) a n d p h a r m a c o k i n e t i c
 Bo ODLIND et al.
studies following administration of AZA have
therefore mainly focused on blood levels of AZA
and especially of 6-MP.
The dosage of AZA in renal transplantation is
empirical. An especially difficult situation arises
when the transplant fails to function properly;
should AZA doses then be reduced? (cf. Calne, 1967;
Simmons, Kjellstrand, Buselmeier & Najarian, 1971;
Elion & Hitchings, 1975; Bach, 1975). We have
studied the pharmacokinetics of AZA (indirect
method) and 6-MP following administration of AZA
to uremic patients using a gas chromatography mass
spectrometry method for determination of 6-MP
(Floberg et al., 1981). In some patients we also
performed an in vitro immunosuppressive test by
studying the inhibition of mitogen stimulated
lymphocyte proliferation by patient serum. Our aim
was to evaluate whether there is any rationale for a
more individualized dosing of AZA in uremic
patients based on plasma concentration deter-
minations of AZA and/or 6-MP.
EXPERIMENTAL PROCEDURES
The patients
Clinical information about the patients is
summarized in Table 1. In fifteen uremic patients we
studied the pharmacokinetics of AZA (indirect
method; see below) and 6-MP after intravenous
administration of 100 mg AZA. AZA was provided
by The Wellcome Foundation Ltd, U.K., as the
freeze dried sodium salt and dissolved in saline
immediately before the injection. In five of these
patients the red blood cell concentrations of 6-MP
and AZA were also determined. The pharma-
cokinetic study was repeated in two of the patients
(MR and TM) following successful renal trans-
plantation. The inhibition of mitogen stimulated
lymphocyte proliferation by serum from six of the
patients (LA, L-E J, K-AL, MR and TM [the latter
post transplant] and EJ [no pbarmacokinetic infor-
mation]) was studied following the same schedule for
blood sampling as for the kinetic study. On a
separate day the pharmacokinetics of AZA and 6-
MP (Table 3) was studied in eight of the patients
following administration of 100 mg AZA orally after
having fasted overnight.
Blood and plasma sampling and analys&
Ten ml blood samples were drawn from an
indwelling venous catheter in one arm at 0, 5, 10, 15,
20, 30, 45, 60, 90 min and at 2, 3, 4, 5 and 6 h after
injection of 100 mg of AZA into a vein in the other
arm. On the following day, an oral dose of 100 mg
of AZA was given and blood samples were drawn at
0, 15, 30, 45, 60 and 90 rain and at 2, 3, 4, 5 and 6 h.
Immediately after sampling, half the blood volume
was put on ice and centrifuged at + 4°C, whereafter
plasma and red blood ceils were frozen separately at
- 18°C until analysis. The remaining blood volume
was kept at room temperature for at least 6 h,
whereupon plasma and red blood cells were
separated and stored frozen as above.
Direct extractive alkylation of 6-MP in plasma
with pentafluorobenzyl bromide as alkylating
reagent was done prior to analysis with gas
chromatography-mass spectrometry with multiple
ion detection (Floberg et al., 1981). The lower limit
of detection of the gas chromatographic-mass
spectrometric method used for the determination of
6-MP (Floberg et al., 1981) was 2 ng in a 1 ml
sample. The relative standard deviations at the 10
and 100 ng/ml levels of 6-MP in plasma were 6 and
10%, respectively. Samples of 6-MP were stable in
blood and plasma (Ding & Benet, 1979; Floberg et
al., 1981). Some degradation of 6-MP was seen after
thawing and therefore re-analysis of samples was not
performed. AZA was virtually stable in neutral and
weakly alkaline aqueous buffer and in plasma. A
rapid conversion to 6-MP occurred in the presence of
red blood cells (Elion & Hitchings, 1975; Ding &
Benet, 1979; Floberg et al., 1981), probably due to
an effect of glutathione in these cells (Elion &
Hitchings, 1975). This reaction was used for the
determination of azathioprine concentrations in the
samples assuming a quantitative conversion of AZA
to 6-MP (cf. Floberg et al., 1981). The total
concentration of AZA and 6-MP expressed as 6-MP
concentration was obtained from the samples stored
at room temperature. After subtraction of the 6-MP
concentrations determined in the samples put on ice
(and immediately separated), the AZA concentration
in the samples was obtained after correction for
molecular weight difference between AZA and
6-MP. From the above it follows that it is mandatory
to cool the blood samples immediately in an ice bath
to obtain accurate plasma values for 6-MP and AZA
following AZA administration (Ding & Benet, 1979;
Floberg et al., 1981).
Pharmacokinetic analysis
Data obtained after intravenous administration of
the drug were fitted to a linear two-compartment
model. The area under the concentration-time
curve (AUC) was estimated by the log-trapezoidal
 Serum Azathioprine and 6-Mercaptopurine Levels

Table 1. Patient data

Subject Sex Age Body Diagnosis Renal function Medication*

weight dialysis/creatinine
(kg) clearance
BA F 36 60 Diabetes CAPD Insulin
Prazosine (1 mg x 3)
Levothyroxine (0.1 mg x 1)
MB F 29 50 Chronic glomerulonephritis HD Norethisteron (10 mg x 1)
Levothyroxine (0.1 mg x 2)
Digitoxin (0.1 mg x 1)
MC M 43 60 Chronic glomerulonephritis HD Propranolol (160 mg × 2)
Hydralazine (50 mg x 2)
Bendroflumethiazide (2.5 mg x 1)
KD M 59 60 Chronic glomerulonephritis HD Digitoxin (0.1 mg x 1)
Metoprolol (100 mg x 2)
Hydralazine (50 mg x 2)
Prednisolone (10 mg x 1)
BH F 25 60 Chronic glomerulonephritis 5 ml/min Propranolol (80 mg x 2)
Hydralazine (50 mg x 2)
L-EJ M 44 80 Chronic glomerulonephritis HD Hydralazine (50 mg x 2)
Prazosine (1 mg x 2)
Atenolol (100 mg × 2)
Digitoxin (0.1 mg x 1)
Furosemide (500 mg x 2)
JL M 61 60 Mb Wegener CAPD Pindolol (5 mg x 1)
Penicilline (0.8 g x 2)
Warfarin
K-AL M 50 70 Diabetes CAPD Insulin
TM M 36 60 Chronic glomerulonephritis HD Prednisolone (5 mg x 1)
Transplanted 85 ml/min Isoxacilline (500 mg x 3)
Azathioprine (175 mg x 1)
Prednisolone (7.5 mg x 1)
Metoprolol (50 mg x 1)
UN F 33 50 Nephrosclerosist HD 0
JO M 41 70 Diabetes CAPD Metoprolol (100 mg x 1)
Insulin
TO M 32 65 Chronic glomerulonephritis HD Metoprolol (100 mg x 2)
HP F 50 80 Chronic glomerulonephritis HD T Digitoxin (0.1 mg x 1)
MR M 35 55 Chronic glomerulonephritis HD
36 55 Transplanted 55 ml/min Azathioprine (125 mg x 1)
Prednisolone (5 mg x 2)
Prazosine (1 mg x 2)
GU M 63 70 Chronic glomerulonephritis HD
* All medication withheld during the experiment.
-t Unclassified liver disease with markedly elevated alkaline phosphatases of hepatic origin. Galactose excretion test was
slightly impaired.

m e t h o d , adding the area to infinite time b e y o n d the
last sampling point as calculated f r o m C p / ~ where
Cp is the c o n c e n t r a t i o n at the last sampling point,
and /3 is the elimination rate constant o f the
terminal phase. The latter was calculated by linear
least square regression analysis. The clearance, CI,
was calculated by the equation: CI = d o s e / A U C .
The volume o f distribution, Vo, was calculated f r o m
Vo = Cl/fl. In the case o f 6 - M P a quanti-
tative conversion f r o m A Z A was assumed and the
 Bo ODLINDet al.
clearance of 6-MP was calculated from the corrected
" d o s e " i.e. a dose of 100 mg A Z A corresponds to a
calculated " d o s e " of 55 mg 6-MP. The following
parameters were also calculated from the data: t~/,~ =
half-time of elimination of the first phase (i.e. the
phase of distribution); t~p = half-time of elimination
of the terminal phase; t,~ = point of time for
maximal concentration.
A U C after oral administration was calculated as
above. No other pharmacokinetic parameters were
estimated owing to the relatively few plasma
concentration values obtained in each patient after
oral intake.
All pharmacokinetic data were analyzed by the
pharmacokinetic computer program I G P H A R M
(Gomeini & Gomeini, 1978). Mean values and
standard deviations were calculated by standard
equations using a Texas Instruments TI-51-III
programmable calculator.
Inhibition o f mitogen stimulated lymphocyte
proliferation by patient plasma
Mitogen stimulation was performed mainly as
described previously (Juhlin, Tufveson & Aim,
1977), with modifications for human lymphocytes.
Peripheral blood lymphocytes were prepared from
unrelated group O, Rh-healthy donors by puri-
fication of heparinized peripheral blood on
Leucoprep density gradients (Pharmacia, Uppsala,
Sweden). The lymphocyte layer on the gradients was
washed three times, counted in a hemocytometer and
the cell concentrations were adjusted.
The cells were cultured in RPMI 1640, containing
L-glutamine (2mM) (both from Flow Laboratories,
Irvine, Scotland), benzyl-penicillin (100 U/ml) and
streptomycin (100 tag/ml). The culture medium was
supplemented with 5% serum from the patients as
described above or 5% serum from the lymphocyte
donors. Cell suspensions in 0.i ml volumes per
culture, containing 5 x 104 peripheral blood
lymphocytes were cultured in round bottomed
Cooke microtiter plates in water saturated air with
507o CO2 at 37°C for 3 days. Leucoagglutin ( L A /
Pharmacia, Uppsala, Sweden), a phytohemagglutin
derivative, was added to the cultures at the time of
initiation at previously titrated optimal con-
centration. The incorporation of (3H)-methyl-
thymidine [(3H)-TdR] in the cultures during a 5 h
period on culture day 3 was taken as an indicator of
cell proliferation. 0.02/aCi of (~H)-TdR specific
activity 0.8 C i / m M (Amersham, U.K.) in 0.025 ml
0.85% NaCI was added per culture. The pulsed
cultures were processed with a flow multiple cell
culture harvester. The radioactivity was measured by
conventional liquid scintillation spectrometry.
The results are expressed as: percent of maximal
stimulation at
counts/min in culture at indicated time
-5min = 100 x
counts/min in culture at - 5 rain
RESULTS
A Z A and 6-MP in plasma after i.v. administration
The maximal plasma concentration ( C , , J of A ZA
after i.v. administration was generally found in the
first blood sample, obtained 5 min after dose. The
mean Cmox was 3.14/aM (Table 2), but the
interpatient variability was considerable (range
0 . 7 9 - 8.22/aM).
The plasma levels of A ZA fell rapidly after i.v.
administration (Fig. I). In most patients (12/15) the
best fit to the data was obtained with a bi-
exponential equation, but in three patients only one
phase of elimination from plasma was observed. The
mean A UC was 63/aM × rain with a range of
3 0 - 109. The mean total plasma clearance (CI) was
6.9 1./min (range 3 . 3 - 12.0) and the mean VD was
438 1. (range 115 - 1010). The mean half-time for the
first, rapid phase of elimination (tv~o) was 6.1 rain,
with a range of 1.5 - 14.5 min (Table 2). The mean
half-time of elimination of the terminal phase (t~/~)
was 50 rain, range 1 5 - 1 3 7 min (Table 2). After
240 rain AZA was not detectable in the plasma of
most of the patients.
The maximal plasma concentrations of 6-MP were
also generally found already in the first blood
sample, collected 5 min after dose (Table 3). The
mean Cmo~ was 1.66/aM (range 0.39-3.09). There
was no obvious correlation between the Cmo~values
for 6-MP and A ZA (cf. Tables 2 and 3). The mean
A U C for 6-MP was 64/aM x min (range 1 4 - 138).
The mean total plasma clearance was 8.0 l./min
(range 2 . 6 - 24.9) and the mean V/~ was 887 1. (range
150-3260). The mean t~,~ for 6-MP was 4.6 min
(range 1.6 - 9.3), and the mean t~:,o 74 min (Table 3).
With the exception of one patient, UN, who had a t:~:
value of 264 rain, the interpatient variability in t,:,~
was less than four-fold (range 2 4 - 9 8 ) .
After 240 min the plasma levels of 6-MP were
either very low or not detectable. In some patients
the plasma concentration curve tended to level off,
suggesting the presence of a third phase of
elimination. Whether a third phase of very slow
elimination of 6-MP exists cannot be established,
 Serum Azathioprine and 6-Mercaptopurine Levels 5

Table 2. A Z A in plasma after i.v. administration of 100 m g A Z A

Subject C~ax tin= t,~,o t,~,o A UC CI V~

(/aM) (rain) (rain) (rain) ~ M x rain) (l./min) (1.)
BA 7.43 5 1.5 33 71 5.0 240
MB 1.51 5 14.5 137 74 4.9 965
MC 2.57 10 5.6 66 109 3.3 315
KD 0,79 15 * 30 40 9.0 395
BH 3,55 5 5.2 22 100 3.6 115
LEJ 1,97 5 3.2 67 34 10.5 1010
JL 1.97 5 3.1 29 64 5.6 235
KAL 2.83 5 5.3 46 108 3.3 225
TM 1 2.17 5 6.8 15 38 9.4 205
2t (0.92) (5) (4.5) (19) (11) (29.8) (910)
UN 8.22 5 4.1 28 30 12.0 490
JO 1.45 10 * 34 71 5.1 255
TO 5.00 5 3.2 64 83 4.4 405
HP 2.11 5 13.9 62 37 9.8 885
MR 1 2.96 5 3.4 60 44 8.2 710
2t (2.37) (5) (14.3) (90) (72) (5.0) (655)
GU 2.57 5 8.9 * 37 9.8 125
2 3.14 6.05 49.5 62.7 6.92 438
S.D. 2.14 4.08 30.9 28.0 2.97 307
For explanation of abbreviations, see text.
* Only one phase o f elimination from plasma.
tPost-transplant values not included in statistics.

PM I o AZA in plasma Ig/ml

• 6-MP in plasma 400

12l&l & 6-MP in erythrocytes 1200 IJM

AZA in plasma nglml

• 6-MP in plasma

lOO 0.2 40

!! [OOo0 Time after dose intake (hours)

20

10

1 2 3 4
Time after dose intake (hours)

Figs 1 and 2. A Z A and 6-MP concentrations after administration of 100 mg A Z A intravenously (Fig. 1) and orally (Fig. 2)
in one patient, H P . Five hours after dose intake, A Z A and 6-MP were undetectable in all samples from this patient.
 Bo ODLIND et al.

Table 3. 6-MP in plasma after i.v. administration of 100 mg AZA

Subject C,n~ t,,,a, t ,:, t~.a A UC Cl Vc,

(/aM) (min) (min) (rain) (/aM × min) (l./min) (I.)
BA 2. l I 5 1.6 24 55 6.5 230
MB 1.38 5 3.0 84 59 6.1 740
MC 3.09 10 4.0 33 93 3.9 190
KD 0.39 15 * 41 23 15.7 935
BH 2.37 5 5.4 36 59 6.2 325
LEJ 1.05 5 1.8 91 14 24.9 3260
JL 2.17 5 4.6 37 42 8.5 450
KAL 2.30 5 3.0 98 55 6.6 940
TM 1 0.66 5 * 81 38 12.4 1450
2t (0.59) (5) (4.2) (40) (21 ) (17.2) (980)
UN 0.46 5 5.9 264 51 7.1 2710
JO 2.89 5 7.5 53 75 4.8 370
TO 2.37 5 5.0 40 138 2.6 150
HP 0.99 5 9.3 70 137 2.6 265
MR 1 1.38 5 5.3 65 66 5.4 510
2t (1.64) (5) (1.6) (22) (20) (18.2) (575)
GU 1.25 5 3.0 87 59 6.2 780
.f 1.66 4.57 73.6 64.3 7.97 887
S.D. 0.87 2.21 58.0 35.4 5.80 929
For explanation of abbreviations, see text.
* Only one phase of elimination from plasma.
tPost-transplant values not included in statistics.

since the p l a s m a c o n c e n t r a t i o n s o f 6 - M P during this erythrocytes occured simultaneous with that in

" p h a s e " were always close to the limit o f
d e t e r m i n a t i o n o f the m e t h o d .
The p h a r m a c o k i n e t i c study was repeated in two
patients, T M a n d M R , after successful renal
transplantation. Both patients showed higher
clearance for 6 - M P after the t r a n s p l a n t a t i o n (Tables
2 a n d 3).
AZA a n d 6 - M P in erythrocytes after i. v.
administration
The c o n c e n t r a t i o n o f 6 - M P in erythrocytes was
d e t e r m i n e d in seven o f the patients (Table 4). In five
o f these (KD, BH, H P , M R , G U ) the c o n c e n t r a t i o n
o f A Z A was also d e t e r m i n e d . In all five patients the
levels o f A Z A were very low or undetectable,
indicating t h a t virtually all A Z A h a d been converted
to 6 - M P within the erythrocytes. Only the 6 - M P
values are t h e r e f o r e s h o w n in Table 4.
T h e general shape o f the c o n c e n t r a t i o n - t i m e
curve for 6 - M P in erythrocytes was similar to t h a t in
p l a s m a (Fig. 1). T h u s , the m a x i m a l c o n c e n t r a t i o n in
p l a s m a in all seven patients. In four of the patients
a n initial phase o f very rapid elimination o f 6-MP
f r o m the erythrocytes (a) was followed by a slower
phase (/3). In three patients only one phase was
observed (Table 4). The 6 - M P c o n c e n t r a t i o n s fell
significantly slower in erythrocytes t h a n in p l a s m a
d u r i n g the terminal phase o f elimination, with m e a n
t~,2o values o f 96 a n d 61 min, respectively (P < 0.05,
t-test for paired data, two-tailed). Cm~x a n d A U C o f
6 - M P tended to be higher in erythrocytes t h a n in
plasma, but the difference was not statistically
significant ( P > 0.05).
A Z A a n d 6 - M P in p l a s m a after oral administration
Oral a d m i n i s t r a t i o n o f A Z A generated flat p l a s m a
c o n c e n t r a t i o n - time curves o f b o t h A Z A a n d 6 - M P
(Fig. 2). The m e a n C,,,x o f A Z A was less t h a n one
t e n t h o f that f o u n d after i.v. injection (Table 5) a n d
A Z A was generally detectable only between
15 - 30 m i n a n d 1 2 0 - 240 m i n after dose intake. The
m e a n oral A U C o f A Z A was 9.8/aM x min,
 Serum Azathioprine and 6-Mercaptopurine Levels
corresponding to 18% of the A UC of AZA after i.v.
administration.
The mean Cma,of 6-MP was 0.18/aM, with a range
of 0.05-0.53 laM (Table 5). The time for the
maximal plasma concentration varied markedly
(range 15 - 240 min), with a median value of 90 min.
6-MP was generally detectable between 3 0 - 6 0 rain
and 2 4 0 - 3 6 0 min after dose intake. The mean oral
A U C of 6-MP was 33 laM x min, representing 41%
of the A UC of 6-MP after i.v. administration of
AZA.
Inhibition o f mitogen stimulated lymphocyte
proliferation by patient serum
Figure 3 shows results from experiments, where
the LA-induced proliferation of peripheral blood
lymphocytes from an unrelated donor was inhibited
to various degrees by patient sera. Sera drawn 5, 10
and 30 min after intravenous administration of
100 mg AZA significantly (P < 0.001) inhibited the
LA-induced proliferation, independent of whether
the blood samples had been stored at room
temperature, or kept on ice (see above).
Sera drawn at 60, 180 and 360 min (data from the
later time not shown) did not significantly alter the
LA-induced proliferation. Sera drawn from these
patients 5 min prior to injection of AZA were never
inhibitory. The same was true for sera derived from
the lymphocyte donor (data not shown).
DISCUSSION
After intravenous administration AZA rapidly
disappeared from plasma in the uremic patients (Fig.
1, Table 2). This could partly reflect distribution of
AZA to blood cells and tissues, but is apparently
mainly due to the rapid conversion of AZA to 6-MP
as evidenced by the very rapid appearance in plasma
of 6-MP following AZA administration (Fig. 1,
Table 3). This reaction, which involves glutathione
and cysteine (Elion & Hitchings, 1975), occurs
mainly in the liver (where it is thought to be catalyzed
by glutathione-S-transferase) (Kaplowitz &
Kuhlenkamp, 1978; Watanabe, Hobara &
Nagashima, 1978). This conversion also occurs in
blood in vivo and in vitro, as a result of a
temperature and concentration dependent reaction
(Elion & Hitchings, 1975; Ding & Benet, 1979;
Floberg et al., 1981; Odlind, Grefberg, Hartvig,
Lindstr6m & L6nnerholm, 1981), which probably
involves the glutathione in the red blood cells.
Consequently, it is an absolute requirement in
pharmacokinetic studies of AZA, that blood samples
are immediately cooled and freeze-centrifuged to
obtain accurate plasma concentrations of AZA and
6-MP (Floberg et al., 1981). Unfortunately, this
procedure has not been clearly documented in some
previous studies (Maddocks, 1979b; Lin et al., 1980).
This might have resulted in falsely high 6-MP
concentrations and correspondingly low AZA
concentrations in the plasma.
Our previous finding that AZA added to whole
blood samples was converted to equimolar
concentrations of 6-MP (Floberg et al., 1981), is a
prerequisite for our use of this in vitro conversion as
an indirect method of determination of AZA. It does
not, however, exclude the possibility that AZA in
vivo to some extent can be metabolized in other ways
e.g. by oxidation (Chalmers, Knight & Atkinson,
1969; Elion & Hitchings, 1975) or by cleavage to
liberate 1-methyl-4-nitro-5-thioimidazole (cf. Elion
& Hitchings, 1975).

Table 4.6-MP in erythrocytes after i.v. administration of 100 mg AZA

Subject Cm~x t~ax t ,/~o t ,~,~ A UC

0aM) (rain) (rain) (min) (/aM x rain)
KD 2.82 15 * 89 315
BH 1.71 5 * 102 164
LEJ 1.38 5 4.7 135 57
TM 2 0.79 5 4.4 56 24
HP 1.12 5 * 49 59
MR 1 6.11 5 10 99 171
GU 1.84 5 10 142 105
.~ 2.25 7.3 96 128
S.D. 1.82 35.4 99.4
P NS <0.05 NS
* Only one phase of elimination from the erythrocytes.
 BO ODLIND et al.

Table 5. AZA and 6-MP in plasma after oral administration of 100 mg AZA

AZA 6-MP
Subject C,,~x t A UC A UCor~t × 1O0 Cmox t,,ax A UC AUCo,ol x 100
(/~M) (min) (pM × min) A UC,~ (pM) (min) ~ M × min) AUC,.~.
MB 0.03 15 1.7 2.3 0.08 240 19 32
MC 0.86 15 9.9 9.0 0.53 15 62 67
KD 0.01 90 0.7 1.8 0.05 60 5 22
BH 0.13 30 * * 0.13 60 11 19
TM 0.05 60 2.2 5.8 0.16 90 19 50
JO 0.10 240 11.6 16 0.20 240 47 63
TO 0.10 120 16.2 20 0.12 90 30 22
HP 0.16 30 26.4 72 0.16 120 68 49
£ 0.18 9.81 18.1 0.18 32.6 40.5
S.D. 0.28 9.36 24.7 0.15 23.7 19.2
* Too few samples to allow determination of AUC.

Mean plasma half lives for A Z A of 28, 10 and
12.5 min in patients with good renal function have
been reported by Maddocks (1979b), Ding,
Gambertoglio, Amend, Birnbaum & Benet (1980)
and Lin et al. (1980), respectively. The latter group
also found a similar plasma half life o f A Z A in
patients with reduced renal function. However, these
100
~ 8o
60
~ 40
o
n 20
/y---.q
510 3~0 60 180 (rain)
~ o,t
o,,
E accumulation of either o f these two substances in
Fig. 3. Capacity of serum from intravenously injected
patients to inhibit lymphocyte induced proliferation by LA.
Mean values + / - / 1 S.D. from six patients are shown.
*----* represents serum processed at room temperature and
O O serum kept on ice. The lower part of the figure
shows final plasma concentrations of 6-MP (striped bars)
and AZA (open bars) in the cultures (mean values).
values cannot be directly compared with our mean
half-time of elimination for the terminal phase of
50 rain, since neither of these authors identified
more than one phase of the plasma
c o n c e n t r a t i o n - t i m e curve. Ding et aL (1980)
reported a mean plasma clearance of A Z A of 57 m l /
m i n / k g which is approximately half the value found
in our patients (Table 2).
6-MP also disappeared rapidly from plasma; a
mean half-time of elimination for the terminal phase
(t,~:~) of 74 min in the uremic patients. Previous
studies of patients with acceptable graft function
have reported mean t,~,a values of 6-MP of 38 rain
(Lin et aL, 1980), 63 rain (Ding et al., 1980) and
114 min (Maddocks, 1979), respectively. The two
patients, who were retested after successful renal
transplantation, had higher plasma clearance values
for 6-MP. Plasma clearance values o f 6-MP have not
been reported p r e ~ - ~ , in normal subjects or
uremic patients.
Nevertheless, taking all observations into
consideration, it seems that renal function does not
influence the elimination rate of 6-MP (or of A Z A ;
cf. Table 2), and there appears to be no risk of
uremia. This is what one would expect since 6-MP is
metabolized mainly through oxidation by hepatic
xanthine oxidase to thiouric acid which is the main
product excreted in the urine. Thus, it has recently
been shown that pretreatment with allopurinol, a
xanthine oxidase inhibitor, increased the systemic
bioavailability of oral 6-MP considerably (Zimm,
Collins, Riccardi, O'Neill, Narang, Chabner &
Poplack, 1983). It is o f interest in this context that
the patient UN, who had the by far highest value o f
 Serum Azathioprine and 6-Mercaptopurine Levels
t,/2afor 6-MP, 264 min, was the only patient who also
had impaired liver function. On the other hand,
using isotope labelled A Z A , no great change in the
pharmacokinetics of the thiopurines was observed in
patients with reduced liver function (Bach, 1975).
Our finding that A Z A levels in erythrocytes were
low or undetectable support the hypothesis that once
within the cell, A Z A is rapidly converted to 6-MP,
probably due to the glutathione content of the
erythrocytes. The elimination of 6-MP from
erythrocytes was somewhat slower than the
elimination from plasma with t~/2~values of 96 and 61
rain, respectively. The explanation for this difference
is not clear.
After oral administration of 100 mg A Z A the
plasma concentrations of A Z A and 6-MP were quite
low (Fig. 2, Table 5) with mean maximal values of
20 ng/ml. This is in agreement with the plasma 6-MP
levels after oral A Z A reported by Maddocks (1979b)
and Lin et al. (1980) while Ding & Benet (1979)
found somewhat higher peak values of 4 5 - 7 5 ng/
ml. These findings are probably not due to poor
gastrointestinal absorption of AZA, since 70% of
the radiolabel was recovered from urine 0 - 4 8 h
after S3~ labelled A Z A was administered orally
(Elion, 1972). Therefore, as previously suggested
(Maddocks, 1979b), the low A Z A levels probably
reflect rapid conversion to 6-MP in portal blood and
liver ("first-pass" metabolism). Hence, the low
circulating levels of 6-MP after A Z A orally may be
somewhat surprising; this could reflect "first-pass"
metabolism also of 6-MP (Zimm et al., 1983) and
perhaps uptake of 6-MP in lymphocytes in portal
venous blood as suggested by Maddocks (1979b).
It is often difficult to define a relevant
bioavailability when studying a pro-drug. We found
m e a n AUC6_MPafter A Z A orally to be 41% of mean
A UC6_Mpafter A Z A i.v.; Ding et al. (1980) found this
value to be 60%. In a recent paper by Zimm et al.
(1983) a very low (average 16°/0) and variable
bioavailability of oral 6-MP was found. As
emphasized by Zimm et al. the low bioavailability of
oral 6-MP could have clinical significance in
maintenance therapy of leukemia. Comparative
studies of the pharmacokinetics of A Z A and 6-MP,
based on adequate methods of determination, are
presently missing. However, using isotope labelled
drug Elion (1972) found a higher gastrointestinal
absorption of A Z A than of 6-MP. In this context, it
is important to realise that, although differences in
immunosuppressive effects between A Z A and 6-MP
have been documented in vitro (Bach, Dardenne &
Fouriner, 1969; Medzihradsky, Hollowell & Elion,
1981; A1-Safi & Maddocks, 1983; cf. also Elion &
9
Hitchings, 1975) and in vivo (Calne, Alexandre &
Murray, 1962) there are also species differences
(Shehadeh, Gattman & Lindquist, 1970). No
controlled clinical trials of the comparative
immunosuppressive activity or toxicity of the two
drugs have been performed. Moreover, it seems that
the 5-mercapto-l-methyl-4-nitroimidazole moiety,
which is also liberated when A Z A is converted to
6-MP (cf. Elion & Hitchings, 1975) has little or no
immunosuppressive activity (Smith & Forbes, 1970;
A1-Safi & Maddocks, 1983) (contrary to previous
suggestions [Spreafico, Donelli, Bossi, Vecchi,
Standen & Garattini, 1973; Elion & Hitchings,
1975]). Thus, if the rapid in vivo conversion of A Z A
to 6-MP in humans is also taken into account, it is
more likely that A Z A in humans primarily acts as a
pro-drug to release 6-MP. 6-MP, in turn, is subject
to intracellular conversion - - by influence of
hypoxanthine-guanine-phosphoribosyl transferase
(HGPRT) - - to the corresponding ribonucleotide,
thioinosinic acid, which is thought to be the main
active metabolite, and to other thioanalogues of
purine derivates (Elion & Hitchings, 1975).
Thioinosinic acid inhibits the synthesis of adenine
and guanine-nucleotides in target cells, which affects
both RNA and DNA synthesis. Ribonucleotides do
not penetrate cell membranes and consequently their
formation can only be followed by determination of
intracellular concentrations, which is extremely
difficult (cf. Breter et al., 1978). On these grounds,
Elion & Hitchings (1975) have proposed that
detection of plasma levels of A Z A and 6-MP may
offer no predictive value with respect to the
therapeutic effectiveness or toxicity of the two
compounds. However, along with other investigators
(Ding & Benet, 1979; Zimm et al., 1983) we believe
that information of potential clinical value can be
obtained provided that one has access to highly
sensitive and specific methods of determination of
these drugs. A good example of this is the
demonstration of increased plasma concentration of
6-MP following pre-treatment with the xanthine
oxidase inhibitor allopurinol (Zimm et al., 1983);
this drug combination is well known to enhance the
immunosuppressive activity of A Z A / 6 - M P . In this
context, it is of interest that furosemide has been
found to inhibit the hepatic metabolism of A Z A in
human liver in vitro (von Bahr, Glaumann, Gudas &
Kaplowitz, 1980). This potentially important drug
interaction has, however, not yet been studied in the
clinical setting.*
An important finding in support of the biological
relevance of measuring 6-MP plasma levels is given
by the parallel decline in inhibition of mitogen
10 Bo ODLIND et aL

induced lymphocyte proliferation in vitro by patient
sera and the A Z A / 6 - M P concentrations in the same
sera (Fig. 3). The sensitivity of the immunological
test system used here is considerably higher than
those previously reported e.g. by A1-Safi &
Maddocks (1983), most likely because our culture
conditions were carefully titrated to a point where
the number o f lymphoc~,tes added to the cultures
were linearly proportional to the tritium methyl-
thymidine uptake in the cultures. We found a
minimum effective concentration o f 6-MP to inhibit
mitogen induced lymphocyte proliferation in the
cultures of about 0 . 0 2 - 0 . 0 4 taM.
Interestingly, this suggests the possibility of a
biological effect even of the low plasma
concentrations of 6-MP obtained after oral
administration of A Z A , contrary to the proposal by
Maddocks (1979b). In spite of the high sensitivity of
the in vitro immunological test system we could not
detect any difference in inhibition of lymphocyte
proliferation between sera from erythrocyte exposed
and non-exposed samples. This could either mean
that A Z A did not significantly add to the immuno-
suppressive action of 6-MP in the test system or that
A Z A could have been converted to 6-MP in the
tissue culture system. Previous attempts to relate
pharmacokinetics to immunosuppressive effects
have employed isotope labelled A Z A / 6 - M P and the
rosette inhibition assay and have generally shown a
bad correlation (cf. Bach, 1975). The use of a
sensitive and specific method of determination of
A Z A / 6 - M P and a sensitive immunological test
system as described in this paper offers a better tool
in further studies of plasma c o n c e n t r a t i o n -
immunosuppressive effect/toxicity relationships of
these drugs. Pending such studies, immuno-
suppressive treatment with A Z A in renal trans-
plantation and auto-immune diseases must still
follow a rather fixed dose regimen with dose
reduction when leukopenia appears. Although there
seems to be little danger of accumulation of A Z A /
6-MP itself in patients with severely reduced renal
function, the possibility exists that metabolite(s)
normally excreted by the kidneys, may be retained
and increase the toxicity of A Z A / 6 - M P in uremia.
Moreover, it has been suggested that the uremic
condition p e r s e exerts some immunosuppressive
effects. A careful supervision of the patients is
therefore mandatory.
Acknowledgements - - This work was supported by the
Wellcome Foundation Ltd., Sweden, the Tore Nilsson fund
and the Maud & Birger Gustavsson fund.

REFERENCES

AL-SAFI, S. A. & MADDOCKS, J. L. (1983). Effects of azathioprine on the human mixed lymphocyte reaction (MLR). Br. J.
clin. Pharmac., 15, 203 - 209.
BACH, J. F. (1975). The mode of action of immunosuppressive agents. In Frontiers o f Biology, Vol. 41 (eds Neuberger, A.
and Tatum, E. L.) North-Holland, Amsterdam.
BACH, J. F., DARDENNE, M. & FOUR1NER, C. (1969). In vitro evaluation of immunosuppressive drugs. Nature, Lond., 222,
998- 1000.
BAHR, C. VON, GLAUMANN,H., GUDAS, J. & KAPLOWlTZ,W. (1980). Inhibition of hepatic metabolism of azathioprine by
furosemide in human liver in vitro. Biochem. Pharmac., 29, 1439-1441.
BERENBAUM, M. C. (1971). Is azathioprine a better immunosuppressive than 6-mercaptopurine? Clin. exp. I m m u n . , 8,
1-8.
BRETER, H.-J., MAIDHOF,A. & ZAHN, R. K. (1978). The quantitative determination of metabolites of 6-mercaptopurine in
biological materials. III. The determination of '4C labeled 6-thiopurines in L51784 cell extracts using high-pressure
liquid cation-exchange chromatography. Biochim. biophys. Acta, 518, 205- 215,
CALNE, R. Y. (1967). Renal Transplantation, 2nd edn. Edward Arnold, London.
CALNE, R. Y., ALEXANDRE, G. P. J. • MURRAY, J. E. (1962). A study of the effects of drugs in prolonging survival of
homologous renal transplants in dogs. Ann. N . Y . Acad. Sci., 99, 749-761.
CHALMERS,A. H., KNIGHT, P. R. 8/; ATKINSON,W. R, (1969). 6-thiopurines as substrates and inhibitors of purine oxidase:
a pathway for conversion of azathioprine into 6-thiouric acid without release of 6-mercaptopurine. A ust. J. exp. Biol.
med. Sci., 47, 263- 273.

* The only patient (L-E J) in this study with concomittant furosemide therapy did not differ markedly from the other
patients in AZA pharmacokinetics (Table 2).
 Serum Azathioprine and 6-Mercaptopurine Levels 11

DING, T. L. & BENET, L. Z. (1979). Determination of 6-mercaptopurine and azathioprine in plasma by high-performance
liquid chromatography. J. Chromat., 163, 281-288.
DING, T. L., GAMBERTOGLIO,J. G., AMEND, W. J. C., BmNBAUM, J. & BENET, L. Z. (1980). Azathioprine bioavailability
and pharmacokinetics in kidney transplant patients. Abstract. 81st Annual Meeting o f the American Society o f
Clinical Pharmacology and Therapeutics, San Francisco, 19-21 March, p. 26.
ELION, G. B. (1972). The significance of azathioprine metabolites. Proc. R. Soc. Med., 65, 2 5 7 - 260.
ELION, G. B. & HITCHINGS, G. H. (1975). Azathioprine. In Handbuch der Experimentellen Pharmacologie, Vol. 38, pp.
4 0 4 - 425. Springer, Berlin.
FLOBERG, S., HARTVIG, P., LINDSTROM, B., LONNERHOLM, G. & ODL1ND, B. (1981). Extractive alkylation of
6-mercaptopurine and determination in plasma by gas chromatography mass spectrometry. J. Chromatog., 225,
73-81.
GOMEINI, C. & GOMEINI, R. (1978). IGPHARM-Interactive package of pharmacokinetic analysis. J. Pharmac.
Biopharmac., 11, 3 4 5 - 361.
JUHLIN, R., TUFVESON, G. & ALM, G. V. (1977). Role of the thymus in generation of lymphocyte functions. II. Serum
mediated inhibition of development of mitogen reactive lymphocytes in embryonic mouse thymus organ culture. Cell
Immun., 32, 193-202.
KAPLOWITZ, N. & KUHLENKAMP, J. (1978). Inhibition of hepatic metabolism of azathioprine in vivo. Gastroenterology, 74,
90 - 92.
LIN, S.-N., JESSUP, K., FLOYD, M., WANG,T.-P. F., BUREN, C. VAN T., CAPRIOLI,R. M. & KAHAN,B. D. (1980).
Quantitation of plasma azathioprine and 6-mercaptopurine levels in renal transplant patients. Transplantation, 29,
290 - 294.
MADDOCKS, J. L. (1979a). Assay of azathioprine, 6-mercaptopurine and a novel thiopurine metabolite in human plasma.
Br. J. Clin. Pharmac., 8, 273-278.
MADDOCKS, J. L. (1979b). Progress in immunosuppression. In Topics in Therapeutics, Vol. 5 (eds Davies, D. M. and
Rawlins, M. D.) pp. 3 3 - 44.
MEDZIHRADSKY, J. L., HOELOWELL, R. P. & ELION, G. B. (1981). Differential inhibition by azathioprine and
6-mercaptopurine of specific suppressor T cell generation in mice. J. Immunopharmac., 3, 1 - 16.
NARANG, P. K., YEAGER, R. L. & CHATTERJI, D. C. (1982). Quantitation of 6-mercaptopurine in biological fluids using
high-performance liquid chromatography: a selective and novel procedure. J. Chromat., 230, 3 7 3 - 380.
ODLIND, B., GREEBERG, N., HARTVIG, P., LINDSTR()M, B. & LONNERHOLM, G. (1981). Pharmacokinetics of azathioprine
and 6-mercaptopurine: Methodological aspects and preliminary results in uremic patients. Scand. J. Urol. Nephrol.,
Suppl 64, 213-219.
SHEHADEH,I. n., GATTMAN, R. D. & L1NDQUIST, R. R. (1970). Renal transplantation in the inbred rat. XV. An assay study
of three immunosuppressive drugs. Transplantation, 10, 66.
SIMMONS, R. L., KJELLSTRAND, C. M., BUSELMEIER, T. J. & NAJARIAN, J. S. (1971). Current practice of renal
transplantation at the University of Minnesota. Minn. Med., 54, 115- 120.
SMITH, J. L. & FORBES, I. J. (1970). Inhibition of protein synthesis in human lymphocytes by thiopurines. Aust. J. exp.
Biol. reed. Sci., 48, 2 6 7 - 276.
SPREAFICO, F., DONELLI, M. G., BOSSI, A., VECCHI, A., STANDEN, S. & GARATTINI, S. (1973). Immunodepressant activity
and 6-mercaptopurine levels after administration of 6-mercaptopurine and azathioprine. Transplantation, 16,
269 - 278.
WATANABE,A., HOBARA, N. & NAGASHIMA,H. (1978). Demonstration of enzymatic activity converting azathioprine to
6-mercaptopurine. Acta Med. Okayama, 32, 173 - 179.
Z1MM, S., COLLINS, J. M., R1CCARDI, R., O'NEILL, D., NARANG, P. K., CHABNER, B. & POPLACK, D. G. (1983). Variable
bioavailability of oral mercaptopurine. Is maintenance chemotherapy in acute lymphoblastic leukemia being
optimally delivered? New Engl. J. Med., 308, 1005- 1009.