Cumulative Pharmacokinetic Study of Oxaliplatin...
Transcript of Cumulative Pharmacokinetic Study of Oxaliplatin...
Vol. 3. 891-899. June 1997 Clinical Cancer Research 891
Cumulative Pharmacokinetic Study of Oxaliplatin, Administered
Every Three Weeks, Combined with 5-Fluorouracil in
Colorectal Cancer Patients’
Erick Gamelin,2 Anne Le Boull,
Mich#{232}le Boisdron-Celle, Alain Turcant,
R#{233}myDelva, Annick Cailleux, Ana#{239}sKrikorian,
Silvano Brienza, Esteban Cvitkovic,
Jacques Robert, Francis Larra, and Pierre Allain
Department of Medical Oncology and Clinical Pharmacology. CentrePaul Papin. Centre Regional de Lutte Contre Ic Cancer, 49033Angers Cedex lE. G.. M. B-C.. R. D.. F. LI: Laboratory ofPharmacology. Centre Hospitalier Universitaire, 49000 Angers
IA. L. B.. A. T.. A. C.. P. Al: Department of Medical Oncology.Centre Hospitalier Universitaire Paul Brousse, 94800 Villejuif IE. C.]:
Laboratory of Biochemistry and Pharmacology. Institut Bergonnie.
33076 Bordeaux [J. RI: and Debiopharni France, 94220 Charenton
Le Pont IA. K.. S. B.l. France
ABSTRACT
The cumulative pharmacokinetic pattern of oxaliplatin,
a new diamminecyclohexane platinum derivative, was stud-
ied in patients with metastatic colorectal cancer. Oxaliplatin
was administered by i.v. infusion (130 mg/m2) over 2 h every
3 weeks, and 5-fluorouracil and leucovorin were adminis-
tered weekly. A very sensitive method, inductively coupled
plasma-mass spectrometry, allowed for the determination of
total plasma and ultracentrifugable (UC) and RBC platinum
levels on day 1, at 0, 2, and S h, and on days 8, 15, and 22.
Sixteen patients underwent three or more courses, and six of
them underwent six or more courses. The platinum concen-
tration curves were quite similar from one course to an-
other, with a high peak value 2 h after administration (day
1, Cm,,x 3201 ± 609 �.Lgfliter) and a rapid decrease (day 8,
443 ± 99 jig/liter). Cmax of both total and UC platinum
levels in plasma remained unchanged throughout the treat-
ment. The mean total platinum half-life in plasma was 9
days. We found residual levels of total platinum on day 22
(161 ± 45 rig/liter), but we observed no significant accumu-
lation for the four first cycles (P 0.57). In contrast, plat-mum accumulated significantly in RBCs after three courses
Received 8/5/96: revised 2/24/97: accepted 3/5/97.
The costs of publication of this article were defrayed in part by the
payment of page charges. This article must thereftre be hereby marked
advertise,neiii in accordance with I 8 U.S.C. Section 1734 solely to
indicate this fact.
I This work was sponsored in part by Debiopharm France. S.A.. Charen-
ton Le Pont. France. We acknowledge the Coniit#{233}D#{233}parteniental de
Maine et Loire de Ia Ligue Nationale Contre Ic Cancer and the Fonda-tion Langlois for their financial support. This work was presented in part
at the 87th Annual Meeting of the AACR, held in April 1996 inWashington. DC.2 To whom requests for reprints should be addressed. Phone:(33)41 3527(8): Fax: (33)41 4831 90.
(+91 % at day 22 of the third cycle versus day 22 of the first
cycle, P = 0.000018), and its half-life there was equivalent to
that of RBCs. The patterns of UC and total platinum con-
centration curves were very similar and correlated signifi-
cantly (P < 106) at all sampling times. The mean UC:total
platinum ratio was 15% at day 1 and 5% at days 8, 15, and
22 in the 3-week treatment course. Unlike cisplatin, which
rapidly accumulates in plasma as both free and bound plat-
mum, oxaliplatin does not accumulate in plasma, but it does
accumulate in RBCs, after repeated cycles at the currently
recommended dose ( 130 mg/rn2) and schedule of adminis-
tration (every 3 weeks).
INTRODUCTION
Oxaliplatin is a diammine cyclohexane platinuni complex
that is active in several solid tumor types. especially in some
cisplatin/carboplatin refractory diseases such as colorectal can-
cers ( 1-3). Its toxicity profile is different from that of cisplatin
because it is devoid of nephrotoxicity at the recommended doses
and schedule (2). Oxaliplatin induces acute peripheral sensitive
neurotoxicity. characterized by acral paresthesia or dysesthesia
affecting the laryngo-pharyngeal area and the upper and lower
limbs. Acral paresthesia increases in incidence. intensity. and
duration with repeated treatments and is consistently reversible
(4).Some preliminary results concerned the early pharmacoki-
netics. distribution. and elimination of ultrafiltrable and total
platinum after a single oxaliplatin administration to patients (5).
but the long-term platinum kinetics remain unknown. The parent
drug. cisplatin. accumulates in plasma with repeated adminis-
trations (6). and it induces a peripheral neurotoxicity dependent
on the cumulative dose, as does oxaliplatin. Thus. for the de-
tection of any accumulation occurring with iterative courses of
therapy, the evaluation of oxali platin pharmacokinetics ap-
peared to be of major interest.
Previous studies have been carried out regarding the long-
term behavior of free and bound platinum after administration of
organoplatinum drugs. They were hampered due to the lack of
sensitivity of the analytical technique used. which was atomic
absorption spectrometry (7-9). A more recent technique.
ICPMS.3 is much more sensitive (10-13). This technique has
allowed us to establish the long-term pharmacokinetics of plat-
mum after iterative cisplatin administrations (6). We detected
significant levels of total and UC platinum in plasma. up to 3
weeks after cisplatin administration. showing that a progressive
:i The abbreviations used are: ICPMS. inductively coupled plasma-mass
spectrometry: UC. ultracentrifugable: 5-FU. 5-fluorouracil: Hh. hemo-
globin: Cl-C6. course I-course 6.
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
892 Oxaliplatin in Colorectal Cancer
accumulation of both total and UC platinums occurred upon
iterative infusions.
We have applied this technique to the study of platinum
pharmacokinetics after several administrations of oxaliplatin.
Actually. little is known about oxaliplatin mnetabolites. In
plasma, oxaliplatin loosens oxaloacetate residue and is quickly
metabolized in dach-platinuni. No oxaliplatin is detectable 2 h
after administration. Other nietabolites probably exist, but they
are not known. The techniques currently used measure platinum
and do not permit the separate study of oxaliplatin and other
platinum derivative metabolites.
The patients presented with advanced colorectal cancer that
was resistant to 5-FU. They were treated with oxaliplatin plus
5-FU and leucovorin, because some publications have shown
previously the synergistic clinical activity between both drugs
and a good response rate in 5-FU-resistant colorectal cancer (3,
14). The objective of the present study was to characterize the
long-term retention of oxaliplatin. by determining total and UC
plasma platinuni levels, as well as RBC platinum concentra-
tions. We can assume that 5-FU does not modify oxaliplatin
pharmacokinetics. It also does not influence the pharmacokinet-
ics of cisplatin and carboplatin. Moreover, it is poorly bound to
proteins. and its metabolism is completely different from that of
oxaliplatin.
PATIENTS AND METHODS
Patients. All of the patients had histologically proven.
metastatic colorectal cancer that was refractory to the 5-FU and
leucovorin combination. All patients gave written informed
consent and were more than 18 and less than 70 years old.
Perfcrmance status < III (by WHO criteria) was required. as
well as normal bone marrow function. Noninclusion criteria
were brain metastasis. expected survival of <3 months. and a
cardiac condition contraindicating 5-FU treatment. All patients
except two had normal renal function (creatinine clearance, >65
ml/min/l .73 m2). All of them had normal hepatic function and
no sign of ascitis or pleural effusion.
Treatment. Chemotherapy consisted of oxaliplatin given
in combination with 5-FU and folinic acid. All patients had
long-term venous access, established either by means of a cath-
eter or by means of an implantable chamber device. Oxaliplatin
was given at I 30 mg/m2 in 250 ml of sterile 5% glucose solution
as a 2-h infusion. every 3 weeks. 5-FU was administered after
oxaliplatin infusion and repeated weekly. by means of an 8-h
infusion in I liter of 0.9% saline serum through a battery-
operated pump. The initial dose of 1300 mg/m2 was adjusted
weekly for each patient. according to 5-FU plasma assays and to
a table of 5-FU dose adjustment (15. Folinic acid was admin-
istered at 2(X) mg/m2 by iv. bolus just before (0 h) and at 4 h
after 5-FU infusion.
Oxaliplatin was administered in association with prophy-
lactic antiemetics. ondansetron, and corticosteroids at conven-
tional doses. Other symptomatic treatments were allowed when-
ever required. Previous treatments for concomitant conditions
were continued at the same dose.
Blood Sampling. For each cycle. on the basis of oxali-
platin infusion. collection times of 5-ml blood samples were:
day I, 0. 2. and 5 h: days 8, 15. and day 22, befare the next
oxaliplatin infusion. They were collected in heparinized tubes.
The plasma was immediately separated by centrifugation at 4#{176}C.
Globular pellets were stored, and plasma was split into two
aliquots. one for the total platinum assay. without further prep-
aration, and one for the UC platinum assay.
Platinum Assay. Platinum quantification was performed
with the ICPMS method described previously ( 10). The spec-
trometer used was an Elan 5000 (Perkin-Elmer Corp.).
Total platinum in plasma and RBC platinum levels were
measured after a 20-fold dilution in a solution containing Eu-
ropium (Sigma Cheniical Co.: 100 g/liter). an internal standard.
For UC platinum assay. the plasma was centrifuged for 3 h at
100,000 rpm using rotor TLA 100.2 (Beckman Instruments,
Fullerton. CA). The supernatant was subsequently diluted 10-
fold in the Europium solution.
The calibration was performed with I g/liter platinum
standard solution (Sigma). diluted in a saline solution (Titrisol;
Merck) that contained 8.39 g/liter NaCI. The diluted solutions
were introduced into the nebulization chamber of the ICPMS,
using a peristaltic pump from Gilson.
The method was linear between 0 and 5000 jig/liter of
platinum. The recoveries were close to 100C/c. The within- and
between-day coefficients of variation were lower than 2C/c for
total. UC. and RBC platinum. The estimated limit of quantifi-
cation was I jig/liter.
Under the conditions of this study. the limit of quantifica-
tion was estimated at 1 jig/liter in plasma. The sample concen-
trations ranged from I to 5000 jig/liter.
Platinum levels in RBC were expressed both as jig/liter of
cell volume and as jig/g of Hb.
RESULTS
Patient Characteristics. Seventeen patients were in-
eluded in the study between June 1994 and April 1995. One
patient refused plasma assays after inclusion. Sixteen patients
could be investigated throughout their treatment (mean number
of cycles. 5; range, 3-12). Eight of them underwent 3 courses,
two underwent 4 courses. two had 6 courses, two had 7 courses,
one patient had I I courses. and one patient had 12 courses.
Eighty-five courses of oxaliplatin were administered. The
3-week interval between courses was maintained in 58 of 68
intervals (85�k). and 64 of 68 intervals were no longer than 4
weeks. Treatment was delayed for 10 courses in nine patients
due to gastrointestinal toxicity (2 courses). myelosuppression (2
courses), personal convenience (2 courses), surgery (I course),
and other reasons not related to toxicity (3 courses). The inter-
vals between delayed courses were 28 (six courses), 36 (one
course), 43 (two courses), and 108 days (one course, for
surgery).
The treatment was stopped due to: (a) tumor progression in
seven cases (three after 3 cycles and four after 4. 6. 7. and 12
cycles): (b) drug-related toxicity in four cases (cardiac toxicity.
gastrointestinal tract toxicity. myelosuppression, and skin tox-
icity); (�) a quite stable disease in two cases (after 8 and 1 1
cycles); and (d) patient treatment refusal in two cases. after 3
and 4 courses.
All oxaliplatin treatments were administered at the recom-
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Table I Incidence of toxic events in the population of patients for the first three courses and throughout the whole treatment
First 3 months treatment. WHO grade
Toxicity
Mucositis
Diarrhea
Peripheral neuropathy
Anemia
LeukopeniaThrombopenia
5
4
65
0
II
5I
31
3
III
5
1
0
7
69
I 04
Whole treatment. WHO grade
II
8
1
6
Ill
7
1
0
35261131173349±796 (ii 6)
3201±60 (n = 16) 3095+542
(n=l6)9
2267±276
242±64.
(n- l6)
2619±805
(n � 8)2316±463
� 2142±229 2108±227
266±34 310±43 302±46
161±45 187±31 � 1104i31
2914±571
(n -�6)
2316i343
302t31l
2011±13
2664i365
364±61
225±22
1475±933
(n 4)
3287±424
7311± 125
�
3276 t’187
(ii 2)
2503/625
Clinical Cancer Research 893
4500
4000
3500
r� 3000C
0
� 2500Ca)C)
C8 2000
C
.� 1500
0.
1000
500
0
0 20 40 60 80 days 100 12(1 141) l(l) ISO
Fig. I Mean total platinum concentrations in plasma. Concentrations did not increase significantly throughout eight courses. Only a trend towarda slow increase of residual levels was observed from the first to the seventh course ( + 50’4).
mended dose of I 30 mg/m2 and given as a 2-h infusion. The
median cumulative dose of oxaliplatin was 520 mg/m2 (range,
260-1560 mg/m2).
Toxicity. Toxic events, evaluated before each new cycle
according to WHO criteria, are displayed in Table I . No toxic
death occurred during this trial. A patient presented with 5-FU
related acute angina pectoris at home, the night after a treatment.
It spontaneously resolved before the physician came. Another
patient had a toxidermia. In these two cases, the treatment was
stopped. and the toxic event disappeared.
Total Platinum Characteristics. The mean plasma 1ev-
els of total platinum are presented in Fig. I . All courses are
characterized by a high peak value at 2 h after the end of
administration, followed by a rapid decrease. The plasma levels
measured after S h were 70% of the levels measured after 2 h,
and those measured after 8, 1 5, and 22 days were I 5. 7, and 3%
of the peak value, respectively. It should be noticed that signif-
icant plasma levels of platinum ( 16 1 ± 45 jig/liter after the first
course) were still measured 22 days after administration. The
mean terminal half-life of total platinum in plasma determined
from three points, measured on days 8, 15, and 22, was 9 ± 0.7
days (range. 8.42-9.66; Table 2). These results illustrate the
reproducibility of the kinetics after iterative administrations.
For 12 patients. late samplings performed on days 28 (six
patients). 36 (four patients). 58 (one patient). and 108 (one
patient) showed residual platinum levels: 183 jig/liter (range.
I I 1-272 jig/liter). 109 jig/liter (range. 94-I 29 jig/liter). 52
jig/liter. and 13 jig/liter. respectively. These results illustrate the
very long retention of platinum in plasma. In these cases. a
breakdown of the platinum concentration slope was observed on
day 22, and the terminal half-life was 15 days.
Total platinum levels at early sampling times did not in-
crease significantly between courses (Fig. I and Table 3). Only
a trend toward a slow increase of residual levels was observed
from the first to the seventh course (+SOC/d. The slope of the
residual levels regression curve was 0.71 (P < 10 �). However,
the difference was not statistically significant when all of the
sampling times were taken together and when the first course
was compared with the third (P 0.13) or the fourth (P 0.57)
course.
Pharmacokinetics of UC Platinum. The pattern of the
UC platinum plasma concentration curves is similar to that of
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
I
Total platinum 8.4 ± 1.1 8.4 ± 1.31 8.7 ± 0.78
UC platinum 6.8 ± 2 6.4 ± I 7.2 ± 0.7
RBC platinum 50 ± 10 46.6 ± I I 45.5 ± 14
3 4 6 7 8
9.4 ± 0.7 8.5 ± 0.5
8.5
894 Oxaliplatin in Colorectal Cancer
Table 2 Mean platinum half-lives in plasma (total and UC platinum) and in RBCs for successive courses. Half-lives were very stable
throughout the treatment. Mean platinum half-life in RBCs was almost equivalent to that of RBCs. suggesting that the binding might be covalent.
Course (days)
5
9.7 ± 0.47 9.3 ± 0.94 9.5 ± 0.5
8 6.5 6.8 ± 0.8
Table 3 Comparison of platinum concentrations in plasma and RBCs at the third and sixth courses vs. the first course. Total platinum did not
accumulate in plasma. and neither did UC platinutii. In contrast. platinum significantly and rapidly increased in RBCs.
C3 vs. Cl C6 vs. Cl C6 vs. C3
TP UP RBC P TP UP TP UP
+I2�/c +13% -1%NS NS NS NS
+17% -6% +24% +27%
<0.05 - <0.01 -
+50% +20% +17%
<0.01 NS - NS
Sampling time
2h
(7( change
P NS” NS
+73%
<0.01
5h
(/( change
P
-5%
NS
-24%
NS
+57%
<0.01
8 days
(4 changeP
+18%
<0.05
�20%
NS
+ 107%
<0.01
15 daysC/� change
P
+27%
<0.01
+40%
NS
+96%
<0.01
22 days
%change
P
+16C/
NS
+21%
NS
+91%
<0.01
- ‘, �‘ not significant.
total platinum (Fig. 2). Early high peaks appeared at 2 and 5 h
after oxaliplatin administration and were followed by a rapid
decrease. The UC platinum elimination half-lives were some-
what shorter than the total platinum ones and did not change
throughout the treatment (7.15 ± 0.7 days; Table 2). UC plat-
mum did not accumulate in plasma (Table 3). although small but
significant levels of UC platinum (10 jig/liter) were measured
until 22 days after every course (Fig. 2). The values were not
statistically different between the courses. whatever the sam-
pling time.
UC platinum plasma levels were closely and strongly cor-
related with total platinum plasma levels. The mean ratio of UC
versus total platinum was 7.7%, and the coefficient of correla-
tion was r = 0.94 (P < l0�6). It did not change throughout the
courses but changed with the sampling time within each course,
and two different patterns could be observed. For early sam-
plings (2 and 5 h). with platinum levels > 1,000 jig/liter, the
mean percentage of UC versus total platinum was I 5%. and the
coefficient of correlation was r = 0.69 (P < l0�6; Fig. 3). For
late samplings (days 8. 15, and 22) and for total platinum plasma
levels less than bOo jig/liter, the percentage was 5%, and the
coefficient of correlation was r = 0.65 (P < l0�6; Fig. 3).
Platinum Accumulation in RBCs. Platinum was meas-
ured in RBCs for eight patients over three courses of treatment
(Fig. 4). Platinum levels in RBCs increased rapidly during
infusion and reached a maximal value at 2 and 5 h; afterward,
they declined slowly (Fig. 4). Compared to the values obtained
after the first course. the levels at the second and third cycles
were significantly higher (Fig. 4 and Table 3). For the first
cycle, platinum levels in RBCs were close to the total platinum
values on day 1, 2 and S h (83 ± 12 and 130 ± 9%. respec-
tively), but they were strikingly higher afterward, reaching
8 10 ± 40% of plasma levels on day 22. On day 22, the residual
platinum concentrations in RBCs represented one-half of the
peak values, indicating a long storage in this compartment. The
mean platinum terminal half-life in RBCs, calculated with the
concentrations on days 8, 15, and 22, was 48 ± 10.5 days. It did
not change with the course (Table 2). In three instances, it
reached 70 days.
We looked for a link between the variations of RBC
platinum levels and those of Hb because, for three patients,
platinum RBC levels increased unexpectedly from day 8 to day
15. A spontaneous rise of Hb level occurred at the same time.
which could explain the trapping and the increase of platinum in
RBCs. Thus, we calculated platinum RBC concentrations in
jig/g Hb, and these platinum/Mb ratios were always lower on
day 15 than on day 8. We observed a progressive accumulation
from the first to the third treatment course. Thus, we calculated
platinum RBC concentrations in jig/g Hb, and these plati-
numlHb ratios were always lower on day I 5 than on day 8. We
observed a progressive accumulation from the first to the third
treatment course: on day I, 2 h, 7.4 ± 1.67, 10.4 ± 2.11, and
14.28 jig/g Hb; on day 8, 4.91 ± 1.23, 8.43 ± 1.98, and 10.78
�.Lg/g Hb; on day 15: 4.82 ± 1.51, 7.02, and 8.8 jig/g Hb: and on
day 22, 3.92 ± 0.77, 6.2 ± 1.43, and 7.14 ± 0.64 jig/g Hb
(values are for Cl, C2, and C3, respectively).
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
700
600
�. 500
C0
� 400CC)C)C
I 300
�. 200
100
0
0 20 40 60 80 00 20 4(1
days
900
800
700
600 �
�a
500 �C)C)C0
‘iou C)
300 .��a0
200 L)
I 00
0
I 0(1
01)
81) �
71)C0
60 �
CC)
50 �0
U
40 E
C
30 .�
20
I0
0
Fig. 3 Close correlation between UC and total platinum concentrations in plasma (346 samples. Two different patterns could he observed. A. 6rearly sampling times (2 and 5 h) and high total platinum levels (>1000 �ig/1iter), the coefficient of correlation r was 0.69, P < 10 “. the ratio UC:totalplatinum in plasma was 15%. and the coefficient of the linear regression curve was 0. 1 1. B. for late sampling times (days 8. I 5. and 22) and lowplatinum levels (< 1000 jig/liter). the coefficient of correlation r was 0.65, P < l06. the ratio was 5%. and the coefficient of the linear regressioncurve was 0.048.
3800 3300
total platinum
2800
concentration
2300 1800
( .tgft.)
800 6(1(1
total I)latinLita
400
concentration
20))
( (tgi’I.)
Clinical Cancer Research 895
Fig. 2 Mean UC platinum con-
centrations in plasma throughout
seven courses. UC platinum did
not accumulate, although small
but significant levels of UC plati-
nuni were measured until 22 days
after every course.
At late sampling times (days 8, 15 and 22), RBC platinum
levels were higher than total plasma platinum levels but corre-
lated significantly with them (coefficients of correlation: r =
0.67 on day 8, r = 0.75 on day 15, and r 0.55 on day 22, P <
10 6; Fig. 5). The same observation could be made with UC
platinum. This relationship was not found for the early sam-
plings.
As mentioned above, two patients had a renal function at
baseline characterized by a low creatinine clearance (26 and 32
mI/mm). At all of the sampling times, their total and RBC
platinum levels were 20-40% higher than the whole group.
Likewise, UC platinum levels in these patients on day I . 5 h.
were either higher or very close to those on day 1 . 2 h. Because
these two patients had a different pharmacokinetic behavior, a
separate statistical analysis compared the different courses after
their exclusion and showed no difference (data not shown).
We looked for a link between accumulation of platinum in
RBCs and anemia incidence. RBC levels were higher in patients
who experienced subsequent anemia (ii 4). The differences of
the mean levels between the two groups were significant at day
8 (P = 0.015), day 15 (P = 0.05), and day 22. 2 h (P = 0.014).
after the first administration of oxaliplatin.
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
896 Oxaliplatin in Colorectal Cancer
5000
4500
4000
0 10 20 30 40 50 60 70days
1000
500
0
A B
Fig. 4 Mean platinum concentration inRBCs throughout three courses. Platinum
levels increased rapidly after infusion and
reached a niaximal value at 2 and 5 h:afterward, they declined slowly. and, thus.
a progressive accumulation could be ob-
served.
I600
1400
1200
1000
800
600
400
200
0
C0
CC)UC0
U
EC
0.
800 600 400 200 0
total platinum concentration (�.tg/L)
Fig. 5 Correlation between RBCs and total platinum in plasma (120 samples). A. there was no correlation at early sampling times. B. the correlation
was high for late sampling times: r 0.65. P < 10 �. and the linear regression curve was 1(8)9.
�; 35000.
C0
.� 3000
CC)
� 2500U
E
.� 2000ca
0.
Ld� 1500
21)00
. 1800
. 1600.
I 400.
#{149}. . .� 1200
.
.. ... 1000. . .
.� 800
.
. . 600.
.400
� 200
4300 3800 3300 2800 2300 1800
total platinum concentration (�iWI�)
C
0
a
CC)
C0U
EC
ca
0.
DISCUSSION
Some preliminary results on early pharmacokinetics of
oxaliplatin have been reported after a single course (5). How-
ever. no data were available regarding its long-term retention.
ICPMS allowed us to study total. UC, and RBC platinum
concentrations for successive courses of chemotherapy, as had
been done for other organoplatinum drugs (6, 10-13). Eighty-
five cycles have been administered, with a median number of 5
cycles per patient. Toxicity was consistent with the data previ-
ously reported with this type of association (I. 2). Myelosup-
pression remained moderate, and mucositis resulted from 5-FU,
whereas neurotoxicity was related to oxaliplatin.
The platinum concentration curves were very similar
throughout the treatment, and the terminal half-lives remained
unchanged. The � of total platinum were constant over
eight courses. Significant residual levels of total platinum per-
sisted until the beginning of the following course and showed a
slightly increasing trend. These results, combined with a pro-
longed terminal half-life of total platinum, may cause us to fear
a significant platinum accumulation if the interval between two
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Clinical Cancer Research 897
courses is reduced to 15 days. This accumulation could increase
the incidence of neurotoxicity.
If we compare the pharmacokinetics of oxaliplatin and
cisplatin after repeated administrations, the main difference is
the absence of clear accumulation of plasma UC and total
platinum after iterative administrations of oxaliplatin, whereas a
quite significant accumulation of platinum in plasma has been
described after repeated cisplatin administration (6, 16, 17). We
have shown previously for cisplatin that there was a progressive
and significant increase in both Cri�ax and Cmin of total and UC
plasma platinum, with a doubling of these levels after three
courses (6). In contrast, the increase of total platinum residual
levels after oxaliplatin administration is much lower (50% in-
crease after seven courses), at the limit of statistical signifi-
cance. This difference between oxaliplatin and cisplatin may be
due to a different platinum binding to plasma proteins and/or to
a difference in the volume of distribution of the two drugs. We
used an ultracentrifugation technique for measuring free plati-
num concentrations in plasma, as we did previously for studying
free platinum after cisplatin administration (6). Ultracentrifuga-
tion was compared previously with other techniques and was
found comparable for reliability to equilibrium dialysis, which is
the technique of reference (8, 18-21). It appeared superior for
reproducibility, especially when measuring low free drug con-
centrations. Moreover, ultracentrifugation, as a physical
method, does not involve a membrane and does not take into
account the binding of drugs to low molecular weight proteins
or peptides, as can be the case with low-density lipoproteins
(21). Thus, this technique appeared to be particularly adapted
for the assessment of non-protein-bound platinum. A method-
ological study was carried out in our laboratory comparing
ultracentrifugation and ultrafiltration (Amicon, model 4104)
with 22 plasma samples. The results of the two techniques were
closely correlated (Pearson correlation test, r = 0.994).
UC platinum did not accumulate in plasma after iterative
courses of oxaliplatin. Cnaax as well as residual levels remained
constant, although significant amounts of UC platinum persisted
throughout the whole treatment period and until 22 days after
oxaliplatin administration (Figs. 3 and 4).
Three of four studies of cisplatin administration reported
free platinum accumulation after cisplatin was administered (6,
7, 22). In the fourth discrepant study, the platinum determination
method in plasma had a high detection limit that did not allow
for the detection of small variations in free platinum concentra-
tions (16). Two mechanisms could explain the progressive ac-
cumulation of free platinum: either a reduced ability to clear
ultrafiltrable platinum with repeated courses (22) or a close
correlation between total and free platinum, the free species
following the progressive accumulation of the total one (6).
These results could explain, in part, the differences of toxicity
between oxaliplatin and cisplatin, especially the long-term cu-
mulative nephrotoxicity of cisplatin (23-25). A quick distribu-
tion, a progressive accumulation, and a very long-term platinum
retention in deep compartments characterize cisplatin pharma-
cokinetics ( I 2, 1 3, 26, 27). This slow storage for several months
is linked with long-standing neurotoxicity and irreversible
chronic nephrotoxicity (12, 13, 17, 23, 26-30). In contrast, the
absence of oxaliplatin accumulation in plasma could explain
why oxaliplatin neurotoxicity is most often reversible upon
treatment discontinuation (4).
UC platinum levels are closely correlated with total plati-
num values after oxaliplatin administration. Two studies showed
a similar correlation between UC and total platinum after cis-
platin administration (6, 3 1), but most authors reported that the
platinum binding to proteins after cisplatin administration was
irreversible (8, 9, 29, 32, 33). It is important to stress that these
two studies were either carried out ex viva, so that the influences
of rate constants between different compartments other than the
protein binding system were eliminated or they were not sensi-
tive enough to detect platinum level modifications lower than
100 jig/liter. These studies concluded that the half-life of free
platinum was very short, with a maximum of a few hours (9, 3 1,
34-37), whereas with ICPMS, we could detect significant levels
throughout the whole cycle (6). The proportion of UC platinum
in plasma after oxaliplatin administration varied according to
the sampling time. It was approximately 15% at early sampling
times, reaching even 25-30% in some patients, and was 5%
from day 8 to day 22 after each administration. No study has
reported this phenomenon with cisplatin, for which the ratio
between UC and total platinum was constant at approximately
6% throughout the sampling period, from day I to day 22 (6, 33.
38).
It has been shown previously that for the first 2 h after
cisplatin administration, platinum could bind to several protein
species, including low molecular weight thiols such as sulfliy-
dryl groups, cysteine, methionine, and glutathione (30, 39, 40).
Later, it binds with a high affinity to high molecular weight
proteins (>25 kDa) such as albumin, globulins, and transferrin
(8, 9, 33, 38, 40). On the other hand, after carboplatin admin-
istration, platinum only had affinity with low molecular weight
thiols and did not bind with high molecular weight proteins (40).
These results and ours suggest that, after oxaliplatin adminis-
tration, platinum binds first with both high and low molecular
weight proteins and only later with high molecular weight
proteins.
In the present study, we showed that platinum underwent a
long-term retention process in RBCs. The RBC platinum uptake
was rapid. The RBC platinum levels decreased much more
slowly than those in plasma and subsequently became higher
than the plasma levels in the late samplings. Consequently, RBC
platinum levels increased markedly from one course to the
following one. The difference between cycle I and cycle 3 (an
increase of 57-107%) was highly significant for all sampling
times.
Our results show that at early times, RBC platinum be-
haved independently from UC and bound plasma platinum.
However, on days 8, 15, and 22, RBC platinum levels show a
strong correlation with total and UC platinum in plasma.
Some arguments suggest that Hb, the major protein in
RBC, binds with platinum and plays a major role in its storage.
When an increase of Hb occurred, RBC platinum levels on day
22 were higher than on day 15. The binding of platinum to Hb
might be covalent because the terminal half-life of platinum in
RBCs was similar to that of RBCs, as determined by radioactive
phosphate labeling techniques, which is about 60 days (41). In
three instances, it reached 70 days, which may suggest a see-
ondary uptake by RBCs after a first release.
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
898 Oxaliplatin in Colorectal Cancer
As for oxaliplatin, a particular accumulation in RBCs has
been previously reported for cisplatin, but not for carboplatin.
Carboplatin distribution in RBCs and its extrusion of them were
very fast so that its half-life was very short (42-44). The
concentration-time curves of RBC platinum were similar to
those of free platinum. In contrast, after cisplatin administration,
the platinum uptake in RBCs was rapid, the concentrations
reached high peaks at 2.5 h. and platinum accumulated signif-
icantly. A few studies have emphasized the importance of RBCs
in the distribution volume and the long-term retention of plati-
num (8. 9. 38, 45). Some of them reported an irreversible
RBC-platinum binding (8. 9. 38). whereas some others found a
slow release (45). Vermorken et a!. (8. 9) found that the elim-
ination of platinum from RBCs was biphasic, with a second
terminal half-life of about 30 days. They concluded that this
terminal half-life involved an increased breakdown of RBCs
resulting from cisplatin cytotoxicity (8, 9). To our knowledge.
the long-term kinetics of platinum in RBCs after repeated ad-
ministrations of cisplatin has not been studied thus far.
This important and prolonged retention of platinum in
RBCs might play a role in some toxic manifestations. The
probable interaction with Hb could be of major interest for the
study and understanding of platinum adduct formation and
liability.
Although a prospective pharmacodynamic evaluation was
not the goal of the present study, we did attempt to correlate
platinum plasma levels with neurotoxicity, because a pharma-
cokinetic-pharmacodynamic correlation has been reported for
cisplatin (17. 23. 28). Total platinum levels varied, as has been
reported with cisplatin (8, 9. 24. 38). However, the number of
values was perhaps too small in our limited cohort, and we could
not elicit any clear relationship between platinum levels in
plasma and the neurotoxicity pattern. Moreover. neurotoxicity
occurs after a 700 mg/m2 total cumulated dose of oxaliplatin,
whereas, in the present study. the mean cumulated dose of
oxaliplatin was 520 mg/rn2.
Likewise. we looked for a relationship between anemia
incidence and individual pharmacokinetic data. Patients who
experienced subsequent anemia had significantly higher RBC
platinum levels at days 8. 15, and 22 after the first administra-
tion of oxaliplatin, but once again. the groups were small.
In the present study, we showed that oxaliplatin does not
accumulate in plasma but accumulates in RBCs with a very long
half-life after repeated administration at the currently recom-
mended dose and schedule of administration. Unlike cisplatin,
which rapidly accumulates as both free and bound platinum,
Ci,,ax5 of total and UC platinum levels in plasma remain un-
changed throughout the treatment, despite significant total re-
sidual levels on day 22. These differences might explain why
oxaliplatin neurotoxicity is consistently reversible.
REFERENCES
I . Machover. D.. Diaz-Rubio. E.. de Gramont. A.. Schilf. A..
Gastiaburu, J. J., Brienza S., Itzhaki, M., Metzger, G.. N’Daw, D.,Vignoud. J.. Abad, A.. Francois. E., Gamelin, E.. Marty, M., Sastre, J.,
Seitz, J. F.. and Ychou, M. Two consecutive phase II studies of oxali-
platin (t-OHP) for treatment of patients with advanced colorectal car-
cinoma who were resistant to previous treatment with fluoropyrimi-
dines. Ann. Oncol., 7: 95-98. 1996.
2. Extra. J. M.. Espie. M.. Calvo. F.. Feriiie. C.. Mignot. L.. and Marty,
M. Phase I study of oxaliplatin in patients with advanced cancer. Cancer
Chemother. Pharmacol., 25: 299-303, 1990.
3. Louvet, C., Bleiberg. H.. Gamelin, E.. Fran#{231}ois.E.. Andre, T.. Extra.
J. M.. Nole, F., Bensmaine, M. A., Itzhaki, M., Brienza, S., andCvitkovic, E. Oxaliplatin synergystic clinical activity with 5-fluoroura-
cil in FU-resistant colorectal cancer patients is independent of FU +
folinic acid schedule. Proc. Am. Soc. Clin. Oncol., 15: 206, 1996.
4. Levi. F., Misset, J. L., Brienza, S., Adam, R., Metzger. G.. Itzhaki,
M.. Caussanel, J. P., Kustlinger. F.. Lecouturier, S.. Descorps-Declere.A.. Jasmin. C.. Bismuth, H., and Reinberg, A. A chronopharmacologicphase II clinical trial with 5-fluorouracil. folinic acid. and oxaliplatin
using an ambulatory multichannel programmable pump. Cancer, 69:
893-900, 1992.
5. Bastian, G. Report on the pharmacokinetics of oxaliplatin in patients
with normal and impaired renal function. Ann. Oncol., 5 (Suppl. 5): 126.
I 994.
6. Gamelin, E.. Allain, P., Maillart, P., Turcant, A., Delva, R.,
Lortholary. A.. and Larra, F. Long-term pharmacokinetic behaviour of
platinum after cisplatin administration. Cancer Chemother. Pharmacol..
37: 97-102, 1995.
7. Reece. P. A., Stafford, I., Russel, J., and Grantley. G. Nonlinear renal
clearance of ultrafiltrable platinum in patients treated with cis-diam-
minedichloroplatinum(II). Cancer Chemother. Pharmacol., /5: 295-
299, 1985.
8. Vermorken, J. B.. van der Vijgh. W. J. F, van der Klein. I, Hart, A.
AM., Gall, H. E., and Pinedo, H. M. Pharniacokinetics of free and totalplatinum species after short-term infusion of cisplatin. Cancer. Treat.
Rep.. 68: 505-513. 1984.
9. Vermorken, J. B., van der Vijgh. W. J. F., Klein, I.. Gall, H. E.,
Groeningen. C. J.. Hart, A. A. M.. and Pinedo, H. M. Pharmacokinetics
of free and total platinum species after rapid and prolonged infusions of
cisplatin. Clin. Pharmacol. Ther.. 39: 136-144, 1986.
10. Allain. P.. Berre, S., Mauras, Y.. and Le Bouil, A. Evaluation of
inductively coupled mass spectrometry for the determination of plati-
num in plasma. Biol. Mass Spectrom.. 21: 141-143. 1992.
1 1 . McKay. K. New techniques in the pharniacokinetic analysis of
cancer drugs. The ultratrace determination of platinum in biological
samples by inductively coupled plasma-mass spectrometry. In: P. Work-
man and M. A. Graham (eds.), Pharmacokinetics and Cancer Chemo-
therapy: Cancer Surveys. pp. 407-414. Cold Spring Harbor. NY: Cold
Spring Harbor Laboratory. 1993.
12. Tothill, P., Matheson, L. M.. Smyth. J. F.. and McKay. K. Induc-
tively coupled plasma mass spectrometry for the determination of plat-
mum in animal tissues and a comparison with atomic absorption spec-
tronietry. J. Anal. At. Spectrom., 5: 619-622. 1990.
13. Tothill, P., Matheson, L. M., Robbins, M. E. C., Hopewell. J. W.,
McKay. K.. and Smyth. J. F. Long-term retention and distribution of
platinum in animals after cisplatin administration. Cancer Ther. Control.
2: 275-287, 1992.
14. De Gramont, A., Vignoud. J.. Tournigand. C.. Louvet, C.. Varette,
C.. Raymond, E.. and Krulik, M. Oxaliplatin with high-dose folinic acid
and 5-tluorouracil 48 h infusion in pretreated metastatic colorectal
cancer. Eur. J. Cancer, 33: 214-219, 1997.
15. Ganielin, E. C., Danquechin-Dorval, E. M., Dumesnil, Y. F.,
Maillart, P. J., Goudier, M. J., Burtin, P. C., Delva, R. G., Lortholary.
A. H.. Gesta, P. H., and Lana, F. G. Relationship between 5-fluorouracildose-intensity and therapeutic response in patients with advanced cob-
rectal cancer receiving 5-FU containing infusional therapy. Cancer
(Phila.), 77: 441-451, 1996.
16. Bonetti. A., Franceschi, T., Apostoli, P.. Messori. A.. Sperotto. L.,
and Cetto, G. L. Cisplatin pharmacokinetics using a five day schedule
during repeated courses of chemotherapy in germ cell tumors. Ther.
Drug Monit., 17: 25-32. 1995.
17. Crow, W., Mauer, E., and Green, W. Relation between cisplatin
cytotoxicity and platinuni accumulation in plasma. Proc. Am. Soc. Clin.
Oncol., 28: 109. 1984.
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Clinical Cancer Research 899
18. Kurz, H., Trunk, H., and Weitz, B. Evaluation of methods to
determine protein-binding of drugs. Equilibrium dialysis. ultrafiltration,
ultracentrifugation. gel filtration. Arzneim. Forsch.. 27: 1373-1379.
1977.
19. Matsushita, Y., and Moriguchi. I. Measurement of protein bindingby ultracentrifugation. Chem. Pharm. Bull. (Tokyo), 33: 2948-2955,
1985.
20. Pacitici, G. M.. and Viani, A. Methods of determining plasma and
tissue binding of drugs. Pharmacokinetic consequences. Clin. Pharma-
cokinet., 23: 449-468. 1992.
21. Verbeeck, R. K., and Cardinal, J. A. Plasma protein binding of
salicylic acid. phenytoIn. chborpromazine. propranobol and pethidine
using equilibrium dialysis and ultracentrifugation. Arzneim. Forsch., 35.
903-906, 1985.
22. Reece. P. A.. Stafford, I.. Russel, J., and Grantley. G. Reduced
ability to clear ultrafiltrable platinum with repeated courses of cisplatin.
J. Clin. Oncol., 4: 1392-1398, 1986.
23. Desoize, B., Marechal, F., Millard, H., and Cattan, A. Correlation of
clinical pharmacokinetic parameters of cisplatin with efficacy and tox-
icity. Bionied. Pharmacother.. 45: 203-207. 1991.
24. Forastiere, A. A., Belliveau, J. F., Goren, M. P., Vogel. W. C..
Posner. M. R.. and O’Leary. G. P., Jr. Pharmacokinetic and toxicityevaluation of 5-day continuous infusion versus intermittent bolus cis-
diamminedichloroplatinum (II) in head and neck cancer patients. Cancer
Res., 48: 3869-3874, 1988.
25. Reece P. A.. Stafford I., and Russel J. Creatinine clearance as apredictor of ultrafiltrable platinum deposition in cancer patients treated
with cisplatin: relationship between peak ultrafiltrable platinum plasma
levels and nephrotoxicity. J. Clin. Oncol., 5: 304-309, 1987.
26. Litterst, C. L.. Gram, T. E.. Dechik, R. L.. Leroy, A., and Guavino,
A. M. Distribution and disposition of platinum following intravenousadministration of cisplatin to dogs. Cancer Res., 36: 2340-2344, 1976.
27. Litterst. C. L., and Schweitzer, V. G. Increased tissue deposition
and decreased excretion of platinum following administration of cispla-
tin to cisplatin-pretreated animals. Cancer Chemother. Pharmacob.. 12:
46-49. 1984.
28. Campbell. A. B.. Kalman. S. M., and Jacobs. C. Plasma platinum
levels: relationship to platinum dose and nephrotoxicity. Cancer Treat.
Rep.. 67: 169-172. 1983.
29. Himmelstein, K. J., Patton, T. F., Belt, R. J., Taylor, S., Repta. A. J..
and Sternson, L. A. Clinical kinetics of intact cisplatin and some related
species. Clin. Pharmacol. Ther.. 29: 658-664, 1981.
30. Mistry, P., Lee, C., and McBrien, D. C. H. Intracellular metabolites
of cisplatin in the rat kidney. Cancer Chemother. Pharmacol., 24: 73-79,
1989.
31. Fournier, C.. Vennin, P., and Hecquet, B. Correlation between free
platinum AUC and total platinum measurement 24 h after iv. bolus
injection of cisplatin in humans. Cancer Chemother. Pharmacol., 21:
75-77. 1988.
32. Momburg. R., Bourdeaux, M. Sarrazin, M., Roux, F., and Briand,
C. In vitro plasma binding of some second generation antitumor plati-num complexes. Eur. J. Metab. Pharmacokinet., 10: 77-83, 1985.
33. Repta. A. J., and Long, D. F. Reactions of cisplatin with human
plasma and plasma fractions. in: Cisplatin. Vol. 18. pp. 285-304. New
York: Academic Press, 1980.
34. Bajorin. D. F.. BosI. G. J.. Alcock, N. W., Niedzwiecki. D.. Gallina.
E.. and Shurgot. B. Pharmacokinetics of cis-diamminedichloroplatinum(II) after administration in hypertonic saline. Cancer Res., 46: 5969-
5972, 1986.
35. Belliveau, J. F., Posner, M. R., Ferran, L., Crabtree, G. W.,Cummings, F. J., Wiemann, M. C., O’Leary, G. P.. Griffin, H..
Phaneuf. M. A.. O’Rourke, A.. and Calabresi. P. Cisplatin administeredas a continuous 5-day infusion: plasma platinum levels and urine plat-
mum excretion. Cancer Treat. Rep., 70: 1215-1217. 1986.
36. Cano. J. P., Catalin, J., and Rigault. J. P. Pharmacokinetic study of
cis-diammineplatinuni II in children after rapid infusion. Biomedicine
(Paris). 34: 146-153, 1981.
37. Gulbo, J. J., Litterst, C. L., Maguire, P. J., Sikic. B. I., Hoth,
D. F.. and Woolley, P. V. Pharmacokinetics and protein binding of
cis-diamminedichloroplatinum (II) administered as a one-hour or as
a twenty-hour infusion. Cancer Chemother. Pharmacol., 5: 21-26.
1980.
38. Manaka, R.. and Wolf, W. Distribution of cis-platin in blood.
Chem.-BioI. Interact., 22: 353-358, 1978.
39. Dalez-Yates, P. T. The metabolites of platinum antitumor drugs and
their biological significance. In: D. C. H. McBrien and T. F. Slater(eds.). Biochemical Mechanisms of Platinum Antitumor Drugs. pp.
121-146. Oxford: IRL, 1986.
40. Tosetti, F., Rocco, M., Fulco, R. A., Chiara, S., Bruzzone, M.,
Campora, E., and Esposito, M. Serial determination of platinum, protein
content and free sulfliydryl levels in plasma of patients treated with
cisplatin or carboplatin. Anticancer Res., 8: 381-386. 1988.
41. Harris, J. W., and Kellermeyer. R. W. The Red Cell. Revised Ed.,
pp. 517-658. Cambridge: Harvard University Press, 1972.
42. Gayer, R. C., George, A. M., and Deeb, G. In vitro stability, plasma
protein binding and blood partitioning of Cl4-carboplatin. Cancer Che-
mother. Pharmacol., 20: 271-276, 1987.
43. Oguri, S., Sakakibara, T., Mase, H., Shimizu, T., Ishikawa. K..
Kimura, K., and Smyth. R. D. Clinical pharmacokinetics of carboplatin.
J. Clin. Pharmacol., 28: 208-215, 1988.
44. Van der Vijgh. W. J. F. Clinical pharmacokinetics of carboplatin.
Clin. Pharmacokinet., 21: 242-261, 1991.
45. Long. D. F.. Patton, T. F.. and Repta. A. J. Platinum levels in human
erythrocytes following intravenous administration of cisplatin: impor-
tance of erythrocytes as a distribution site for platinum species.
Biopharm. Drug Dispos.. 2: 137-146, 1981.
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
1997;3:891-899. Clin Cancer Res E Gamelin, A L Bouil, M Boisdron-Celle, et al. cancer patients.every three weeks, combined with 5-fluorouracil in colorectal Cumulative pharmacokinetic study of oxaliplatin, administered
Updated version
http://clincancerres.aacrjournals.org/content/3/6/891
Access the most recent version of this article at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://clincancerres.aacrjournals.org/content/3/6/891To request permission to re-use all or part of this article, use this link
Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from