nature publishing group
`
`ARTICLES
`
`Plasma and Tissue Pharmacokinetics of
`Epirubicin and Paclitaxel
`in Patients Receiving
`Neoadjuvant Chemotherapy for Locally
`Advanced Primary Breast Cancer
`M Hunzi A Jetteri M Warm2 E Pantke2 M Tuscher2 G Hempe13 U Jaehde4 M Untch56
`C Kurbacher27 and U Fuhrl
`
`The objective of the study was to assess individual distribution of antineoplastic drugs into the tumor Twelve
`advanced stage primary breast cancer patients with neoadjuvant epirubicin + paclitaxel chemotherapy were studied
`Plasma concentrations of epirubicin and paclitaxel were monitored for 24 h Epirubicin concentrations in subcutaneous
`and tumor tissues were measured using microdialysis up to 12 h postdose Epirubicin concentrations were described by
`a compartmental population pharmacokinetic model NONMEM Noncompartmental analysis was used for paclitaxel
`Plasma pharmacokinetics corresponded to published data Mean epirubicin exposure in the tumor and in subcutaneous
`tissue was very similar but tissue Cmax and area under the curve values reached only means 1 and 11 respectively
`
`of plasma values Epirubicin doses were significantly correlated to tumor exposure irrespective of body surface area
`tumors
`There is no specific barrier for epirubicin to reach primary breast cancer
`
`which
`
`is
`
`based
`
`on
`
`interstitial
`
`brane4
`
`patients
`
`for
`
`of
`
`from the
`the diffusion of analytes
`through a semipermeable mem
`compartment
`In 1996 an initial pilot study in cancer
`reported the
`of microdialysis
`assessing
`feasibility
`Thus far clinical
`intratumoral carboplatin concentrations5
`microdialysis has been employed for the characterization
`of carboplatin5 cisplatin6
`intratumoral drug concentrations
`dacarbacine1°
`cap ecita
`5 fluor ouraci17 methotrexate89
`bineil and melphalanu in various
`types of cancer such as
`cancer7841 malignant melanoma51012
`breast
`osteosar
`coma1213 Merkel cell tumor12 and cancer of the oral cavity6
`As a potential caveat
`for the use of microdialysis in cancer
`patients puncturing solid tumors have been suspected to
`induce metastases raising ethical concerns However clinical
`studies following fine needle biopsy of tumors reported an
`in the range of 00030005c014
`incidence of such
`events
`tumor
`of
`that puncture
`
`It
`
`at the
`
`is obvious that achieving adequate concentrations
`site of action is a prerequisite for therapeutic
`efficacy of any
`agents Therefore although
`drugs including antineoplastic
`tumor resistance at a molecular level has been recognized as a
`failure current attention is
`key component
`in therapeutic
`on local drug transport and drug delivery
`also focused
`Among various distribution processes drug transfer from the
`space is considered a
`central compartment into the interstitial
`step in clinical resistance of tumors12
`critical
`Solid tumors have
`several potential barriers to drug
`in the distribution
`of blood
`delivery such as alterations
`vessels blood flow capillary permeability interstitial
`pres
`sure and lymphatic drainage that may limit drug distribution
`from the blood vessel and provide inherent mechanisms of
`between
`resistance
`These
`factors may vary considerably
`tumors13 A method that allows the quantification of local
`in vivo is the microdialysis technique
`drug concentrations
`
`There was no evidence
`
`lesions
`
`Department of Pharmacology Clinical Pharmacology University of Cologne Cologne Germany 2Faculty of Medicine Divisions of Breast Surgery and Gyne
`cologic Oncology Department of Gynecology
`and Obstetrics University of Cologne Cologne Germany 3Institute for Pharmaceutical
`and Medical Chemistry
`University of Munster Munster Germany Institute for Clinical Pharmacy
`University of Bonn Bonn Germany 5Department
`of Gynecology
`and Obstetrics
`Klinikum Grosshadern Munich Germany 6Current address Department of Gynecology
`and Obstetrics Klinikum Berlin Buch Berlin Germany Current address
`Oncology Department of Gynecology
`and Obstetrics Medical Center Bonn Friedensplatz Bonn Germany Correspondence
`of Gynecologic
`U Fuhr uwefuhruk
`koelnde
`
`Division
`
`II
`
`Received
`
`1 September
`
`2006 accepted
`
`10 November
`
`2006 published online 14 February 2007 doi101038sjclpt6100067
`
`CLINICAL
`
`PHARMACOLOGY
`
`THERAPEUTICS
`
`I VOLUME 81 NUMBER 5 I MAY 2007
`
`659
`
`Abraxis EX2037
`Actavis LLC v Abraxis Bioscience LLC
`1PR201701101 1PR201701103 1PR201701104
`
`

`

`e 70
`c 60
`c3
`13 50
`a 40
`8 30
`
`8 10
`
`cc
`
`o
`
`o
`
`1
`
`2
`
`3
`
`1
`
`Time h
`
`6
`
`7
`
`8
`
`In vitro recovery of epirubicin In this experiment
`Figure 1
`membrane was placed for 300 min in artificial
`solution with 27 human serum albumin containing 635 figI of epirubicin
`the membrane was placed in the same medium without
`
`a microdialysis
`
`interstitial
`
`fluid Ringers
`
`thereafter
`
`epirubicin
`
`10000
`
`1000
`
`100
`
`10
`
`pgl
`
`fepirubicin
`
`Concentrationo
`
`300
`600
`900
`1200
`Time after start of infusion min
`
`1500
`
`Figure 2 Individual concentration time profiles of epirubicin in plasma after
`a 1 h infusion of 90 mgm2 protocol A black or 150 mgm2 protocol B
`dotted lines
`
`Plasma tumor and subcutaneous
`
`tissue pharmacokinetics
`
`of
`
`profiles
`
`after
`
`epirubicin
`The individual
`time vs total plasma concentration
`of epirubicin are shown in Figure 2 As expected a rapid
`the end of
`of concentrations
`occurred
`the
`decay
`infusion followed by a slower elimination period Evaluation
`in plasma was most appro
`of pharmacokinetic
`parameters
`priate using a three compartment model Figure 3 see
`Methods section The obtained tissue concentrations were
`much lower than total plasma concentrations Figure 4 In
`the three compartment model
`initially
`only the predicted
`plasma concentrations
`time profiles and pharmacokinetic
`parameters for compart
`ment no 3 were very similar to
`those measured
`in
`subcutaneous
`tissue corrected by a scaling factor
`reflecting
`binding differences of epirubicin between plasma where
`were measured and interstitial
`total concentrations
`fluid
`obtained by microdialysis Thus the subcutaneous
`concen
`trations were allocated to the existing compartment no 3
`and included into
`from the analysis of plasma concentrations
`the model Table 1 This model provided plausible results
`
`developed
`
`from
`
`concentration
`
`vs
`
`with narrow confidence intervals
`istics Table 1 Age and body mass index BMI had an
`
`for the various character
`
`VOLUME
`
`81 NUMBER 5 I MAY 2007 I wwwnaturecomkpt
`
`ARTICLES
`
`affected either
`disease
`
`the course or prognosis of
`
`the underlying
`
`is one of
`
`distribution
`
`of
`
`these
`
`of
`
`The combination of epirubicin and paclitaxel
`treatment of breast
`used in neoadjuvant
`several regimens
`cancer eg ref 15 There is no information on tissue
`two
`drugs with the exception
`reported in tumor tissue after
`
`epirubicin concentrations
`surgery16 and paclitaxel
`concentrations
`single patient17 The objective of
`this study was to assess
`into the tumor and as
`individual
`drug distribution
`site into the abdominal subcutaneous
`
`reference
`
`in ascites
`
`fluid in a
`
`a
`
`adipose
`
`tissue in breast cancer patients assuming that
`intensity which
`tumor dose
`is defined as
`the product of
`at the effect site and time of exposure5 would
`concentration
`be more predictive for clinical tumor response than plasma
`concentrations
`
`the interstitial
`
`RESULTS
`Evaluation of the microdialysis method
`In the in vitro tests a negligible adsorption of
`Epirubicin
`epirubicin to the equipment was found Already in the first
`30 min microdialysate sample an equilibrium was
`reached
`Figure 1 After transfer of the membrane into a solution
`
`without
`
`it
`
`60 min until
`took less
`than
`the
`epirubicin
`was below the quantification limit of 2 itgl
`concentration
`The in vitro recovery and delivery of epirubicin was constant
`range of 142650 tg1 and reached
`over a wide concentration
`a mean+ SD value of 528±14 The in vivo micro
`as a proof of
`dialysis calibration by retrodialysis conducted
`principle in patients 11 and 12 gave a recovery of 485 and
`567 in tumor and of 478 and 485 in subcutaneous
`confirming the in vitro results
`adipose tissue respectively
`Thus the mean in vitro recovery was used as a correction
`ie 0528 for
`all microdialysis concentrations
`from measured
`in vivo to calculate tissue concentrations
`
`factor
`
`concentrations
`
`Paclitaxel Paclitaxel showed pronounced adsorption to the
`microdialysis membrane as well as to the outlet tubes The
`amount bound to the equipment exceeded
`that recovered in
`the perfusate Therefore even with high albumin concentra
`tions in the perfusion fluid in order to bind the drug it took
`approximately 2 h to reach an equilibrium between external
`and perfusate paclicaxel concentrations with a pronounced
`variation Owing to this result we
`decided
`to exclude
`from further
`in vitro and in vivo microdialysis
`
`paclitaxel
`
`experiments
`
`Outcome of treatment
`Of the 12 patients two responded completely
`to the initial
`before surgery five responded partially and
`chemotherapy
`five had stable disease Local
`the disease after
`relapse of
`surgery which was monitored for a median postoperative
`observation period of 68 range 5575 months occurred in
`the patients including disease related death in two
`six of
`
`patients
`
`660
`
`

`

`Compadment no 1
`
`Compartment roa 3
`
`Central
`compartment
`
`total plasma
`concentrations
`
`Deep peripheral
`
`compartment
`scaled subcutaneous
`
`microdialysis
`
`concentrations
`
`Elimination
`
`Compartment no 2
`
`Compartment no 4
`
`Superficial
`
`peripheral
`
`compartment
`
`no concentrations
`measured
`
`Tumor
`
`tissue
`scaled tumor
`microdialysis
`
`concentrations
`
`Figure 3 Compartmental pharmacokinetic model used for epirubicin
`A three compartment model was fitted using plasma and subcutaneous
`between
`microdialysis concentrations To account
`for binding differences
`measured plasma and microdialysis concentrations
`a scaring factor for
`microdialysis data was included in the model After obtaining
`pharmacokinetic parameters for the three compartment model all
`parameters were fixed for estimation of the parameters for the tumor
`
`compartment
`
`9 Plasma
`1r Subcutaneous
`
`Tumor
`
`tissue
`
`tissue
`
`12
`
`10
`
`8
`
`4
`
`2
`
`§
`
`a
`
`3
`
`I
`
`300
`600
`900
`Time after start of epirublcin
`
`1200
`Infusion min
`
`1600
`
`Figure 4 Population mean concentration time profiles of epirubidn in
`tissue and tumor following a 1 h infusion of
`plasma subcutaneous
`90 mgm2 BSA dose in a patient with mean demographic data 170cm body
`height 795 kg body weight dose 1719 mg a Linear scale b log scale
`Please note the 100 fold difference in scales between plasma and tissue
`
`concentrations in a only
`
`independent
`
`significant effect on the volume of distribution
`in compartment no 2 Table 1
`Finally the incorporation of tumor concentrations
`the pharmacokinetic model was successful
`as assessed by the
`in the Methods
`section using the same
`criteria described
`
`into
`
`ARTICLES
`
`have reached
`
`the
`
`tissue
`
`scaling factor We also tried to estimate a separate scaling
`found for
`this was almost
`to the factor
`factor
`identical
`subcutaneous fluid but could not be estimated reliably The
`plots are shown in Figure 5
`goodness offit
`corresponding
`the estimation of concentration
`The procedure
`allowed
`vs time profiles in tumor beyond
`the observation period
`despite the higher random noise inherent
`to microdialysis data
`Figure 6 After the distribution processes
`in this case redistribu
`equilibrium the lowest
`rate constant
`tion from compartment no 3 to the central compartment
`Table 1 governs the decay of concentrations
`tissues
`in all
`The corresponding population estimate for the respective half
`life which may be calculated as In 2 x clearance
`1 x volume
`of distribution was 466 min or roughly 8h
`In the tumor of one patient and in the abdominal sub
`cutaneous adipose tissue of four patients epirubicin concen
`trations in the perfusate were always below the limit of
`quantification We did not take these concentrations into account
`for the development of the model see Discussion section
`The expected significant positive correlations were found
`in the correlation matrix dose was correlated to body surkce
`area BSA and related parameters such as body weight AUC
`and Cmax for respective sites that were correlated the AUC in
`by the NONMEM model was
`tumor
`calculated
`correlated to the scaling factor age and BMI were negatively
`correlated to volume of distribution in compartment no 2
`V2 as also seen
`in the evaluation
`of covariates of the
`
`analysis Table 1 etc Despite
`population pharmacokinetic
`the similar mean concentrations
`in tumor and
`tissue Figure 4 individual C
`and AUC
`subcutaneous
`tumor C and AUC0503th were
`values showed no significant correlation between the two
`
`obtained
`
`sites Interestingly
`positively correlated to the absolute dose Spearmans rank
`rs = 0752 P= 0008 and rs = 0633
`correlation coefficient
`P= 0036 and related parameters body weight BSA BMI
`and had a negative correlation to V2 rs = 0655 P= 0029
`and rs = 0582 P=0060 NS
`In Table 2 selected pharmacokinetic
`parameters
`epirubicin in tumor sc tissue and plasma
`are sorted
`to response and relapse of
`the disease Despite
`according
`mean
`there was no
`differences
`difference between groups
`para
`for any pharmacokinetic
`including those not shown
`meter in the KruskalWallis test
`in Table 2 Also none of the demographic characteristics
`
`of
`
`statistically
`
`significant
`
`the two
`treatment However
`were related to efficacy of
`tumor epirubicin concentrations
`patients with the highest
`ie patients no 1 and 3 had at least partial
`response and no
`two patients with
`whereas
`those
`the
`lowest
`ie patients no 8 and 12 showed
`concentrations
`response and also had relapse of cancer
`patient no 8 died from the disease Figure 6
`
`relapse
`
`histological
`
`no
`and
`
`Plasma pharmacokinetics
`of paclitaxel
`The individual concentration
`vs time profiles of paclitaxel
`11 patients are shown in Figure 7 plasma pharmacokinetic
`by noncompartmental methods
`obtained
`are
`
`parameters
`
`in
`
`CLINICAL
`
`PHARMACOLOGY
`
`THERAPEUTICS
`
`I VOLUME 81 NUMBER 5 I MAY 2007
`
`661
`
`

`

`ARTICLES
`
`Table 1 Pharmacokinetic
`
`parameters of epirubicin in plasma tumor and subcutaneous
`
`tissue
`
`Lower limit
`of 95 Cl
`
`Point
`
`estimate
`
`limit
`
`Upper
`of 95 Cl
`
`Interindividual
`CV
`
`Pharmacokinetic parameter of epirubicin unit
`Absolute dose mg
`Maximal concentration in compartment no 1 plasma taken from raw data µg1 n=12
`Maximal concentration in compartment no 3 taken from raw data µg1 n=8
`Ratio of maximal concentration in compartment no 3 over compartment no 1
`Maximal concentration in compartment no 4 tumor taken from raw data µg1 n=11
`Ratio of maximal concentration in compartment no 4 over compartment no 1
`AUC in compartment no 1 mgI x min
`
`n=8
`
`n=11
`
`AUC from predose to 503 min after start of infusion in compartment no 3 calculated
`noncompartmental methods mgI x min n=8
`
`by
`
`by
`
`157
`
`1108
`
`66
`
`044
`
`91
`
`081
`
`116
`
`31
`
`31
`
`188
`
`1264
`
`112
`
`097
`
`133
`
`106
`
`126
`
`45
`
`46
`
`147
`
`116
`
`219
`
`1421
`
`158
`
`149
`
`176
`
`131
`
`136
`
`59
`
`60
`
`172
`
`130
`
`139
`
`NA
`
`NA
`
`NA
`
`NA
`
`NA
`
`NA
`
`NA
`
`NA
`
`NA
`
`NA
`
`NA
`
`141
`
`numerically mgI x min
`
`122
`
`n=11
`
`101
`
`Clearance
`
`Clearance
`
`Clearance
`
`AUC from predose to 503 min after start of infusion in compartment no 4 calculated
`noncompartmental methods mgI x min n=11
`AUC extrapolated to infinity in compartment no 4 tumor calculated
`Ratio of AUC extrapolated to infinity in compartment no 4 over compartment no 1
`Elimination clearance Lmin
`for exchange between compartments no 1 and no 2 Lmin
`for exchange between compartments no 1 and no 3 Lmin
`for exchange between compartments no 1 and no 4 Lmin
`Volume of distribution in compartment no 1 plasma L
`Volume of distribution in compartment no 2 L
`Volume of distribution in compartment no 3 L
`Volume of distribution in compartment no 4 tumor L
`factor accounting for lower protein binding in tissue no unit
`
`Scaling
`
`Effect of BMI on volume of distribution in compartment no 2
`per kgm2 of BMI above mean BMI
`distribution
`
`change of volume of
`
`in compartment no 2
`Effect of age on volume of distribution
`per year of age above mean age
`distribution
`
`change of volume of
`
`127
`
`0245
`
`0799
`
`0148
`
`108
`
`349
`
`537
`
`99
`
`133
`
`0290
`
`0876
`
`0222
`
`117
`
`479
`
`573
`
`135
`
`0335
`
`0953
`
`0296
`
`126
`
`609
`
`609
`
`171
`
`00793
`
`00894
`
`00995
`
`367
`
`399
`
`299
`
`302
`
`232
`
`205
`
`NS
`
`280
`
`569
`
`NS
`
`NS
`
`NS
`
`NS
`
`329
`
`NA
`
`NA
`
`index Cl confidence interval CV coefficient of variation NA not applicable
`AUC area under the curve BMI body mass
`NS inclusion of
`this element of
`AUC and Crx values are normalized to an epirubicin dose of 90 mgm2 BSA Two patients received 150 mgm2 all other
`variation did not
`improve the model significantly
`90 mgm2 Results of compartmental
`evaluation if not
`indicated otherwise
`population
`pharmacokinetic
`
`patients received
`
`intraindividual
`
`presented in Table 3 In one patient plasma volumes were
`to measure paclitaxel concentrations Much of
`not sufficient
`vs time curves
`the apparent variability in the concentration
`was caused by the highly variable duration of infusion range
`87224 mm Tissue concentrations of paclitaxel could not be
`
`determined see above
`
`Tolerability of microdialysis
`Microdialysis experiments were well tolerated by all patients
`No patient
`required administration of local anesthetics before
`the microdialysis syringes No other adverse
`insertion of
`events occurred including hemorrhage or excessive pain at
`the site of probe insertion for both tumor and subcutaneous
`the
`could
`tissue respectively All patients
`easily tolerate
`after having received neoad
`measurement
`pharmacokinetic
`juvant chemotherapy Most importantly no local metastases
`were observed at
`the site of probe insertion during the
`observation period
`
`postoperative
`
`DISCUSSION
`
`In this study beyond plasma pharmacokinetics
`of epirubicin
`and paclitaxel we assessed drug transfer of epirubicin into
`tumors and subcutaneous
`the interstitial
`space of
`tissue in
`breast cancer patients assuming that local exposure would be
`more predictive for treatment efficacy than plasma concen
`trations
`
`were similar to
`plasma
`Epirubicin
`pharmacokinetics
`published data This applies to both the three compartment
`model and the quantitative
`results obtained here1821 It has
`been reported that paclitaxel may increase the AUC of
`epirubicin when given consecutively22 However
`as only two
`patients in our study received epirubicin alone such an effect
`could not be substantiated in the present small study
`Interestingly it was possible to allocate a compartment
`to a real site The scaling factor
`from the plasma model
`introduced in the model ie 009 Table 1 is lower
`than the
`unbound fraction of epirubicin in plasma23 However
`
`662
`
`VOLUME
`
`81 NUMBER 5 I MAY 2007 I wwwnaturecomcpt
`
`

`

`ARTICLES
`
`pet no1
`
`pat no 2
`
`pat no 3
`
`pat na 4
`
`A
`
`A
`
`pat no 6
`
`pat no 0
`
`pat no 7
`
`pat no 11
`
`pet no 9
`
`pot no10
`
`pat no11
`
`pet no 12
`
`fiie52L
`
`50
`
`25
`
`0 50
`
`
`
`2525
`
`6 E
`
`5= soso
`
`
`
`
`
`2525
`
`Tx 0
`
`a 3000
`
`2500
`
`p2000
`
`1600
`
`5 tow
`
`500
`
`0
`
`750
`
`1500
`
`750
`
`1500
`
`750
`
`1500
`
`750
`
`1500
`
`lime after start of eplrubicin Infusion mm
`
`Identity line yx
`
`Individual
`
`data points
`
`10
`
`16
`
`Figure 6 Individual concentration time profiles of epirubicin in tumor
`tissue Triangles represent the raw tumor concentration data only those
`ie 2 pgI are shown lines
`above the lower limit of quantification
`represent the individual estimates obtained by population
`pharmacoltinetics see Methods section Lines also indicate
`histological
`response dotted no change continuous partial response continuous with
`hatched areas complete response Open symbols were used for patients
`relapse of disease closed symbols for patients with no relapse
`with local
`during the period of observation patient no 6 had no concentrations
`above the limit of quantification therefore the population mean was used
`for this patient who had no relapse The dose was 90 mgm2 BSA protocol
`A for all patients except nos 2 and 3 who obtained 150 mgm2 BSA
`protocol B see the text Patients nos 2 and 8 have died from progressive
`disease 13 and 30 months after surgery respectively
`
`epirubicin
`
`Observed
`
`concentrations pin
`Figure 5 Goodness of fit plots for epirubicin concentrations in a plasma
`no 1 b subcutaneous tissue compartment
`no 3 and
`compartment
`c tumor compartment no 4 Predicted concentrations were calculated by
`the pharmacokinetic model as described in the Methods and Results
`
`sections using the population estimates for pharmacokinetic parameters
`see Table 1 The concentrations grossly deviating from the population
`estimates on the right side of c originate from patient no 3 but were
`properly estimated with individual parameters see Figure 6
`
`we used an artificial
`fluid for microdialysis both
`interstitial
`for in vitro validation and for perfusion of the system which
`concentrations
`rather than
`should provide total
`interstitial
`unbound concentrations The complex interaction of epi
`induding proteins and
`rubicin with blood components
`the indirect access microdialysis provides to
`and the differences
`in binding of
`tissue concentrations
`epirubicin at the different sites make it difficult
`to provide a
`physiological explanation for the magnitude of the observed
`to model
`It was however
`factor
`useful
`the
`vs time curves
`both
`
`erythrocytes24
`
`scaling
`
`concentration
`
`in the subcutaneous
`
`tissue and the tumor and thus to obtain information on the
`the
`time profile in the tissue beyond
`concentration
`vs
`observation period
`The individual measured epirubicin plasma concentra
`tions were not predictive of
`concentrations
`interstitial
`in
`tumor or subcutaneous adipose tissue The lack of correlation
`and derived
`between plasma and tumor concentrations
`parameters has also been described
`in
`pharmacokinetic
`and may be explained by
`other microdialysis studies878
`tumor
`and
`of
`vascular
`characteristics
`biology
`However mean
`values in tumor
`subcutaneous tissue were almost identical as was interindi
`vidual variation Table 2 Furthermore individual
`tissue
`vs time profiles showed some irregularities
`Figure 6 and could not be assessed in the tumor in one
`tissue in four patients We
`patient and in the subcutaneous
`attributed these
`findings to technical
`the
`problems of
`microdialysis method such as intermittent placement
`in
`fluid to allow more
`areas with insufficient
`interstitial
`complete exchange of concentrations Finally the method
`we used to estimate in vivo recovery is a compromise between
`and
`for
`an
`extensive
`discussion
`accuracy
`see ref 25 Thus it cannot
`be exduded
`the lack of
`that
`correlation between tissue and plasma concentrations
`is a
`problem rather than a true phenomenon
`methodological
`The same mean concentrations
`in both tissues tested suggests
`
`typical
`
`architecture78
`
`concentration
`
`feasibility
`
`and in
`
`CLINICAL
`
`PHARMACOLOGY
`
`THERAPEUTICS
`
`I VOLUME 81 NUMBER 5 I MAY 2007
`
`663
`
`

`

`ARTICLES
`
`Table 2 Parameters directly related to exposure to epirubicin in relationship to histopathological
`and to relapse of disease
`chemotherapy
`
`response to the neoadjuvant
`
`Pharmacokinetic parameter of epirubicin unit
`Absolute dose mg
`Maximal concentration in compartment no 1 plasma taken from
`raw data µg1
`
`Maximal concentration in compartment no 3 taken from
`raw data µg1
`
`Maximal concentration in compartment no 4 tumor taken from
`raw data µg1
`AUC in compartment no 1 mgI x min
`AUC from predose to 503 min after start of infusion in
`compartment no 3 calculated by noncompartmental methods
`mgI x min
`
`AUC from predose to 503 min after start of infusion in
`compartment no 4 calculated by noncompartmental methods
`mgI x min
`AUC extrapolated to infinity
`in compartment no 4 tumor
`numerically mgI x min
`calculated
`
`Response
`
`Relapse
`
`No change
`n=5
`
`190 31
`
`1247 113
`
`Partial response
`n=5
`173 4
`1259 124
`
`Complete
`
`response
`
`n=2
`
`146 294
`
`1474 3051
`
`Yes n=6
`
`182 26
`1285 99
`
`No
`n=6
`
`193 20
`
`1558 315
`
`171 50
`
`86 14
`
`95
`
`194 63
`
`91 09
`
`112 21
`
`149 34
`
`93 477
`
`110 18
`
`213 71
`
`130 21
`55 13
`
`133 8
`35 06
`
`136249
`
`37
`
`13317
`60 17
`
`151 21
`37 04
`
`38 08
`
`48 07
`
`37 180
`
`37 07
`
`75 27
`
`168 32
`
`137 26
`
`147 341
`
`161 27
`
`175 39
`
`Abbreviation AUC area under curve Pharmacokinetic parameters are not dose normalized Means SEM individual values for n <3 are shown Compartment no 3
`represents
`no 4 represents data obtained from breast cancer
`tumor microdialysis Numbers for
`individual groups of data may be lower
`subcutaneous adipose tissue data compartment
`could not be quantified in all cases see Table 1
`because
`tissue concentrations
`
`it
`
`identified
`
`Epirubicin AUC in plasma is closely related to the effect
`on white blood cell counts26
`to achieve
`In an attempt
`uniform AUC values
`the drug usually is administered
`at doses normalized by BSA as has been done in the
`has been shown
`study However
`III
`underlying phase
`that BSA is neither
`covariate
`for plasma
`a significant
`nor for neutropenia27 A more recent study
`on
`epirubicin
`pharmacokinetics
`aspartate
`aminotransferase but not BSA as a significant covariate of
`systemic epirubicin exposure2° Here we found that tumor
`exposure is related to absolute rather than to BSA normalized
`reduction in the volume
`doses probably caused
`by a
`of distribution in compartment no 2 in individuals with
`higher BMI As increasing
`age had the same independent
`effect on V2 this compartment may be related to total body
`water and it
`is conceivable that a reduction in total body
`water would result in higher concentrations
`of epirubicin at
`other sites
`
`concentrations
`
`9 8 7 6 5 4 3 2
`
`mgI
`
`fpaclitaxel
`
`Concentrationo
`
`300
`
`600
`900
`Time after start of infusion min
`
`200
`
`1
`
`1500
`
`in plasma after a 3 h
`Figure 7 Concentration time profiles of paclitaxel
`infusion of 175 mgm2 protocol A black and 250 mgm2 protocol B
`dotted lines
`
`that distribution of epirubicin into the relatively small solid
`tumors examined here is not a general
`limiting factor
`tumor exposure and thus for antineoplastic
`
`efficacy
`
`for
`
`Table 3 Main pharmacokinetic
`
`parameters of paclitaxel
`
`Pharmacokinetic parameter of paclitaxel unit
`
`Maximal concentration taken from raw data mgI
`AUC extrapolated to infinity mgI x h
`terminal elimination half life h
`Apparent
`Clearance Lhm2 BSA
`
`in plasma
`limit of 95 Cl
`
`Lower
`
`Point estimate
`
`Upper
`
`limit of 95 Cl
`
`508
`
`133
`
`62
`
`102
`
`590
`
`157
`
`82
`
`117
`
`673
`
`181
`
`102
`
`131
`
`AUC area under curve BSA body surface area Results
`
`are normalized to a dose of 175 mgm2 BSA Data are given for 11 patients no plasma available for patient no 3
`
`664
`
`VOLUME
`
`81 NUMBER 5 MAY 2007 wwwnaturecomkpt
`
`

`

`Paclitaxel
`
`in plasma were
`pharmacokinetic
`parameters
`similar to published data192829 taking different administra
`infusions into account With
`tion routes or duration of
`to microdialysis of the highly lipophilic and poorly
`respect
`have been
`watersoluble drug paclitaxe13° many difficulties
`reported for microdialysis of lipophilic drugs High protein
`binding binding to the membrane and outlet
`tubes is
`among the most important problems3132 Even the addition
`of albumin to the perfusion fluid which is recommended to
`to plastics and glass33 did
`decrease the binding of paclitaxel
`to membrane and tubes
`the binding of paclitaxel
`not prevent
`As a constant
`recovery was not attainable we were unable to
`using the micro
`tissue pharmacokinetics
`study paclitaxel
`
`expectation
`
`taking
`
`dialysis technique
`The merely descriptive
`evaluation
`of
`the relationship
`between epirubicin tumor exposure and therapeutic
`efficacy
`in this small study gave no clear results The apparent
`trend
`to higher tumor exposure in patients with better efficacy as
`opposed to the other sites examined
`to the
`corresponded
`but was not statistically
`significant However
`interindividual
`in the
`chemotherapy
`differences
`doses and paclitaxel plasma pharmacokinetics
`into
`protocol
`in addition to methodological problems see above
`account
`the power of this study to identify epirubicin tumor exposure
`as a major source of variation in therapeutic efficacy was low
`Published data suffer from the same limitation Muller et al
`reported a positive relationship between efficacy and 5fluoro
`uracil exposure in tumor but not at other sites in patients
`che
`with primary breast
`cancer
`undergoing
`neoadjuvant
`motherapy In contrast no methothrexate
`parameter was related to clinical
`responses in a microdialysis
`study conducted in nine breast cancer patients One potential
`problem is highlighted in a preclinical study with melphalan in
`between
`limb malignancies where a significant
`correlation
`tumor response and melphalan concentration in subcutaneous
`tissue but not in tumor was seen The authors discussed
`that
`in this case microdialysis probes may have been located in
`necrotic tumor
`regions which would result
`in microdialysis
`drug concentrations not representative for other
`regions of the
`tumor with living malignant cells12
`In summary this study clearly shows that
`specific barrier for epirubicin to reach primary breast cancer
`tumors It appears that total rather than BSA adjusted doses
`are related to tumor exposure and that low total body water
`increases tumor exposure These assumptions however need
`to be addressed in further studies
`
`pharmacokinetic
`
`there is no
`
`METHODS
`Twelve
`advanced stage
`with biopsy proven
`Patients
`patients
`3 cm showing no evidence of
`primary breast cancer tumor size
`and
`neoadjuvant
`overt metastatic
`disease
`thus
`qualified for
`were enrolled All patients were treated within a
`randomized controlled phase III
`for neoadjuvant
`of locally advanced or inflammatory primary breast
`Gynakologische Onko
`initiated by the Arbeitsgemeinschaft
`cancer
`logic Breast Study Group This trial
`a standard dose
`compared
`combination
`chemotherapy with
`of
`intravenous
`four
`epirubicin at 90 mgm2 and intravenous
`at 175 mgm2
`
`trial
`
`cycles
`
`paclitaxel
`
`chemotherapy
`
`multicenter
`
`chemotherapy
`
`ARTICLES
`
`3 weeks protocol A with a biweekly
`dose dense
`150 mgm2
`of epirubicin at
`at 250 mgm2 protocol B
`
`given every
`comprising three
`protocol
`cycles
`followed by three cycles of paclitaxel
`Protocol B was obligatory supported by the administration of
`Amgen Munich Germany given on days
`filgrastim Neupogencw
`3 10 after every chemotherapy course All patients with hormone
`responsive disease received tamoxifen after completing chemother
`apy Local radiotherapy was performed according to the institutional
`guidelines Patients had a mean age of 52 +6 years mean + SD and
`received protocol A
`a mean BMI of 292 + 54 kgm2 Ten patients
`two patients nos 2 and 3 received protocol B The actual mean
`± SD duration of the intravenous
`infusion of epirubicin and
`paclitaxel was 59 +12 min and 160 + 44 min respectively
`
`The study protocol which was closely synchro
`Study procedure
`nized with the underlying phase III
`trials was approved by the
`Ethics Committee of the Medical
`Faculty of
`the University of
`Cologne Germany and
`conducted
`to
`the updated
`according
`Declaration of Helsinki published online httpwwwwmanete
`policyb3htm and the Good Clinical Practice Guidelines of the
`European Commission All patients
`their written informed
`gave
`consent
`to participate in both the Arbeitsgemeinschaft
`Gynakolo
`trial and the pharmacokinetic study reported here
`gische Onkologie
`On the day of the first epirubicin and paclitaxel administration
`of epirubicin and paclitaxel were monitored
`plasma concentrations
`up to 24 h postdose Additionally
`of epirubicin
`the concentrations
`in the tumor as well as in abdominal subcutaneous
`adipose tissue
`in 60 min intervals
`were measured
`12 h postdose
`to
`up
`microdialysis Definitive surgery was performed 2 4 weeks
`cessation of protocol B and 3 5 weeks after completing protocol A
`
`by
`after
`
`respectively
`
`Assessment of tumor response Clinical assessment of response to
`was based on both tumor
`neoadjuvant
`chemotherapy
`imaging
`treatment mammography
`performed before and after
`ultrasound
`resonance tomography and histological
`examination and magnetic
`examination of the tissue removed during surgery using standard
`Eastern Cooperative Oncology Group criteria complete
`response
`was defined as complete disappearance of all measurable tumor for
`response was defined as a reduction of the
`at least 4 weeks partial
`tumor size by 50 or more measured by two perpendicular
`diameters progressive disease was defined as any increase of tumor
`size by more than 25 or occurrence of new lesions Any case other
`than CR PR or PD was
`classified
`Categorization of pathological
`required the
`complete
`response
`the time of
`microscopical absence of any invasive tumor lesion at
`definitive surgery
`
`stable disease
`
`as having
`
`Microdialysis procedure
`
`The principles of microdialysis have been
`
`described previouslyL431
`A commercially available microdialysis probe CMA 60 CMA
`Solna Sweden with a molecular cutoff of 20 kDa an outer diameter
`of 600 yin and a membrane length of 30 mm was used The probe
`was constantly perfused at a flow rate of 200 µImin by a CMA 107
`microdialysis pump CMA Solna Sweden Owing to technical
`problems when inserting the membrane into the dense breast tumor
`tissue34 we used a special
`introducer
`for
`the implantation in the
`by CMA Solna Sweden The perfusion fluid
`tumor manufactured
`was Ringers solution with human serum albumin in a final
`of 27 in order
`to mimic in vivo interstitial
`concentration
`and to prevent
`conditions3536
`fluid loss into the tissue36
`The in vivo microdialysis