PHARMACOKINETIC-PHARMACODYNAMIC RELATIONSHIPS Clin Pharmacokinet 2002; 41 (6): 431-444

0312-5963/02/0006-0431/$25.00/0

© Adis International Limited. All rights reserved.

Pharmacokinetic-Pharmacodynamic
Relationships of the Anthracycline
Anticancer Drugs
Romano Danesi,1 Stefano Fogli,1 Alessandra Gennari,2 Pierfranco Conte2 and
Mario Del Tacca1
1 Division of Pharmacology and Chemotherapy, Department of Oncology, Transplants and
Advanced Technologies in Medicine, University of Pisa, Pisa, Italy
2 Division of Medical Oncology, University Hospital, Pisa, Italy

Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431
1. Clinical Pharmacology of Anthracyclines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
1.1 Doxorubicin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436
1.2 Epirubicin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
1.3 Idarubicin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
1.4 Anthracycline Prodrugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
1.5 Anthracycline Delivery Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
1.6 New Anthracyclines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
2. Biological Basis of Pharmacokinetic-Pharmacodynamic Modelling . . . . . . . . . . . . . . . . . 439
3. Theoretical Basis of Pharmacokinetic-Pharmacodynamic Modelling . . . . . . . . . . . . . . . . . 440
4. Clinical Application of Pharmacokinetic-Pharmacodynamic Relationship . . . . . . . . . . . . . 441
5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442

Abstract The anthracycline glycoside antibiotics represent a group of potent anticancer

agents with a wide spectrum of activity against solid tumours and haematological
malignancies, and are the mainstay of a large number of clinical protocols for the
treatment of adult and childhood neoplastic diseases. Their clinical activity is
limited, however, by acute and chronic adverse effects. Myelosuppression, pre-
dominantly neutropenia and leucopenia, is the dose-limiting toxicity; in addition
to this, mucositis, nausea, vomiting and alopecia are frequent, whereas hepato-
pathy, characterised by elevated bilirubin concentrations, occurs less frequently.
Cardiotoxicity is a major adverse effect of the anthracycline antibiotics and can
be acute or chronic; in the acute setting, electrocardiographic abnormalities may
be seen, including ST-T elevations and arrhythmias, but chronic cardiotoxicity
represents a serious adverse effect that may be lethal due to the development of
irreversible, cumulative dose–dependent, congestive cardiomyopathy.
The occurrence of toxicity displays a marked interindividual variation, and
for this reason the pharmacokinetics and pharmacodynamics of anthracyclines
have been extensively investigated in order to identify integrated models that can
be used in the clinical setting to prevent the development of serious toxicity,
432 Danesi et al.

mainly leucopenia, and maximise tumour exposure. Pharmacokinetics has been

recognised to influence both the toxicity and the activity of anthracyclines; in
particular, there is increasing evidence that the mode of administration plays an
important role for cumulative cardiotoxicity and data indicate that bolus admin-
istration, rather than continuous infusion, appears to be an important risk factor
for anthracycline-induced cardiomyopathy, thus implying that this type of toxic-
ity is maximum concentration-dependent. On the contrary, exposure to the drug,
as measured by area under the curve, seems best related to the occurrence of
leucopenia. Finally, the development of pharmacokinetic-pharmacodynamic
models allows the simulation of drug effects and ultimately dose optimisation in
order to anticipate important toxicities and prevent their occurrence by the ad-
ministration of prophylactic treatments.

The importance of pharmacokinetic-pharmaco- of drug dosages and schedules to maximise thera-

dynamic (PK-PD) modelling in drug development peutic effects and minimise toxicities have been
is becoming increasingly recognised. In preclinical addressed in a number of pharmacodynamic stud-
studies, PK-PD is used to support drug discovery, ies to establish a relationship between serum con-
interpret toxicokinetic experiments and, by the use centrations and dose-limiting toxicities for anti-
of physiological modelling, to extrapolate from ani- cancer agents, including anthracyclines.[2]
mal models to humans. In the clinical programme, The anthracycline antibiotics were introduced
PK-PD is used to support dose-finding and dose- into clinical practice more than 20 years ago, and
escalation studies, and more recent applications in- since then they have been characterised by wide-
clude concentration-controlled clinical trials and spread clinical use that makes them one of the most
population PK-PD. For drugs such as anticancer successful drug categories. The anthracycline an-
agents with a very narrow therapeutic index, the tibiotics comprise a group of cytotoxic compounds
minimisation of interpatient variability in drug ex- with wide-ranging activity against human malig-
posure allows the maximisation of the benefit, nancies. They are used extensively for curative,
while keeping the risk of serious adverse effects to adjuvant and palliative therapy, both as single
an acceptable level, and this strategy is particularly agents and in combination regimens. The anthra-
important when the treatment is given with cura- cyclines produce a number of adverse effects; how-
tive intent.[1] ever, the most important of them is cardiotoxicity.
Pharmacokinetic models are currently used in Although clinically significant cardiac damage
phase I clinical trials to characterise drug distribu- may be caused by the administration of other
tion and elimination; in addition to this, special agents, including mitoxantrone, fluorouracil, al-
attention is devoted to drug effects, as described by kylating agents and amsacrine, the frequency and
a pharmacodynamic approach, since the clinical severity observed with anthracyclines justify spe-
and therapeutic value of a drug depends upon its cial attention to this problem. Strategies have been
dynamic effect. Several factors may limit the use devised to circumvent this adverse effect, includ-
of this approach for cancer chemotherapeutic ing development of less toxic analogues, alter-
agents, including poorly defined concentration- ations in scheduling, administration of cardio-
effect relationships, heterogeneous biological be- protective agents and advanced methods for
haviour of cancer and individual variability in the monitoring cardiac abnormalities.[3]
toxic effects of anticancer agents. However, the The profile of drug activity and pharmacokinetics
measurement and interpretation of drug concentra- may be further modulated by the coadministration
tions in biological fluids and the individualisation of a number of drugs, including chemotherapeutic

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
Pharmacokinetics-Pharmacodynamics of the Anthracyclines 433

agents (taxanes, alkylating agents, antimetabolites 1. Clinical Pharmacology

and monoclonal anti-CD20 and anti-HER2/neu of Anthracyclines
antibodies), multidrug-resistance reverting agents
(verapamil, cyclosporin, valspodar) and cardio- Several anthracyclines have reached the level
protective drugs, such as dexrazoxane (ICRF- of clinical development and are represented in fig-
187), that are able to affect drug disposition, me- ure 1, while table I summarises the relevant phar-
tabolism and pharmacodynamics. In particular, macokinetic and pharmacodynamic characteris-
taxanes and the excipient Cremophor EL interact tics and PK-PD relationships. Doxorubicin and
with the P-glycoprotein transport system of anthra- daunorubicin represent first generation anthracy-
cyclines at the biliary level and strongly influence clines, and epirubicin and idarubicin are second
the disposition of doxorubicin and epirubicin as generation compounds. Doxorubicin is a deriva-
well as their metabolites, with enhanced myelo- tive of daunorubicin (14-hydroxy-daunorubicin)
toxicity and stomatitis,[4,5] whereas trastuzumab, a and is an essential component of the treatment of
monoclonal antibody against the product of the aggressive lymphoma, childhood solid tumours,
HER2/neu oncogene for the treatment of breast bone and soft tissue sarcomas, as well as breast
cancer, enhances the cardiotoxic effect of anthra- cancer. Epirubicin has a spectrum of activity very
cyclines.[6] Finally, the entrapment of anthracy- similar to that of doxorubicin but with lower tox-
clines in lipid carriers (pegylated liposomal doxo- icity. On the other hand, daunorubicin has occu-
rubicin or liposomal daunorubicin) is associated pied the central position of interest in the treatment
with a substantial change in clinical activity and of acute lymphocytic leukaemia, whereas idaru-
toxicity, mostly dependent on changes in pharma- bicin is an effective agent for the management of
cokinetic profile with respect to the parent drugs.[7] acute myeloblastic leukaemia and is reportedly
Most anthracyclines, including epirubicin and more potent and less cardiotoxic than dauno-
idarubicin, have a predictable correlation between rubicin or doxorubicin. Anthracyclines with less
dose and pharmacokinetics, particularly area un- extensive clinical development include iododoxo-
der the concentration-time curve (AUC), and this rubicin[8,9] and methoxymorpholino-doxorubicin
feature facilitates dose adjustment with the adop- (nemorubicin, PNU-152243).[10] Finally, prodrugs
tion of simple algorithms. In other instances, how- of doxorubicin have also been developed, includ-
ever, more complex approaches are required to ing N-L-leucyl-doxorubicin,[11] pirarubicin (4′-O-
predict erratic pharmacokinetic behaviour. tetrahydropyranyl-doxorubicin)[12] and valrubicin
The aim of this article is to review the theoret- (N-trifluoroacetyl-doxorubicin-14-valerate, AD-
ical and clinical bases of the relationship between 32),[13,14] as well as novel delivery systems, includ-
drug distribution and effects of anthracycline anti- ing polymer conjugates[15] and liposomal carri-
tumour drugs, in order to find clinically useful ers.[16] The improved tolerability profile of
guidelines for improved drug use in cancer pa- anthracycline prodrugs and of delivery systems is
tients. A specific section on the clinical relevance a consequence of the slow release of daunorubicin
of the issues examined is included at the end of the or doxorubicin, and their PK-PD relationships are
manuscript, and the implications of the PK-PD ap- dependent on the distribution parameters of the ac-
proach to treatment optimisation are underscored. tive drugs. The ability of dexrazoxane to protect
The present paper reviews the scientific literature against anthracycline cardiotoxicity further en-
indexed in the MEDLINE database on the issues hances clinical interest in exploiting dose intensity
of PK-PD relationships, with particular focus on modifications to therapeutic advantage.[17] New
clinical pharmacology and oncology applications areas of research in relation to anthracycline anti-
in which a detailed analysis of drug distribution biotics include the introduction of novel analogues
and effects is reported. with improved DNA-binding properties, the rever-

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
434 Danesi et al.

O OH O O OH O

C CH2 OH C CH3

OH OH

CH3O O OH O CH 3O O OH O

H3C O H 3C O

NH2 NH 2
HO HO
Doxorubicin Daunorubicin

O OH O O OH O

C CH3 C CH2 OH

OH OH

H O OH CH3O O OH

O O

H3C O H3C O

HO
NH2 NH2
HO
Idarubicin Epirubicin

O OH O O OH O

C CH2 OH C CH2 OH

OH OH

CH3O O OH CH3O O OH O

O
H3C O

H3C O NH2
O

NH2 O
I
4'-lodo-doxorubicin Pirarubicin

Fig. 1. Chemical structures of selected anthracyclines for human use.

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
 Table I. Summary of pharmacokinetic-pharmacodynamic relationships of representative anthracyclines in human use

Anthracycline Cytotoxic metabolites PK-PD relationship Dose-limiting toxicity References

PK parameter PD parameter

Doxorubicin (DXR) Doxorubicinol (DXR-ol) AUCDXR Leucocytes, platelets Bone marrow 18-20

AUCDXR-ol PMN, platelets

Epirubicin (EPI) Epirubicinol (EPI-ol) AUCEPI, AUCEPI + EPI-ol Stomatitis Bone marrow 21-24

Cmax,EPI, C2,EPI, t1⁄2,EPI Stomatitis

AUCEPI, AUCEPI + EPI-ol PMN, leucocytes

© Adis International Limited. All rights reserved.

AUMCEPI, Cssp,EPI Leucocytes (AOC)

Idarubicin (IDA)a Idarubicinol (IDA-ol) AUCIDA, AUCIDA-ol PMN Bone marrow 25

Cmax,IDA, Cmax,IDA-ol PMN

Iododoxorubicin (I-DXR) Iododoxorubicinol (I-DXR-ol) AUCI-DXR PMN, platelets Bone marrow 8,9

AUCI-DXR + I-DXR-ol PMN, platelets

Pharmacokinetics-Pharmacodynamics of the Anthracyclines

MEN-10755 N/A N/A N/A Bone marrow 26

Methoxymorpholino-doxorubicin Methoxymorpholino- None None Bone marrow 10

doxorubicinol

Anthracycline prodrugs and delivery systems

Pirarubicin DXR, DXR-ol AUCDXR Leucocytes Bone marrow 12

N-L-leucyl-doxorubicin DXR, DXR-ol AUCDXR Leucocytes, PMN Bone marrow 11,27

Valrubicin (AD-32) DXR, DXR-ol AUCDXR Leucocytes Bone marrow 13,14

Pegylated liposomal doxorubicin DXR-ol Cmax,DXR Stomatitis, leucocyte nadir Stomatitis 28

t1⁄2,DXR Palmar-plantar Skin toxicity

erythrodysaesthesia

a Oral treatment.
AOC = area of neutrophil-time curve (AOC) below 4000 cells/μl; AUC = area under the plasma-concentration-time curve; AUMC = first moment AUC; Cmax = peak plasma concentration;
C2 = plasma concentration at 2h post-dose; Cssp = plasma concentration at steady state; N/A = not available; None = no PK-PD demonstrated; PK-PD = pharmacokinetic-pharma-
codynamic relationship; PMN = polymorphonuclear neutrophils; t1⁄2 = terminal half-life.

Clin Pharmacokinet 2002; 41 (6)

435
436 Danesi et al.

sal of multidrug resistance and the rational devel- U/L, only 25% of the planned dose should be ad-
opment of drug combinations. ministered; if serum bilirubin concentrations ex-
ceed 85.5 μmol/L, doxorubicin, epirubicin and
1.1 Doxorubicin daunorubicin are contraindicated.[29] In patients
with mild hepatic function abnormalities and se-
Doxorubicin displays linear pharmacokinetics rum bilirubin less than 25.65 μmol/L and AST con-
after intravenous administration, is widely distrib- centrations less than 60 U/L, drug doses should not
uted in the plasma and tissues, with a volume of be reduced, because there is some evidence that
distribution (Vd) exceeding 500 L/m2, and has a lower peak plasma concentrations (Cmax) may
plasma protein binding ranging from 50 to 85%. yield a shorter duration of response and sur-
The drug is extensively metabolised in the liver by vival.[29,32]
aldo-keto reductase, to yield the dihydrodiol deriv- Investigation of PK-PD correlation, by linear or
ative doxorubicinol, which retains antitumour ac- sigmoidal maximum-effect (Emax) modelling, re-
tivity, and by the NADPH-dependent cytochrome vealed significant relationships between the AUC
P450 (CYP) reductase to cleave the glycosidic of doxorubicin and leucocyte (r = –0.69, p = 0.006)
bond and release aglycone metabolites that are in- and platelet (r = –0.71, p = 0.004) counts,[18] as well
volved in the development of cardiotoxicity.
as between the AUC of doxorubicinol and the de-
Doxorubicin undergoes a triphasic plasma decay,
crease in neutrophils and platelets (p < 0.02),[19]
with an initial half life (t1⁄2α) of 4.8 min, a t1⁄2β of 2.6
respectively. In patients with small cell lung cancer
hours and a terminal t1⁄2γ of about 48 hours.[29] Renal
given combination treatment with cyclophosph-
clearance of doxorubicin is low, and about 12% of
amide and vincristine, a significant relationship
total dose is recovered in the urine during 6 days
was found between the AUC of doxorubicin and
after treatment; however, hepatic clearance is high,
leucocyte counts (table I), whereas no correlations
with more than 50% of the drug excreted in the bile
were noted between the AUC of doxorubicinol and
within 7 days after treatment. Therefore, biliary
elimination and faecal excretion of the parent com- thrombocytopenia or neutropenia.[20]
pound and its metabolites are of major clinical rel- Doxorubicin-induced cardiomyopathy may be
evance. About 10 to 20% and 40 to 50% of the dose significantly reduced by prolongation of intrave-
is excreted in faeces within 24 and 150 hours, re- nous infusion. A clinical study in children treated
spectively. About 50% of the biliary eliminated with doxorubicin 60 mg/m2 over 24 hours and 75
dose is unchanged drug, 23% is doxorubicinol and mg/m2 over 72 hours showed a reduced incidence
the remainder consists of other metabolites.[29] of congestive heart failure (1/97 patients) com-
Due to the important role of hepatic metabolism pared with a conventional bolus schedule (13% of
and biliary excretion in the clinical pharmacokinet- patients with congestive heart failure). No substan-
ics of anthracyclines, patients with liver dysfunc- tial differences were noted in the tumour response
tion, as evidenced by hyperbilirubinaemia and/or between continuous infusion and bolus administra-
elevated liver enzymes, may show significant ab- tion.[33] This evidence indicates that Cmax, rather
normalities in drug distribution parameters.[30,31] than AUC, is related to the development of chronic
A widely agreed clinical practice, although cardiotoxicity. However, a prolonged infusion
based on empirical considerations, recommends time increases the incidence and severity of
the administration of 50% of the planned dose of mucositis and bone marrow suppression, which
epirubicin, doxorubicin or daunorubicin if serum then become the dose-limiting toxicities.[29] The
total bilirubin and AST (SGOT) concentrations plethora of studies under way reflects the current
range from 25.65 to 51.3 μmol/L and 60 to 180 dilemma on the optimal dose and schedule of an-
U/L, respectively. If serum bilirubin ranges from thracyclines, and it has not yet been precisely es-
53.01 to 85.5 μmol/L and AST levels exceed 180 tablished which infusion schedule may result in op-

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
Pharmacokinetics-Pharmacodynamics of the Anthracyclines 437

timal effectiveness of therapy. There is some evi- ides and epirubicinol, and 35% undergoes biliary
dence that the volume of distribution at steady state elimination mainly as metabolites, with a renal
(Vss) is higher after prolonged administration than clearance of epirubicin of only 7.39 ± 1.87 L/h/m2.
after conventional short intravenous injection. A clinical study of the administration of
Thus, it has been suggested that the prolongation epirubicin to patients with metastatic breast cancer
of infusion duration may enhance the tissue uptake revealed no correlations of pharmacokinetic pa-
of the drug, including tumour tissue.[29] Finally, rameters with response to therapy or nausea/vomi-
high Cmax of doxorubicin and doxorubicinol may ting. On the contrary, the AUC of epirubicin and
be associated with longer remission phases in pa- epirubicinol, the Cmax, the concentration at 2 hours
tients with acute non-lymphocytic leukaemia.[32] after administration and t1⁄2γ of epirubicin, corre-
lated with the occurrence of stomatitis.[21] PK-PD
1.2 Epirubicin correlations were observed between (see table I):
(i) epirubicin AUC and neutrophil nadir; [22] (ii)
Epirubicin is a semisynthetic derivative of first moment AUC (AUMC) and steady-state plasma
doxorubicin that has been extensively evaluated in concentration (Cssp) and the area of the neutrophil-
patients with breast cancer. It is effective in the time curve (AOC) below 4 000 cells/μl;[23] and (iii)
management of metastatic disease and as adjuvant the logarithmic ratio of nadir/initial leucocyte
therapy in patients with early breast cancer.[34] In count and AUC of epirubicin alone as well as of
the adjuvant setting, epirubicin-based therapy ap- epirubicin plus epirubicinol.[24] However, a study
pears to have effectiveness at least equivalent to in patients with advanced malignancies treated
that of cyclophosphamide, methotrexate and fluo- with epirubicin 90 to 150 mg/m2 failed to demon-
rouracil (CMF); recent trials, predominantly in strate a relationship between Cmax or AUC of epiru-
premenopausal patients, report significant in- bicin and the severity of haematological and non-
crease in relapse-free and overall survivals for haematological toxicity. Furthermore, no difference
epirubicin-based treatments as compared with was noted between patients receiving bolus injec-
CMF.[35] In patients with metastatic disease, tion and 6-hour drug infusion, suggesting that Cmax
epirubicin- and doxorubicin-containing regimens, does not influence the occurrence of toxicity.[38]
in combination with cyclophosphamide and fluo-
rouracil, are therapeutically equivalent.[35] The 1.3 Idarubicin
major adverse effects of epirubicin are acute dose-
limiting myelotoxicity and cumulative dose- Idarubicin, a 4-demethoxy-anthracycline ana-
related cardiotoxicity. Other important adverse ef- logue of daunorubicin, is an effective treatment for
fects include mucositis, nausea and vomiting, acute myelogenous leukaemia and has also shown
reversible alopecia and local cutaneous reactions; activity in the treatment of advanced breast cancer,
however, the tolerability of epirubicin is better multiple myeloma and non-Hodgkin’s lymphoma.
than that of doxorubicin at equimolar doses[35] and The absence of a methoxyl group at position 4 of
the cardiotoxicity profile allows higher doses to be idarubicin results in increased lipophilicity and
administered with acceptable cardiac damage.[36] cellular uptake compared with the related com-
Compared with doxorubicin, epirubicin pharma- pound daunorubicin. Idarubicin is more cytotoxic
cokinetics are characterised by higher total body than doxorubicin in vitro and its major metabolite,
clearance (CL) [mean 74.4 vs 18 to 30 L/h/m2], idarubicinol, demonstrates similar activity to
shorter terminal half-life (mean 16.01 vs 20 to 70 idarubicin. The drug may be administered intrave-
hours) and larger Vss (mean 912 vs 700 L/m2), al- nously or orally (bioavailability 30%); the dose-
beit highly variable among patients.[29,37] A small limiting toxicity is leucopenia, while the incidence
fraction of epirubicin total dose (6 to 7%) is recov- of cardiotoxicity is lower than with doxoru-
ered in the urine as such, less than 5% as glucuron- bicin.[29]

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
438 Danesi et al.

A triphasic plasma decay of idarubicin was ob- established between the AUC of doxorubicin and
served after intravenous administration, with a both white blood cell and neutrophil counts (table
t1⁄2γ ranging from 6 to 35 hours, a large volume of I), whereas other adverse effects (thrombocyto-
distribution (Vd) exceeding 1 800 L/m2 and a CL penia or stomatitis) did not correlate with pharma-
of 30 to 122 L/h/m2.[39] After oral administration cokinetic parameters.[27] These findings suggest that
of 2 to 10 mg/m2/day, significant correlations were the rate of metabolism of N-L-leucyl-doxorubicin
shown between absolute neutrophil count at nadir into doxorubicin is the main factor affecting the
and idarubicin AUC (r = –0.33, p = 0.022) and Cmax occurrence and severity of adverse events.[11]
(r = –0.38, p = 0.0067), and idarubicinol AUC (r = The PK-PD relationships of pirarubicin and
–0.43, p = 0.0009) and Cmax (r = –0.41, p = 0.0016), valrubicin are summarised in table I.
respectively (table I).[25]
1.5 Anthracycline Delivery Systems
1.4 Anthracycline Prodrugs
The administration of 20 to 320 mg/m2 of N-(2-
The clinical pharmacology of doxorubicin pro- hydroxypropyl) methacrylamide (HPMA) copoly-
drugs may be summarised by taking into account mer-bound doxorubicin (PK1) results in the slow
three representative compounds: N-L-leucyl- release of free doxorubicin, with concentrations
doxorubicin, pirarubicin and valrubicin (AD-32). approximately 1000 times lower than unbound
Their pharmacokinetics have been evaluated with doxorubicin.[15] The median distribution and elimin-
respect to the release of the active drug doxo- ation half-lives of the unbound doxorubicin were
rubicin, which dictates the pharmacodynamic pro- estimated to be 2.7 and 49 hours, respectively, and
file of these agents. CL was as low as 0.194 L/h. The apparent CL of
N-L-Leucyl-doxorubicin has been administered free doxorubicin was 180 L/h, with distribution
from 30 to 240 mg/m2; the drug is characterised by and elimination half-lives of 0.13 and 85 hours,
linear pharmacokinetics, with a CL of 41.3 ± 25.7 respectively.[15]
L/h/m2.[11] N-L-Leucyl-doxorubicin is extensively Doxorubicin in polyethyleneglycol (PEG)-
metabolised into doxorubicin, which appears in coated liposomes displays a dose- and schedule-
plasma immediately after infusion of N-L-leucyl- dependent toxicity profile that is correlated with
doxorubicin; the AUC values of N-L-leucyl- pharmacokinetic parameters. Stomatitis is dose-
doxorubicin, doxorubicin and doxorubicinol in- related, with higher incidence and severity at 60 to
crease linearly with drug dose. If compared with 70 mg/m2. Palmar-plantar erythrodysaesthesia
the equimolar amount of the parent drug, the AUC usually develops after more than two courses of
of doxorubicin generated by N-L-leucyl-doxo- treatment and is schedule-dependent, with shorter
rubicin is approximately 23% of the AUC after ad- administration intervals leading to increased fre-
ministration of doxorubicin. The prodrug charac- quency and severity. Leucopenia and neutropenia
teristics of N-L-leucyl-doxorubicin also affect the are dose-dependent, but mild and uncomplicated in
rate of drug metabolism, since after administration most cases. Despite cumulative doses of up to
of N-L-leucyl-doxorubicin the AUC of doxo- 1500 mg/m2, cardiac toxicity is uncommon, al-
rubicinol is 73% of the AUC of doxorubicin, as though more frequent in patients pretreated with
compared with 35% of the AUC of doxorubicin mitoxantrone and radiotherapy. The pharmacokine-
after administration of doxorubicin itself. As ex- tics of doxorubicin released from PEG-liposomes
pected, the toxicity profile of N-L-leucyl-doxo- are described by a monoexponential elimination
rubicin is similar to that of doxorubicin, with curve with a long half-life (median 79 hours),
myelosuppression (leucopenia and thrombocyto- low CL (median 0.04 L/h) and a small Vd (median
penia) and mucositis being the most serious ad- 3.9 L). Cmax and AUC increased linearly with the dose;
verse events. A significant relationship could be stomatitis and leucocyte nadir are correlated with

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
Pharmacokinetics-Pharmacodynamics of the Anthracyclines 439

dose and Cmax, less with AUC, whereas palmar- replication and transcription, as well as induction
plantar erythrodysaesthesia grade correlated sig- of DNA fragmentation. Anthracyclines exert addi-
nificantly with half-life (table I).[28] tional cytotoxic effects mediated by inhibition of
cytochrome c oxidase activity, free radical forma-
1.6 New Anthracyclines tion and lipid peroxidation, producing direct mem-
Third generation anthracyclines, with highly brane effects, and chelation of iron and generation
potent in vitro cytotoxic activity, have been eva- of reactive oxygen species, resulting in oxidative
luated clinically, including the disaccharide stress.[40,41] Recent studies provide evidence of a
doxorubicin analogue MEN-10755 and methoxy- caspase-dependent pathway in anthracycline-
morpholino-doxorubicin. MEN-10755, given in- induced apoptosis in leukaemic cells treated with
travenously weekly for 3 weeks at 15 to 45 mg/m2, idarubicin and daunorubicin,[42] in normal lym-
displayed a correlation between dose and AUC, a phocytes[43] and in solid tumour cells following
mean AUC of 6 mg • h/L at 30 mg/m2, a CL of 5.6 treatment with doxorubicin.[44]
L/h/m2 and a half-life of 15.3 hours. The maximal Experimental studies have provided evidence
tolerated dose (MTD) was 45 mg/m2 and dose es- that mitochondria are the main site of cardiac in-
calation was prevented by neutropenia; however, jury and that oxidative stress is critically involved
no PK-PD relationships were reported.[26] in doxorubicin-induced heart injury, since transge-
Methoxymorpholino-doxorubicin, given orally nic mice overexpressing cardiac-specific metallo-
at doses from 59 to 940 μg/m2 once every 4 weeks thionein are highly resistant to acute cardiotoxicity
to patients with solid tumours, produced neu- induced by doxorubicin.[45] Clinically, a signifi-
tropenia as the dose-limiting toxicity at the MTD cant protection against cardiac damage by anthra-
of 940 μg/m2. Unexpectedly severe neutropenia cyclines may be obtained with the iron-chelating
and lack of relationship with plasma concentra- agent dexrazoxane.[46]
tions of methoxymorpholino-doxorubicin and its Two different parameters have been adopted to
C-13 dihydro metabolite were observed, suggest- assess the myelotoxic effect of a drug: the nadir,
ing the intracellular formation of potent cytotoxic the measure of the maximum level of neutrophil
metabolites only present in human plasma in trace fall with respect to the pretreatment value, and the
amounts.[10] These findings underscore the com- area over the curve (AOC), obtained by plotting
plexity of the pharmacology of newer anthracy- the neutrophil count as a function of time and cal-
clines with highly enhanced cytotoxic activity, and culating the AOC below 4 000 cells/μl.[23] The ad-
the need for careful preclinical investigation be- ministration of anthracyclines affects the viability
fore their use in the clinical setting. of haematopoietic progenitor cells and impairs
their ability to maintain a normal output of differ-
2. Biological Basis of entiated, nonproliferating cells, whereas the via-
Pharmacokinetic-Pharmacodynamic bility and lifespan of leucocytes in peripheral
Modelling blood is unlikely to be reduced by treatment. These
Anthracyclines are known for their complex observations have relevance to a better definition
cytotoxic mechanism, involving intercalation be- of the concept of surviving fraction applied to pe-
tween DNA base pairs by their positively charged ripheral blood cell counts as an estimate of the se-
amino sugar moieties, which interact with the nega- verity of myelotoxicity.[47] The interval between
tively charged phosphate bridges of DNA. As a the administration of the anthracycline dose and
consequence, anthracyclines interfere with DNA the reduction of neutrophils below baseline values
strand separation and inhibit helicase, DNA is approximately 10 days, and a further 4 to 6 days
topoisomerase II and DNA and RNA polymerase are required to reach the nadir, for a total time lag
activities, which results in the inhibition of DNA of 2 weeks.[23] Since the maturation of a committed

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
440 Danesi et al.

granulocyte stem cell to a mature neutrophil takes ated and highly cytotoxic metabolites of third-
about 10 days and, once in the systemic circulation, generation anthracyclines, including methoxy-
granulocytes remain viable for about 8 to 12 morpholino analogues, will require further
hours,[48] it is clear that the definition of surviving consideration of the biological characteristics un-
fraction applied to peripheral blood cells might sat- derlying their pharmacodynamics in order to bet-
isfy some PK-PD applications but does not reflect ter define the PK-PD relationships that will emerge
the dynamics of the cytotoxic effect of anthracy- from clinical studies on activity and toxicity. An
clines on blood cells, and new investigations for integrated view of factors involved in PK-PD rela-
improved PK-PD modelling are required. tionships of anthracyclines is reported in figure 2.
The development of a life-threatening form of
cardiomyopathy following treatment with anthra- 3. Theoretical Basis of
cyclines has also been the object of intensive in- Pharmacokinetic-Pharmacodynamic
vestigation. Human myocardial tissue generates Modelling
doxorubicinol as well as doxorubicin deoxya-
glycone and doxorubicinol hydroxyaglycone, re- Under pharmacokinetic steady-state conditions,
flecting the ability of myocardial tissue to reduce concentration-effect relationships can be described
the side chain carbonyl group, as well as to perform by several pharmacodynamic models, which can
reductase- and hydrolase-type deglycosylation of be classified as fixed effect, linear, log-linear,
doxorubicin followed by carbonyl reduction, re- maximum effect (Emax) and sigmoid Emax mod-
spectively.[49] Commonly accepted hypotheses els.[51] Under non–steady-state conditions, how-
have assigned a pivotal role to iron, which would ever, more complex PK-PD models are required to
act as a catalyst for free radical reactions and oxi- account for a temporal dissociation between drug
dative stress. However, the role of free radical– disposition and pharmacological effect.
based mechanisms in long-term effects has been The anthracyclines distribute into the bone mar-
challenged. More recently, studies on the role of row and elicit a response (decrease in cell count of
C-13 hydroxy metabolites of anthracyclines have peripheral blood) that is delayed with respect to the
provided new perspectives on the role of iron in the pharmacokinetic phase; the interval before the oc-
cardiotoxicity of these drugs, and have shown that currence of toxicity is mainly dependent on the ef-
the life-threatening chronic toxicity should be at- fect of anthracyclines on the kinetics of cellular
tributed to alterations of iron homeostasis by turnover within the bone marrow, while the gener-
doxorubicinol.[49,50] Finally, intracellularly gener- ation of active metabolites and drug tissue distri-

Drug-specific factors Patient-specific factors

Schedule of administration General conditions

PK
(bolus vs prolonged infusion) PD (pretreatment PS and
organ function - bone marrow)

Intrinsic activity PK-PD

PD
(high potency - MMX-DXR vs lower potency - epirubicin) relationship

Metabolism/elimination
Drug chemistry PK (age, nutritional status,
PK
(prodrug vs active drug vs lipophilic drug) AST, bilirubin)

Fig. 2. Diagrammatic representation of factors affecting the pharmacokinetic-pharmacodynamic (PK-PD) profile of anthracyclines.
AST = aspartate aminotransferase; MMX-DXR = methoxymorpholino-doxorubicin; PS = performance status.

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
Pharmacokinetics-Pharmacodynamics of the Anthracyclines 441

bution are factors marginally related to the time lag Input

between administration and event. Assuming that kel kcc0
anthracyclines exert their effects on granulocyte
Peripheral k21 k13 Peripheral
precursors in the bone marrow, the most predictive shallow
Central deep
simulation of activity is the irreversible cell killing (Vps) 2 k12 (Vc) 1 k31 (Vpd) 3
pharmacodynamic model.[52-54] Important para- kicc
k1ec
meters in this model are: (i) production rate of ma- tlag kec1 −
Baseline
ture cells (i.e. neutrophils and platelets), a zero- Effect compartment Anthracyclines
E50
order process; and (ii) turnover rate of mature Progenitor +
cells, a first-order process. A graphic represent- (Vec) cells
Emax
t50/PK50 kck
ation of the PK-PD model is shown in figure 3. The
pharmacodynamic effect of cytotoxic agents has
been previously described using a counter-clock-
Fig. 3. Graphic representation of the pharmacokinetic-pharmaco-
wise hysteresis loop to correlate the observed phar- dynamic model of anthracyclines. Input = route of introduction
macological event with plasma concentrations un- of the drug in the body; central (Vc), peripheral shallow (Vps),
der conditions of a time delay between distribution peripheral deep (Vpd) and effect (Vec) compartments = body
compartments of drug distribution; kcc0 = rate constant of cell
and response. Moreover, the pharmacodynamic output from the central compartment (physiological turnover);
parameter survival fraction (SF), the ratio of the kel = rate constant of drug output from the central compartment;
number of cells of the affected tissue surviving af- k12/k21 and k13/k31 = intercompartmental rate constants; k1ec/kec1
= rate constants between the central and the effect compart-
ter exposure to a cytotoxic drug (e.g. doxorubicin) ments; kicc = rate constant of input of mature cells into the central
versus the number of cells before exposure, has compartment; kck = rate constant of progenitor cell kill by a cyto-
been used to describe the effect of doxorubicin on toxic agent (anthracycline); Progenitor cell = committed stem
cell to mature granulocyte or platelet; tlag = time interval between
mature circulating blood cells (neutrophils and drug administration and effect; Baseline, E50 and Emax = base-
platelets), and is defined as (equation 1): line cell count, 50% effect and maximal effect; t50/PK50 = time
interval to 50% reduction in cell count/pharmacokinetic param-
eter associated with 50% reduction in cell count. The positive
SF = e–kCt
and negative modulation of kck and kicc by anthracyclines is
shown in the graph.
where k is a constant determining the slope of the
dose-response curve, C is the drug concentration
and t is the time of exposure.[47] This relationship agent [anthracycline] (kck) and/or decrease in rate
directly relates the cellular effect of anthracycline constant of input of mature cells into the central
to the AUC, since the product C × t is equivalent compartment (kicc). The future challenge will be,
to AUC. This relationship, however, does not sat- therefore, the development of novel algorithms
isfy the underlying biology of cytotoxic agents and that incorporate parameters related to the cellular
does not explain the lack of relationship between biological factors sensitive to anthracyclines.
drug exposure and pharmacodynamics observed in
some studies. The counter-clockwise hysteresis 4. Clinical Application of
loop is suitable for describing effects that arise dur- Pharmacokinetic-Pharmacodynamic
ing drug exposure and reaches the maximum effect Relationship
while the agent is present in the plasma or tis-
sue;[47] however, bone marrow toxicity, as well as Anthracyclines are characterised by major
other toxicities involving cellular damage by inter- haematological and non-haematological toxicities
action with vital macromolecules (e.g. DNA) in- that call for careful administration of chemother-
duced by anthracyclines, appears when the drug is apy and represent the limiting factors that prevent
no longer detectable and it depends on increase in dose escalation or even the delivery of the standard
rate constant of progenitor cell kill by a cytotoxic drug dose. Moreover, the late development of toxi-

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
442 Danesi et al.

city, as a result of unrecognised excessive drug ex- lation was reported to be highly significant (r =
posure due to one of the factors listed below, is a –0.59, p < 0.0001).
cause of reduction of dose intensity with a lower In the case of the doxorubicin prodrug
likelihood of clinical success. A number of factors pirarubicin, its AUC in whole blood from time zero
are responsible for the wide interpatient variation to the end of infusion in patients given cumulative
of tolerability and activity of anthracyclines, and doses up to 90 mg may be calculated as follows
they can be summarised as follows: (equation 5):
• intrinsic variability in drug disposition, includ-
AUC = (Cinf × tinf)/2
ing Vss, CL and AUC
• abnormalities in excretory organ function, par- where Cinf is the concentration of pirarubicin in
ticularly liver whole blood at the end of infusion and tinf is the
• intrinsic variability in bone marrow suscepti- length of infusion. Once the AUC is defined, a for-
bility to anthracycline cytotoxicity. mula to predict leucocyte count at the nadir (ap-
In order to prevent the development of serious proximately 12 days after treatment), on the basis
adverse effects, therapeutic drug monitoring of an- of a single sample of blood obtained at the end of
thracyclines should be performed, on the basis of drug administration, may be established (equation
an established relationship between toxicity, 6):
mainly haematological, and pharmacokinetic para-
meters, particularly systemic exposure (AUC), that leucocytenadir = 0.032404 × age + 2.005 +
leucocyteinitial × e(–0.009316 × AUC + 4.202265)
can be obtained, in the case of epirubicin, with the
following formula (equation 2): with leucocyte count expressed as 10 3 cells/μl and
AUC = 9.44 × C2 + 62.5 × C24 + 157.7 the AUC of pirarubicin from time zero to the end
of infusion as μg • h/L. Bias and precision of this
where C2 and C24 are the plasma concentrations model are 0.001 and 25.8%, respectively, and r =
measured 2 and 24 hours after administration 0.809. The prediction of the extent of leucopenia
(μg/L); this relationship is highly significant (r = allows the anticipation of serious haematological
0.953, p < 0.05).[55] In order to reach a target AUC, toxicity and the decision whether or not to start
the dose may be further refined in the individual prophylactic treatment with haematopoietic
patient with normal bilirubin concentrations by in- growth factors.[12]
troducing an estimate of liver integrity as assessed
by AST, as follows (equation 3):[29] 5. Conclusions
dose (mg/m2) = target AUC × (97.5 – 34.2 × log10 AST) A clear PK-PD correlation has been demon-
strated for some anticancer drugs, including an-
since AST levels (U/L) are related to the clearance thracyclines, and this relationship provides a back-
of anthracyclines, particularly doxorubicin. Fur- ground against which rational dose optimisation
thermore, strategies have been devised in order to can be implemented for individual patients.[1] Un-
evaluate the predictability of changes in leucocyte fortunately, PK-PD modelling has played a rather
counts on the basis of plasma concentrations of limited role in the clinical use of anthracycline anti-
anthracyclines. One approach is applied to epiru- cancer drugs. Methodological issues, limited inter-
bicin (equation 4): actions between the clinical pharmacologist and
log(leucocytenadir) = log(leucocyteinitial) – 0.0073 × C6 – 0.14 the oncologist and the need to move novel active
schedules rapidly to phase III of clinical develop-
where C6 is a single measurement of drug concen- ment have restricted the opportunity for clinical
tration (in μg/L) at 6 hours after treatment. Units pharmacology analysis in phase I/II clinical trials.
of leucocyte count are 109/L.[24] Again, this corre- Another cause of suboptimal application of phar-

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
Pharmacokinetics-Pharmacodynamics of the Anthracyclines 443

macokinetic studies is that therapeutic drug moni- tients with solid tumors: the relation between pharmaco-
kinetic property and toxicity. Cancer 2001; 91 (9): 1826-33
toring would not be easy to perform, due to the 8. Twelves CJ, Dobbs NA, Lawrence MA, et al. Iododoxorubicin
large number of blood samples required, unless in advanced breast cancer: a phase II evaluation of clinical
activity, pharmacology and quality of life. Br J Cancer 1994;
adaptive dosage adjustments with feedback con- 69 (4): 726-31
trol are applied with a limited sampling strategy, 9. Merlini G, Anesi E, Garini P, et al. Treatment of AL amyloido-
sis with 4’-iodo-4’-deoxydoxorubicin: an update. Blood
thus allowing a feedback adjustment of dosage fol- 1999; 93 (3): 1112-3
lowing measurement of plasma drug concentration 10. Sessa C, Zucchetti M, Ghielmini M, et al. Phase I clinical and
and comparison with population pharmacokinetic pharmacological study of oral methoxymorpholinyldoxo-
rubicin (PNU 152243). Cancer Chemother Pharmacol 1999;
values.[1] The application of PK-PD procedures to 44 (5): 403-10
the clinical setting is warranted in order to reduce 11. Canal P, Robert J, Ramon M, et al. Human pharmacokinetics
of N-L-leucyl-doxorubicin, a new anthracycline derivative,
toxicity and to give the full chemotherapy dose and its correlation with clinical toxicities. Clin Pharmacol
under safe conditions. Finally, the PK-PD ap- Ther 1992; 51 (3): 249-59
12. Leca FR, Marchiset-Leca D, Galeani A, et al. Pharmacokinetic-
proach is a theoretical strategy that has not, until pharmacodynamic relationships between pirarubicin expo-
now, been used prospectively in clinical oncology. sure and hematotoxicity: clinical application using only one
However, the availability of simple PK-PD algo- blood sample. Anticancer Drugs 1998; 9 (6): 503-9
13. Onrust SV, Lamb HM. Valrubicin. Drugs Aging 1999; 15 (1):
rithms to be applied in the clinical setting is poten- 69-75
tially of enormous, although still underestimated, 14. Steinberg G, Bahnson R, Brosman S, et al. Efficacy and safety
of valrubicin for the treatment of Bacillus Calmette-Guerin
clinical value. Collaboration with a clinical phar- refractory carcinoma in situ of the bladder. The Valrubicin
macology service for drug monitoring will help Study Group. J Urol 2000; 163 (3): 761-7
15. Thomson AH, Vasey PA, Murray LS, et al. Population pharmaco-
clinicians to deliver drugs in safer conditions in kinetics in phase I drug development: a phase I study of PK1
order to achieve controlled neutropenia and maxi- in patients with solid tumours. Br J Cancer 1999; 81 (1):
99-107
mise the dose to be administered. 16. Harashima H, Iida S, Urakami Y, et al. Optimization of antitu-
mor effect of liposomally encapsulated doxorubicin based on
simulations by pharmacokinetic-pharmacodynamic model-
Acknowledgements ing. J Control Release 1999; 61 (1-2): 93-106
17. Wiseman LR, Spencer CM. Dexrazoxane. A review of its use
This work was supported in part by research grants from as a cardioprotective agent in patients receiving anthracycline-
the University of Pisa to Romano Danesi. There are no based chemotherapy. Drugs 1998; 56 (3): 385-403
potential conflicts of interest relevant to the contents of this 18. Sparreboom A, Planting AS, Jewell RC, et al. Clinical pharma-
cokinetics of doxorubicin in combination with GF120918, a
article. potent inhibitor of MDR1 P-glycoprotein. Anticancer Drugs
1999; 10 (8): 719-28
19. Rushing DA, Raber SR, Rodvold KA, et al. The effects of
References cyclosporine on the pharmacokinetics of doxorubicin in patients
1. Canal P, Chatelut E, Guichard S. Practical treatment guide for with small cell lung cancer. Cancer 1994; 74 (3): 834-41
dose individualisation in cancer chemotherapy. Drugs 1998; 20. Piscitelli SC, Rodvold KA, Rushing DA, et al. Pharmacokinetics
56 (6): 1019-38 and pharmacodynamics of doxorubicin in patients with small
2. Hon YY, Evans WE. Making TDM work to optimize cancer cell lung cancer. Clin Pharmacol Ther 1993; 53 (5): 555-6
chemotherapy: a multidisciplinary team approach. Clin Chem
21. Bastholt L, Dalmark M, Gjedde SB, et al. Dose-response rela-
1998; 44 (2): 388-400
tionship of epirubicin in the treatment of postmenopausal pa-
3. Abraham R, Basser RL, Green MD. A risk-benefit assessment
of anthracycline antibiotics in antineoplastic therapy. Drug tients with metastatic breast cancer: a randomized study of
Saf 1996; 15 (6): 406-29 epirubicin at four different dose levels performed by the Dan-
4. Conte PF, Baldini E, Gennari A, et al. Dose-finding study and ish Breast Cancer Cooperative Group. J Clin Oncol 1996; 14
pharmacokinetics of epirubicin and paclitaxel over 3 hours: (4): 1146-55
a regimen with high activity and low cardiotoxicity in ad- 22. Dobbs NA, Twelves CJ. What is the effect of adjusting
vanced breast cancer. J Clin Oncol 1997; 15 (7): 2510-7 epirubicin doses for body surface area? Br J Cancer 1998; 78
5. Danesi R, Conte PF, Del Tacca M. Pharmacokinetic optimisa- (5): 662-6
tion of treatment schedules for anthracyclines and paclitaxel 23. Gurney HP, Ackland S, Gebski V, et al. Factors affecting
in patients with cancer. Clin Pharmacokinet 1999; 37 (3): epirubicin pharmacokinetics and toxicity: evidence against
195-211 using body-surface area for dose calculation. J Clin Oncol
6. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemother- 1998; 16 (7): 2299-304
apy plus a monoclonal antibody against HER2 for metastatic 24. Jakobsen P, Bastholt L, Dalmark M, et al. A randomized study
breast cancer that overexpresses HER2. N Engl J Med 2001; of epirubicin at four different dose levels in advanced breast
344 (11): 783-92 cancer. Feasibility of myelotoxicity prediction through single
7. Hong RL, Tseng YL. Phase I and pharmacokinetic study of a blood-sample measurement. Cancer Chemother Pharmacol
stable, polyethylene-glycolated lipsomal doxorubicin in pa- 1991; 28 (6): 465-9

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)
444 Danesi et al.

25. Toffoli G, Sorio R, Aita P, et al. Dose-finding and pharmaco- 41. Muller I, Niethammer D, Bruchelt G. Anthracycline-derived
logic study of chronic oral idarubicin therapy in metastatic chemotherapeutics in apoptosis and free radical cytotoxicity.
breast cancer patients. Clin Cancer Res 2000; 6 (6): 2279-87 Int J Mol Med 1998; 1 (2): 491-4
26. Bos AM, de Vries EG, Dombernovsky P, et al. Pharmacokinet- 42. Belaud-Rotureau MA, Durrieu F, Labroille G, et al. Study of
ics of MEN-10755, a novel anthracycline disaccharide ana- apoptosis-related responses of leukemic blast cells to in vitro
logue, in two phase I studies in adults with advanced solid anthracycline treatment. Leukemia 2000; 14 (7): 1266-75
tumours. Cancer Chemother Pharmacol 2001; 48 (5): 361-9 43. Ferraro C, Quemeneur L, Prigent AF, et al. Anthracyclines trig-
27. De Jong J, Geijssen GJ, Munniksma CN, et al. Plasma pharma- ger apoptosis of both G0-G1 and cycling peripheral blood
cokinetics and pharmacodynamics of a new prodrug N-l- lymphocytes and induce massive deletion of mature T and B
leucyldoxorubicin and its metabolites in a phase I clinical cells. Cancer Res 2000; 60 (7): 1901-7
trial. J Clin Oncol 1992; 10 (12): 1897-906 44. Wu XX, Mizutani Y, Kakehi Y, et al. Enhancement of Fas-
28. Lyass O, Uziely B, Ben-Yosef R, et al. Correlation of toxicity mediated apoptosis in renal cell carcinoma cells by adriamy-
with pharmacokinetics of pegylated liposomal doxorubicin cin. Cancer Res 2000; 60 (11): 2912-8
(Doxil) in metastatic breast carcinoma. Cancer 2000; 89 (5): 45. Sun X, Zhou Z, Kang YJ. Attenuation of doxorubicin chronic
1037-47 toxicity in metallothionein-overexpressing transgenic mouse
heart. Cancer Res 2001; 61 (8): 3382-7
29. Lipp HP, Bokemeyer C. Anthracyclines and other intercalating
46. Venturini M, Michelotti A, Del Mastro L, et al. Multicenter
agents. In: Lipp HP, editor. Anticancer drug toxicity: preven-
randomized controlled clinical trial to evaluate cardioprotec-
tion, management and clinical pharmacokinetics. Marcel tion of dexrazoxane versus no cardioprotection in women re-
Dekker, New York, 1999: 81-113 ceiving epirubicin chemotherapy for advanced breast cancer.
30. Twelves CJ, Dobbs NA, Michael Y, et al. Clinical pharmaco- J Clin Oncol 1996; 14 (12): 3112-20
kinetics of epirubicin: the importance of liver biochemistry 47. Ratain MJ, Schilsky RL, Conley BA, et al. Pharmacodynamics
tests. Br J Cancer 1992; 66 (4): 765-9 in cancer therapy. J Clin Oncol 1990; 8 (10): 1739-53
31. Twelves CJ, Dobbs NA, Gillies HC, et al. Doxorubicin phar- 48. Mazza JJ. Hematopoiesis and hematopoietic growth factors. In:
macokinetics: the effect of abnormal liver biochemistry tests. Mazza JJ, editor. Manual of clinical hematology. 2nd ed. Bos-
Cancer Chemother Pharmacol 1998; 42 (3): 229-34 ton: Little Brown, 1995: 1-16
32. Preisler HD, Gessner T, Azarnia N, et al. Relationship between 49. Licata S, Saponiero A, Mordente A, et al. Doxorubicin metabo-
plasma adriamycin levels and the outcome of remission in- lism and toxicity in human myocardium: role of cytoplasmic
duction therapy for acute nonlymphocytic leukemia. Cancer deglycosidation and carbonyl reduction. Chem Res Toxicol
Chemother Pharmacol 1984; 12 (2): 125-30 2000; 13 (5): 414-20
33. Berrak SG, Ewer MS, Jaffe N, et al. Doxorubicin cardiotoxicity 50. Minotti G, Cairo G, Monti E. Role of iron in anthracycline
in children: reduced incidence of cardiac dysfunction associ- cardiotoxicity: new tunes for an old song? FASEB J 1999; 13
ated with continuous-infusion schedules. Oncol Rep 2001; 8 (2): 199-212
(3): 611-4 51. Meibohm B, Derendorf H. Basic concepts of pharmacoki-
34. Biganzoli L, Piccart MJ. The bigger the better? Or what we netic/pharmacodynamic (PK/PD) modelling. Int J Clin Phar-
know and what we still need to learn about anthracycline dose macol Ther 1997; 35 (10): 401-13
per course, dose density and cumulative dose in the treatment 52. Jusko WJ. Pharmacodynamics of chemotherapeutic effects:
of breast cancer. Ann Oncol 1997; 8 (12): 1177-82 dose-time-response relationships for phase-nonspecific
35. Ormrod D, Holm K, Goa K, et al. Epirubicin: a review of its agents. J Pharm Sci 1971; 60 (6): 892-5
efficacy as adjuvant therapy and in the treatment of metastatic 53. Katashima M, Yamada Y, Yamamoto K, et al. Analysis of anti-
disease in breast cancer. Drugs Aging 1999; 15 (5): 389-416 platelet effect of ticlopidine in humans: modeling based on
36. Macchiarini P, Danesi R, Mariotti R, et al. Phase II study of irreversible inhibition of platelet precursors in bone marrow.
high-dose epirubicin in untreated patients with small-cell lung J Pharmacokinet Biopharm 1999; 27 (3): 283-96
cancer. Am J Clin Oncol 1990; 13 (4): 302-7 54. Johnson RD. Population toxicodynamic analysis of pemetrexed
37. Fogli S, Danesi R, Gennari A, et al. Gemcitabine, epirubicin and disodium (Alimta) [abstract no. 729]. Proc Am Soc Clin
paclitaxel: pharmacokinetic and pharmacodynamic interactions Oncol 2000; 19: 188a
in advanced breast cancer. Ann Oncol 2002; 13: 919-27 55. Jakobsen P, Steiness E, Bastholt L, et al. Multiple-dose phar-
38. Tjuljandin SA, Doig RG, Sobol MM, et al. Pharmacokinetics macokinetics of epirubicin at four different dose levels: stud-
ies in patients with metastatic breast cancer. Cancer
and toxicity of two schedules of high dose epirubicin. Cancer
Chemother Pharmacol 1991; 28 (1): 63-8
Res 1990; 50 (16): 5095-101
39. Hollingshead LM, Faulds D. Idarubicin. A review of its phar-
macodynamic and pharmacokinetic properties, and therapeu-
tic potential in the chemotherapy of cancer. Drugs 1991; 42 Correspondence and offprints: Dr Romano Danesi, Division
(4): 690-719 of Pharmacology and Chemotherapy, Department of On-
40. Gewirtz DA. A critical evaluation of the mechanisms of action
proposed for the antitumor effects of the anthracycline anti- cology, Transplants and Advanced Technologies in Medi-
biotics adriamycin and daunorubicin. Biochem Pharmacol cine, University of Pisa, Via Roma 55, Pisa, I-56126, Italy.
1999; 57 (7): 727-41 E-mail: r.danesi@med.unipi.it

© Adis International Limited. All rights reserved. Clin Pharmacokinet 2002; 41 (6)