ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Mar. 2006, p. 874–879
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Vol. 50, No. 3
Pharmacokinetics of Telbivudine following Oral Administration of
Escalating Single and Multiple Doses in Patients with Chronic
Hepatitis B Virus Infection: Pharmacodynamic Implications
Xiao-Jian Zhou,1* Seng-Gee Lim,2Deborah M. Lloyd,1George C. Chao,1
Nathaniel A. Brown,1and Ching-Lung Lai3
Idenix Pharmaceuticals Inc., 60 Hampshire Street, Cambridge, Massachusetts 021391; National University Hospital,
5 Lower Kent Ridge Road, Singapore 119074, Singapore2; and Division of Hepatology, University of Hong Kong,
Queen Mary Hospital, 102 Pok Fu Lam Road, Pok Fu Lam, Hong Kong, China3
Received 11 July 2005/Returned for modification 1 September 2005/Accepted 8 December 2005
The pharmacokinetics of telbivudine were evaluated in adult patients with chronic hepatitis B virus (HBV)
infection following once-daily oral administration at escalating doses of 25, 50, 100, 200, 400, and 800 mg/day for 4
weeks. Telbivudine was rapidly absorbed after oral administration, with the median times Tmaxto the maximum
plasma concentration (Cmax) ranging from 0.8 to 3.0 h postdosing across cohorts. Single-dose and steady-state
maximum Cmaxs and the areas under the plasma concentration-time curve from time zero to time t (AUC0–ts)
increased proportionally with dose. At steady-state, the values of Cmaxand AUC0–twere higher than those obtained
after the administration of a single dose, indicative of a slight accumulation, with the ratios of the steady-state value
to the value after the administration of a single dose ranging from 1.14 to 1.49 for Cmaxand from 1.40 to 1.70 for
the substantial steady-state trough plasma levels observed in the groups receiving doses of 100 to 800 mg were
clearly indicative of the presence of a second slower elimination phase, with the mean estimated half-lives ranging
maximum-effect modeling established a quantitative relationship between a reduction in serum HBV DNA levels
and parameters of drug exposure, in particular, the steady-state Cmaxand AUC0–t. In summary, this study showed
that telbivudine exhibits dose-proportional plasma pharmacokinetics with sustained steady-state drug exposure
and exposure-related antiviral activity, supporting the need for further clinical studies by use of a once-daily
regimen in patients with chronic HBV infection.
Telbivudine (?-L-2?-deoxythymidine) is an L-configured nucle-
oside with potent and specific activities against hepatitis B virus
(HBV) and other hepadnaviruses and no appreciable activity
against human immunodeficiency virus or other viruses (3). The
in vitro median effective concentration of telbivudine for reducing
extracellular DNA levels in HBV-expressing hepatoma cell line
2.2.15 was 0.19 ?M (?0.05 ?g/ml). In woodchucks chronically
infected with woodchuck HBV, up to 28 days of telbivudine treat-
ment produced consistent, multilog reductions in circulating se-
rum woodchuck HBV DNA levels (3).
In vitro toxicological assessments produced no adverse
findings (1–3). The 50% cytotoxic concentration of telbivu-
dine in 2.2.15 cells was ?2,00 ?M (?500 ?g/ml), indicating
that it has an excellent therapeutic index in cell culture (3).
Other in vitro results suggested that telbivudine is unlikely
to be associated with hematologic or mitochondrial toxici-
ties, peripheral neuropathy, or myopathy (3). Mutagenic test
results were negative (1).
Acute and subchronic toxicology studies did not identify any
preclinical safety issues for telbivudine. In acute and sub-
chronic (28-day) toxicity studies conducted with rats and mon-
keys with daily doses up to 2,000 mg/kg of body weight, no
treatment-related clinical abnormalities were observed (1, 2).
Preclinical pharmacological studies conducted with cyno-
molgus monkeys following the administration of oral and in-
travenous doses showed that telbivudine was well absorbed,
with an oral bioavailability of 68% (4). Renal clearance appeared
to be the major pathway of telbivudine elimination (4). Prelimi-
nary results from a mass balance study with rats and14C-labeled
telbivudine indicated no major plasma or urine metabolite
The encouraging preclinical anti-HBV activity and favorable
safety profile of telbivudine prompted a phase I/II study with
patients with chronic HBV infection. The objectives were to
evaluate the safety, pharmacokinetics, and antiviral activity of
telbivudine administered orally at escalating doses starting at
25 mg/day for 4 weeks. Clinical efficacy and safety results have
recently been reported in more detail elsewhere (6). This re-
port provides a detailed pharmacokinetic analysis of telbivu-
dine and elaborates the pharmacodynamic relationships with
implications for telbivudine dose selection in subsequent phase
IIb (7) and phase III clinical safety and efficacy trials.
MATERIALS AND METHODS
Study population. Patients meeting the following inclusion criteria were eli-
gible for the study: age of 18 years or older; documented chronic HBV infection,
as determined by the presence of serum HBsAg for ?6 months; serum HBeAg
positivity for ?1 month; a serum HBV DNA level of ?107copies/ml; and a
serum alanine aminotransferase level less than five times the upper limit of
normal. Written informed consent was obtained from all patients. Major exclu-
sion criteria included pregnancy or breast-feeding; coinfection with hepatitis C
* Corresponding author. Mailing address: Idenix Pharmaceuticals
Inc., 60 Hampshire Street, Cambridge, MA 02139. Phone: (617) 995-
9805. Fax: (617) 995-9801. E-mail: firstname.lastname@example.org.
virus, hepatitis D virus, or human immunodeficiency virus; any previous treat-
ment with a nucleoside analog or any other drug treatment for HBV in the year
before the baseline; treatment with immunomodulatory agents or corticosteroids
within 6 months before the baseline; decompensated liver disease; hepatocellular
carcinoma; a history of other clinically important diseases and current abuse of
alcohol or illicit drugs. The trial was conducted at the Queen Mary Hospital in
Hong Kong and the National University Hospital in Singapore. The trial was
approved by the ethics committees of both centers. The first patient was re-
cruited on 20 July 2000, and last patient completed the trial on 5 July 2002.
Study design. This was a phase I/II, double-blind, placebo-controlled, dose-
escalation trial for evaluation of the safety, pharmacokinetics, and anti-HBV activity
of telbivudine administered once daily for 4 weeks at escalating doses of 25, 50, 100,
200, 400, and 800 mg, with a 12-week follow-up period. Seven patients who were
randomly assigned to receive telbivudine or matching placebo at a 6:1 ratio were
enrolled in each dosing cohort. Telbivudine 25-mg tablets were used in the 25- to
200-mg cohorts; 100-mg tablets were given to the 400-mg cohorts; and 200-mg
tablets were administered to the 800-mg cohort. All telbivudine tablets were formu-
lated for immediate release. The study drug was taken orally once daily in the
morning. There was no restriction regarding food intake.
Fourteen to 7 days prior to the baseline (the time of administration of the first
dose), patients underwent a screening evaluation that included a medical history,
a physical examination, and clinical laboratory measurements (HBV DNA level,
serum chemistry, hematology, and urinalysis). Clinical laboratory measurements
were repeated at the baseline and were monitored weekly during the study.
Blood sample collection. An intensive pharmacokinetic evaluation was con-
ducted over a period of 8 h after the first dose and on any day between weeks 2
and 4 (steady state). The selection of the blood sampling duration was largely
dictated by the outpatient nature of the study and consideration of the conve-
nience of the participants. Blood samples (5 ml at each time point) were drawn
into heparin-containing Vacutainer tubes immediately prior to and after dosing
at 0.5, 1, 2, 3, 4, 6, and 8 h. Plasma was obtained by centrifugation and was stored
at ?20°C until analysis. Serum samples for viral load measurement were ob-
tained at the baseline, weekly through week 8, and thereafter every other week
through week 16.
Blood sample analyses. For plasma samples from the 25- to 400-mg dose
cohorts, plasma concentrations of telbivudine were assessed by a validated high-
performance liquid chromatographic (HPLC) method with UV detection.
Briefly, 50 ?l of internal standard (?-L-2?-deoxyadenosine [LdA] at 40 ?g/ml)
and 6 ?l of thymidine phosphorylase (EC 18.104.22.168; ?1 unit/?l; Sigma Chemical
Co., St. Louis, Mo.) were added to 200 ?l of calibration standards (0.1 to 100
?g/ml), quality controls (QC; 0.1 to 80 ?g/ml), and unknown plasma samples.
The mixture was vortexed thoroughly and incubated at 37°C for 1 h to digest any
endogenous thymidine that may interfere with the telbivudine assay. After incu-
bation, acetonitrile (1.5 ml) was added to precipitate protein. The samples were
centrifuged, and the supernatant was recovered and evaporated to dryness. Prior
to chromatographic analysis, dry residues were reconstituted with 150 ?l of
mobile phase (20 mM phosphate buffer containing 2% [vol/vol] acetonitrile) and
centrifuged. The resulting supernatant was transferred into HPLC injection vials.
Reverse-phase chromatography was performed with 50-?l aliquots on a Novapak
C18column (3.9 by 150 mm; 4-?m pore size; Waters, Milford, Mass.). Elution
was carried out isocratically at 1 ml/min. Under these conditions, the retention
times were approximately 4.9 and 7.7 min for telbivudine and LdA, respectively.
Telbivudine and LdA were monitored at 267 nm. This assay was characterized by
a lower limit of quantitation of 0.1 ?g/ml, intra- and interday precisions (coef-
ficients of variation [CVs]) from 1.1 to 10.6%, and accuracies (percent deviation)
from ?3.3 to 6.1%, based on the results for the QC samples.
Plasma samples from the 800-mg cohort were analyzed by a validated HPLC
method with mass-spectrometry (MS)-MS detection. In this assay, the sample
preparation procedure was similar to that for the HPLC with UV detection
method described above. However, because of the high sensitivity of MS-MS
detection, a much smaller sample volume (100 ?l) was used. Reverse-phase
chromatography was performed on a TSK-GEL Amide-80 column (4.6 by 150
mm; 5 ?m pore size; Tosoh Bioscience, Montgomeryville, Pa.). Elution was
carried out isocratically at 1 ml/min with a mobile phase of 90:10 (vol/vol)
methanol–25 mM ammonium formate (pH 3.5). Under these conditions, the
retention times were approximately 1.68 and 1.73 min for telbivudine and LdA,
respectively. Telbivudine and LdA were monitored by using a PE Sciex API 3000
MS-MS mass analyzer at mass transitions of 243.0 to 127.1 m/z and 252.0 to 136.0
m/z, respectively. The mass analyzer was operated under the positive mode by
using atmospheric pressure chemical ionization. This assay has a lower limit of
quantitation of 0.01 ?g/ml, with a calibration curve range from 0.01 to 5 ?g/ml.
The intra- and interday precisions (CVs) and accuracies (percent deviation) were
from 2.3 to 5.6% and ?4.2 to 1.4%, respectively, based on the results for QC
samples with concentrations ranging from 0.03 to 4 ?g/ml. Both bioanalytical
methods were validated to their respective specifications without further cross-
Serum HBV DNA was quantitated by using the COBAS AMPLICOR PCR
assay (Roche Diagnostics, Branchburg, N.J.), which has a quantitation limit of
400 genome copies/ml).
Pharmacokinetic and pharmacodynamic analyses. The plasma concentration-
time data for telbivudine obtained after the first dose and at steady state were
analyzed by model-independent and compartmental modeling approaches by
using Kinetica (version 4.3; Thermo Electron Corporation, Waltham, Mass.).
The maximum plasma drug concentration (Cmax) and the time to Cmax(Tmax)
were obtained directly from the plasma concentration-time profiles. The area
under the plasma concentration-time curve from time zero to time t (AUC0–t),
where t is the time that the concentration was last measurable in a sample, was
calculated according to the linear trapezoidal rule. The half-life (t1/2) observed
over the 8-h sampling period was calculated as 0.693/?, where ? is the slope of the
linear portion of the natural log-transformed post-peak plasma drug concentra-
tion-time curve estimated by linear regression. Potential drug accumulation was
assessed by determination of the ratio of the steady-state value to the first-dose
value (accumulation factor) of Cmaxand AUC0–t.
To characterize the true terminal elimination phase, a compartmental mod-
eling analysis was applied to individual steady-state plasma concentration-versus-
time data for patients with measurable predose trough levels (100- to 800-mg
doses). Because of the lack of samples during the terminal phase, steady-state
predose plasma levels were “flipped over” and considered 24-h datum points. A
two-compartment open model with first-order input and first-order elimination,
parameterized into microconstants (absorption rate constant [Ka] and intercom-
partment rate constants [k12and k21]) and the volume of the central compart-
ment (V1), was fitted to the data. The estimated microconstants were then used
to calculate the elimination rate constant (?) and the terminal-phase half-life
(t1/2?; which is equal to 0.693/?).
The principal parameters underlying plasma drug exposure, including Cmax
and AUC0–t, were assessed for dose proportionality after the first dose and at
steady state in the 25- to 800-mg daily dose range by using the following power
model: Yij? a ? Djb? eij, where Djis the dose at level j; Yijis the pharmaco-
kinetic parameter for subject i at dose level j; a and b are the mean intercept and
slope, respectively; and eijis the residual error for subject i at dose level j. For
practical reasons, the model presented above is log10linearized as follows:
log(Yij) ? a ? b ? log(Dj) ? eij, and fitted by using the GLM (general linear
models) procedure in SAS software (version 8.0; SAS Institute Inc., Cary, N.C.).
A dose-proportional relationship is concluded if the 95% confidence interval of
the estimated mean slope (b) includes unity.
The relationship between telbivudine antiviral activity (pharmacodynamics)
and plasma exposure was explored by using the following Emaxmodel: E ?
Emax? PK/(EPK50? PK), where E and Emaxare the observed and the maximum
antiviral effects, respectively, measured as the viral load reduction (the HBV
DNA level on a log10scale) at the end of treatment (week 4); PK represents the
pharmacokinetic parameters of exposure, i.e., single-dose and steady-state Cmax
and AUC0–t; and EPK50is the value of the pharmacokinetic parameter that
produces 50% of Emax. Nonlinear regression was performed by using the NLIN
procedure in SAS software.
Patient characteristics and disposition. All patients enrolled
were Asian and were primarily of Chinese ethnicity. The treat-
ment groups were comparable with respect to demographics and
baseline serum HBV DNA levels. Table 1 summarizes the pa-
tients’ demographics and baseline HBV-related data. Forty-two
of the 43 patients (36 patients receiving active treatment and 7
patients receiving placebo) completed the 4-week treatment and
the 12-week follow-up periods. One subject randomized to the
placebo in the 200-mg cohort withdrew from the study for per-
sonal convenience reasons unrelated to the study treatment. One
patient in the 400-mg telbivudine cohort completed the study but
was considered noncompliant by the investigator.
Plasma pharmacokinetics. The plasma pharmacokinetics of
telbivudine were investigated over an 8-h period on day 1 and
on any day between weeks 2 and 4 inclusive. Pharmacokinetic
VOL. 50, 2006TELBIVUDINE PHARMACOKINETICS-PHARMACODYNAMICS 875
data from 35 telbivudine-treated patients were evaluated (Ta-
ble 2); the data for the pharmacokinetic parameters for the
noncompliant subject in the 400-mg cohort were estimated
(Table 2 footnotes) but are not included in the cohort sum-
mary statistics. Figure 1 shows the mean plasma telbivudine
concentration-time profiles obtained after single (day 1) and
repeated (steady-state) daily doses of 25 to 800 mg in HBV-
infected patients. Following oral dosing, telbivudine was rap-
idly absorbed, with peak plasma levels reached between 0.8
and 3.0 h (median values) across all dose cohorts on day 1 and
at steady state. Over the dose range of 25 to 800 mg/day
studied, in which the twofold dose increments imply an antic-
ipated doubling factor of 2.0 between adjacent doses with re-
spect to exposure, the mean Cmaxincreased from 0.20 to 3.97
?g/ml (mean doubling factor, 1.9 ? 0.4) on day 1 and from 0.22
to 5.46 ?g/ml (doubling factor, 1.9 ? 0.4) at steady state, and
the mean AUC0–tincreased from 0.51 to 20.94 ?g · h/ml (dou-
bling factor, 2.1 ? 0.5) on day 1 and from 0.75 to 29.73 ?g ·
h/ml (doubling factor, 2.1 ? 0.4) at steady state. The pharma-
cokinetic dose proportionality of telbivudine was ascertained
by statistical analyses of log-transformed parameters and dose.
The model estimates of the slope were close to unity in all
cases: 0.91 (95% confidence interval, 0.81 to 1.01) and 0.93
(95% confidence interval, 0.84 to 1.02) for day 1 and steady-
state Cmax, respectively, and 1.08 (95% confidence interval,
0.96 to 1.20) and 1.03 (95% confidence interval, 0.93 to 1.14)
for day 1 and steady-state AUC0–t, respectively. The telbivu-
dine Cmaxand AUC0–twere greater at steady state than after
a single dose. The mean ratios of the individual paired values
at steady state to the values after administration of a single
dose (accumulation factor) ranged from 1.14 to 1.49 for Cmax
and from 1.40 to 1.70 for AUC0–t. The detailed pharmacoki-
netic parameter data for each dose cohort are presented in
After peak plasma levels were achieved, the disposition of
telbivudine was essentially monophasic over the 8-h sampling
FIG. 1. Mean (SD) single-dose (day 1) and steady-state (week 2 to 4) plasma concentration-time profiles of telbivudine in HBV-infected patients
following oral administration of escalating doses at 25, 50, 100, 200, 400, and 800 mg.
TABLE 1. Patient characteristics at baselinea
Value at the following telbivudine dose (mg):
2550 100 200400800Total
No. of subjects
Gender (% male)
Serum HBV DNAb
(log10no. of copies/ml)
30.9 ? 9.8
68.0 ? 13.4
8.3 ? 1.2
31.7 ? 7.5
72.5 ? 17.2
9.9 ? 1.1
32.0 ? 8.4
63.3 ? 17.2
8.8 ? 0.8
38.7 ? 11.6
59.4 ? 17.2
9.0 ? 1.1
29.4 ? 7.8
62.6 ? 9.1
7.9 ? 0.8
32.5 ? 10.0
63.6 ? 8.5
8.2 ? (0.6)
41.8 ? 16.5
66.3 ? 9.9
8.7 ? 1.3
34.3 ? 10.9
64.6 ? 13.4
8.8 ? 1.1
33 (27–78)68 (22–75)30 (24–139)63 (21–163)81 (20–169)126 (22–545)52 (31–243)58 (20–545)
aAll patients were Asian and seropositive for HBsAg and HBeAg.
bValues are presented as means ? SDs.
cALT, alanine aminotransferase.
dValues are presented as medians (ranges).
876ZHOU ET AL.ANTIMICROB. AGENTS CHEMOTHER.
period. This initial distribution-elimination phase had an esti-
mated t1/2that ranged from 2.5 to 5.0 h after the first dose and
at steady state across all doses. The presence of a second
slower elimination phase was evidenced by the substantial
steady-state predose plasma levels, which ranged from 0.15 to
2.65 ?g/ml (cohort mean values) for doses ?100 mg (Table 2).
To approximate the half-life of the unobserved second elim-
ination phase (t1/2?), a two-compartment model with first-or-
der input and first-order elimination was fitted to individual
steady-state plasma concentration-time data from the 100- to
800-mg dose cohorts, with the predose data “flipped over” and
considered the 24-h datum points. Mean ? standard error
(SE) model estimates were 68.3 ? 8.3 liters for V1, 0.492 ?
0.083 h?1for Ka, 0.361 ? 0.075 h?1for k12, and 0.052 ? 0.039
h?1for k21, respectively. The mean estimated t1/2?ranged from
29.5 to 41.3 h (Table 2). The mean ? standard deviation (SD)
model-predicted day 1 and steady-state AUCs from 0 to 8 h
were 2.89 ? 0.84 and 4.07 ? 0.67 ?g · h/ml, respectively, for the
100-mg dose; 5.48 ? 1.48 and 6.49 ? 1.39 ?g · h/ml, respec-
tively, for the 200-mg dose; 10.35 ? 5.64 and 12.93 ? 5.68 ?g
· h/ml, respectively, for the 400-mg dose, and 19.40 ? 7.06 and
29.89 ? 9.24 ?g · h/ml, respectively, for the 800-mg dose. The
model-predicted AUCs were therefore in excellent agreement
with the observed data presented in Table 2, indicating the ad-
equacy of the modeling analysis. At steady state, the AUC
from 0 to 8 h represented approximately 60% of the AUC over
the dosing interval (24 h).
Pharmacokinetic-pharmacodynamic relationship. An Emax
model was used to delineate the relationship between patient
serum HBV DNA reduction data (log10scale) at the end of 4
weeks of treatment, a measure of the pharmacodynamic effect
of telbivudine, and pharmacokinetic parameters of drug expo-
sure, i.e., Cmaxand AUC0–t, obtained on day 1 and at steady
state. The Emaxmodel fitted the dose-response data well, in
particular, the steady-state Cmaxand AUC0–t, with better pre-
cision for model estimates. As summarized in Table 3, model
estimates for Emaxranging from 3.20 to 3.44 log10were con-
sistent for all four pharmacokinetic parameters examined; es-
timates of the values of the pharmacokinetic parameters re-
quired to produce 50% of Emaxwere 0.06 and 0.12 ?g/ml for
the day 1 and steady-state Cmax, respectively, and 0.13 and 0.41
?g · h/ml for the day 1 and steady-state AUC0–t, respec-
tively. The steady-state plasma exposures required to pro-
duce a 3-log10or 99.9% reduction in the serum HBV DNA
level at week 4 were predicted to be as low as 0.78 ?g/ml for
Cmaxand 3.36 ?g · h/ml for AUC0–t, comparable to the
exposures achieved by a dose of 100 mg/day (Table 2). As
depicted in Fig. 2, while a more profound virologic response
was achieved with higher plasma drug exposure in patients
receiving higher telbivudine doses, consistent with its dose-
proportional pharmacokinetics, a nearly maximal viral load
reduction was obtained with plasma exposures corresponding
to telbivudine doses in the 400- to 800-mg range.
This dose-escalation study evaluated the safety, pharmacoki-
netics, and antiviral activity of telbivudine in HBV-infected pa-
tients during 4 weeks of oral, once-daily treatment. The results of
studies of the safety and antiviral activity of telbivudine, detailed
elsewhere (6), demonstrated that telbivudine is well tolerated
TABLE 2. Values of pharmacokinetic parameters for telbivudine on day 1 (first dose) and at steady state (weeks 2 to 4) following once-daily
oral administration of escalating doses from 25 to 800 mg in HBV-infected patientsj
AUC0–t(?g · h/ml)
Day 1SSDay 1 SS
an ? 2 on day 1 (means [SEs] are shown), and n ? 4 at steady state for the 25-mg group; n ? 6 for all other doses except for the 400-mg group, where n is equal
bOne subject in the 400-mg group was not compliant. The day 1 and steady-state pharmacokinetic parameters for this subject were 3.00 and 2.25 ?g/ml, respectively,
for Cmax, and 16.24 and 12.55 ?g · h/ml, respectively, for AUC0–t. Tmaxwas 1.9 h on both occasions.
cMean ratio of individual paired values at steady state to the values on day 1 for all cohorts except the 25-mg cohort, for which the ratio of the cohort mean values
dTwo subjects in the 800-mg group took a dose within 12 h of the steady-state intensive pharmacokinetic dose, resulting in higher than expected predose levels. The
mean ? SD of the trough concentration (Ctrough) was 0.88 ? 0.59 ?g/ml when the data for these two subjects are excluded.
fCalculated from the observed log-linear portion of the post-peak concentration-time curve up to 8 h, not the terminal-phase t1/2.
gEstimated t1/2?from modeling analysis, approximating the terminal-phase half-life.
hSS, steady state.
iNA, not available.
jData are means (SDs) unless indicated otherwise.
TABLE 3. Emaxmodel analysis for pharmacokinetic-
3.33 ? 0.20
3.44 ? 0.17
0.06 ? 0.03
0.12 ? 0.04
3.20 ? 0.17
3.36 ? 0.16
0.13 ? 0.09
0.41 ? 0.14
aModel estimates for Emaxand EPK50are means ? SEs.
VOL. 50, 2006TELBIVUDINE PHARMACOKINETICS-PHARMACODYNAMICS877
without clinically significant adverse events throughout treatment
and across all dose levels from 25 to 800 mg/day. Compared to
placebo, there were no significant changes in hematologic values
and there were no significant abnormalities in clinical laboratory
data throughout the trial. Clinical adverse events were all mild,
and there was no apparent dose-related or treatment-related pat-
tern in the incidence of adverse events (6). A profound, dose-
related virologic response was observed with all doses tested, with
serum HBV DNA reductions of 3.5 to 4.0 log10achieved with
doses of 400 to 800 mg/day by the end of 4 weeks of treatment.
The pharmacokinetics of telbivudine were evaluated over a
period of 8 h in this first clinical trial conducted with HBV-
infected patients. While this sampling interval proved to be
inadequate for a full pharmacokinetic characterization of tel-
bivudine, it did capture essential pharmacokinetic parameters,
such as Cmax, Tmax, and a significant portion of AUC (approxi-
mately 60%, based on model-predicted steady-state exposure),
therefore allowing assessment of the dose proportionality and the
dose-response relationship. The study results indicated that tel-
bivudine is rapidly absorbed after oral dosing, with Cmaxreached
within 1 to 3 h. Pharmacokinetic parameters of drug exposure,
including Cmaxand AUC0–t, are dose proportional over the dose
range studied. While the absolute oral bioavailability of telbivu-
dine remains unknown, the dose-proportional behavior of telbi-
vudine plasma kinetics indicates consistent absorption of the drug
with no restriction on food intake. A food-effect study later
showed that a high-calorie, high-fat meal had no effect on the
pharmacokinetics of telbivudine (unpublished data).
Over the 8-h period, telbivudine exhibits an apparent single-
phase decline, with a short observed t1/2. This phase later
proved to represent an early distribution-elimination phase, as
the presence of a second, slower elimination phase was evi-
denced by the substantial steady-state levels of telbivudine that
remained measurable 24 h after dosing (predose trough levels;
Fig. 1 and Table 2) in cohorts receiving ?100 mg/day. A mod-
eling approach was used to characterize this unobserved sec-
ond elimination phase by flipping the steady-state predose data
over to 24 h. Although the lack of datum points between 8 and
24 h prevents the model from generating high-precision esti-
FIG. 2. Plots of serum HBV DNA reduction at week 4 versus single-dose and steady-state telbivudine pharmacokinetic parameters underlying
plasma exposure (Cmaxand AUC0–t). Solid and dashed lines, the Emaxmodel-predicted response curves and associated 95% confidence intervals,
respectively. The symbols represent the dose groups.
878ZHOU ET AL.ANTIMICROB. AGENTS CHEMOTHER.
mates of pharmacokinetic parameters, the analysis neverthe-
less identified the second elimination phase, which had a mean
estimated terminal half-life in the range of 29.5 to 41.3 h. The
model-predicted values of the AUC from 0 to 8 h were in
excellent agreement with the experimental data, indicative of
the adequacy of the fitting results. Recent studies with healthy
volunteers with intensive sampling up to 168 h postdosing
confirmed the existence of this second elimination phase (un-
published data). The second phase starts approximately 16 to
24 h after dosing, with a long observed terminal-phase t1/2of
approximately 40 h. This long plasma terminal-phase t1/2of
telbivudine is consistent with the long intracellular t1/2(14 h)
of the active triphosphate form of the drug observed in HepG2
cells (5). The long half-life of plasma telbivudine and its intra-
cellular triphosphate reflect a sustained exposure of the drug
within HBV-infected cells and support the use of once-daily
evidenced by approximately 15 to 50% increases in Cmaxand
AUC0–tat steady state, as well as substantial steady-state predose
levels. Of note, the 800-mg cohort exhibited a higher than
expected mean predose trough level (2.65 ?g/ml) associated
with a large SD (2.01 ?g/ml). This was apparently caused by
the fact that two patients in this group took the steady-state
intensive pharmacokinetic dose in less than 12 h of the time
that they took the previous day’s dose. Unpublished results
from several pharmacokinetic studies with healthy volunteers
with sampling beyond 24 h showed an accumulation factor of
1.2 to 1.6, based on the ratio of the steady-state AUC from 0
to 24 h to the single-dose AUC from 0 to 24 h, therefore
confirming the findings of the present study. The results from
those studies further demonstrated that the drug does not
accumulate further once steady state is achieved after 5 to 7
days of once-daily treatment (8; unpublished data).
The dose-related virologic response (6) prompted a more
detailed evaluation of the telbivudine pharmacokinetic-phar-
macodynamic relationship. An Emaxmodel was successfully
fitted to the individual serum HBV DNA reduction at week 4
versus the pharmacokinetic exposure (Cmaxand AUC0–t) data.
The dose-response relationship was more pronounced at steady
state, suggesting the importance of maintaining continuous
drug exposure, which is ensured by good adherence to treat-
ment, to achieving better antiviral activity. Consistent with the
dose-proportional pharmacokinetics of telbivudine, a more
profound viral load reduction was observed in patients en-
rolled in the higher-dose groups. The Emaxmodel analyses
further indicated that a nearly maximal virologic response was
obtained with the plasma exposure achieved with 400- to
800-mg telbivudine doses, and only a minimal incremental
virologic response could be gained even with substantially
higher doses. Therefore, the results of this exposure parame-
ter-based pharmacokinetic-pharmacodynamic analysis are in
full agreement with those of a previous dose-based Emaxanal-
ysis, which demonstrated that a nearly maximal reduction of
circulating serum HBV DNA (?3.5 to 4.0 log10) was obtained
by week 4 with telbivudine doses between 400 to 800 mg. Based
on these findings, telbivudine doses appropriate for further
clinical evaluation were selected, and doses higher than 800
mg/day were not evaluated (6), despite the excellent safety
profile and dose-proportional pharmacokinetics.
In summary, this clinical evaluation demonstrates that telbi-
vudine exhibits dose-proportional pharmacokinetics and expo-
sure-dependent pharmacodynamics and is well tolerated by
patients with chronic HBV infection (6). This favorable profile
of telbivudine supports ongoing phase III trials for evaluation
of the safety and efficacy of longer-term treatment with telbi-
vudine in patients with chronic hepatitis B.
We are grateful to R. Boehme for critical review of the manuscript
and helpful suggestions.
1. Bridges, E., E. Cretton-Scott, X. J. Zhou, A. Juodawlkis, G. Gosselin, and M.
Bryant. 2001. The anti-HBV agent ?-L-thymidine (LdT) exhibits no observ-
able effects in chronic developmental toxicity studies, p. 84–85, abstr. 85.
Proc., Frontiers in Drug Development for Viral Hepatitis.
2. Bridges, E. G., M. L. Bryant, L. Placidi, A. Faraj, A. G. Loi, C. Pierra, D.
Dukhan, G. Gosselin, J. L. Imbach, B. Herna ´ndez, A. Juodawlkis, B. Tennant,
E. Cretton-Scott, and J. P. Sommadossi. 2001. Toxicological study of the
anti-HBV agent ?-L-2?-deoxythymidine, p. 263–270. In R. F. Schinazi, J. P.
Sommadossi, and C. M. Rice (ed.), Frontiers in viral hepatitis. Elsevier B.V.,
Amsterdam, The Netherlands.
3. Bryant, M. L., E. G. Bridges, L. Placidi, A. Faraj, A. G. Loi, C. Pierra, D.
Dukhan, G. Gosselin, J. L. Imbach, B. Herna ´ndez, A. Juodawlkis, B. Tennant,
B. Korba, P. Cote, P. Marion, E. Cretton-Scott, R. F. Schinazi, and J. P.
Sommadossi. 2001. Antiviral L-nucleosides specific for hepatitis B virus infec-
tion. Antimicrob. Agents Chemother. 45:229–235.
4. Cretton-Scott, E., X. J. Zhou, E. G. Bridges, B. Tennant, A. Juodawlkis, G.
Gosselin, J. L. Imbach, C. Pierra, D. Dukhan, R. F. Schinazi, J. P. Somma-
dossi, and M. Bryant. 1999. Pharmacokinetics of ?-L-thymidine and ?-L-2?-
deoxycytidine in woodchucks and monkeys. Antivir. Ther. 4:A124.
5. Hernandez-Santiago, B., L. Placidi, E. Cretton-Scott, A. Faraj, E. Bridges,
M. L. Bryant, J. Rodriguez-Orengo, J. L. Imbach, G. Gosselin, C. Pierra, D.
Dukhan, and J. P. Sommadossi. 2001. Pharmacology of ?-L-2?-deoxythymi-
dine and ?-L-2?-deoxycydine in HepG2 cells and primary human hepatocytes:
relevance to chemotherapeutic efficacy against hepatitis B virus. Antimicrob.
Agents Chemother. 46:1728–1733.
6. Lai, C. L., S. G. Lim, N. A. Brown, X. J. Zhou, D. M. Lloyd, Y. M. Lee, M. F.
Yuen, G. C. Chao, and Maureen W. Myers. 2004. A dose-finding study of
once-daily oral telbivudine (LdT) in HBeAg-positive patients with chronic
hepatitis B virus infection. Hepatology 40:719–726.
7. Lai, C. L., N. Leung, E. K. Teo, M. Tong, F. Wong, H. W. Hann, S. Han, T.
Poynard, M. Myers, G. Chao, D. Lloyd, and N. A. Brown. 2005. A 1-year trial
of telbivudine, lamivudine, and the combination in patients with hepatitis B e
antigen-positive chronic hepatitis B. Gastroenterology 129:528–536.
8. Zhou, X. J., S. G. Lim, C. L. Lai, P. F. Lam, D. M. Pow, N. A. Brown, and
M. W. Myers. 2001. Pharmacokinetics of ?-L-deoxythymidine (LdT) in healthy
subjects and patients with chronic hepatitis B virus infection, p. 92, abstr. 100.
Proc., Frontiers in Drug Development for Viral Hepatitis.
VOL. 50, 2006TELBIVUDINE PHARMACOKINETICS-PHARMACODYNAMICS879