Hypoxia-selective 3-alkyl 1,2,4-benzotriazine 1,4-dioxides: The influence of hydrogen bond donors on extravascular transport and antitumor activity

Article abstract of DOI:10.1021/jm701037w

Tirapazamine (TPZ) and related 1,2,4-benzotriazine 1,4 dioxides (BTOs) are selectively toxic under hypoxia, but their ability to kill hypoxic cells in tumors is generally limited by their poor extravascular transport. Here we show that removing hydrogen b

Full text of DOI:10.1021/jm701037w

6
654
J. Med. Chem. 2007, 50, 6654–6664
Hypoxia-Selective 3-Alkyl 1,2,4-Benzotriazine 1,4-Dioxides: The Influence of Hydrogen Bond
Donors on Extravascular Transport and Antitumor Activity
†
‡
Michael P. Hay,* Karin Pchalek, Frederik B. Pruijn, Kevin O. Hicks, Bronwyn G. Siim, Robert F. Anderson,
#
Sujata S. Shinde, Victoria Phillips, William A. Denny, and William R. Wilson
Auckland Cancer Society Research Centre, Faculty of Medical and Health Sciences, The UniVersity of Auckland, Auckland, New Zealand
ReceiVed August 20, 2007
Tirapazamine (TPZ) and related 1,2,4-benzotriazine 1,4 dioxides (BTOs) are selectively toxic under hypoxia,
but their ability to kill hypoxic cells in tumors is generally limited by their poor extravascular transport.
Here we show that removing hydrogen bond donors by replacing the 3-NH group of TPZ with simple
2
alkyl groups increased their tissue diffusion coefficients as measured in multicellular layer cultures. This
advantage was largely retained using solubilizing 3-alkylaminoalkyl substituents provided these were
sufficiently lipophilic at pH 7.4. The high reduction potentials of such compounds resulted in rates of
metabolism too high for optimal penetration into hypoxic tissue, but electron-donating 6- and 7-substituents
moderated metabolism. Pharmacokinetic/pharmacodynamic model-guided screening was used to select BTOs
with optimal extravascular transport and hypoxic cytotoxicity properties for evaluation against HT29 human
tumor xenografts in combination with radiation. This identified four novel 3-alkyl BTOs providing greater
clonogenic killing of hypoxic cells than TPZ at equivalent host toxicity, with the 6-morpholinopropyloxy-
BTO 22 being 3-fold more active.
3
1–37
Introduction
to limited extravascular transport (EVT).
ously shown through modeling
We have previ-
3
2,38
39,40
and analogue studies
Hypoxia has been demonstrated in a wide range of human
tumors, and the prognostic significance of tumor hypoxia is
that improving EVT requires balancing the rate of bioreductive
metabolism relative to the tissue diffusion of the analogues:
analogues with low rates of bioreductive metabolism lack
cytotoxic potency, while analogues with high rates of metabo-
lism suffer poor EVT due to excessive consumption of the
prodrug. Definition of the in vitro structure–activity relationships
(SAR) between the one electron reduction potential, E(1), and
in vitro cytotoxicity for substituted BTO analogues allowed the
rational design of BTO analogues with lowered electron affinity,
derived from electron-donating substituents, which retain high
1
becoming better understood. In particular, hypoxia has a major
2
–6
role in driving progression, invasion, and metastasis, as well
7
–9
10–12
as chemo- and radioresistance.
for agents that selectively target hypoxic tumor cells has become
Consequently, the search
9
,13,14
increasingly important.
One approach to such selective targeting of hypoxic cells is
a
15,16
exemplified by the hypoxic cytotoxin tirapazamine (TPZ, 1).
TPZ, a benzotriazine 1,4-dioxide (BTO), is a prodrug that
4
1
hypoxic selectivity. The analogues in that study were not
designed to display increased aqueous solubility, and EVT was
not explicitly considered. A later study of neutral BTOs
demonstrated an increase in tissue diffusion coefficients,
measured using multicellular layer (MCL) cultures, of analogues
1
7
undergoes selective bioreductive activation under hypoxia to
1
8,19
a DNA damaging species,
which may effect DNA strand
2
0–23
breaks.
TPZ has been studied extensively in vitro and in
and in the clinic. Despite indications of clinical
1
6,24
25
vivo,
39
with increased lipophilicity. Recently, we reported a systematic
evaluation of 3-aminoalkylamino BTOs, seeking analogues with
activity, TPZ has a number of limitations that compromise its
in vivo efficacy. In combination chemo-radiotherapy TPZ
displays a range of significant toxicities that limit optimal
40
improved aqueous solubility and EVT. This study showed that
electron-donating substituents in the 6-position could be use to
counteract increased rates of metabolism caused by the 3-ami-
noalkylamino substituents and demonstrated that sufficiently
lipophilic analogues provide MCL penetration similar to TPZ
and show activity against hypoxic cells in tumor xenografts.
However, any improvements over TPZ were generally due to
higher plasma concentrations at equivalent host toxicity, as in
26–29
delivery,
and its solubility is marginal. TPZ shows signifi-
cantly reduced hypoxic selectivity in vivo (2- to 3-fold)
compared to in vitro (hypoxic cytotoxicity ratio, HCR, ca.
30
5
0–100-fold). This low hypoxic selectivity in tumors has been
ascribed to rapid metabolism of TPZ in hypoxic tissues, leading
a
TPZ, tirapazamine; BTO, 1,2,4-benzotriazine 1,4-dioxide; HCR, hypoxic
*
Corresponding author. Address: Auckland Cancer Society Research
cytotoxicity ratio; EVT, extravascular transport; E(1), one-electron reduction
potential; PK/PD, pharmacokinetic/pharmacodynamic; D, tissue diffusion
coefficient; MCL, multicellular layer; kmet, first-order rate constant for
bioreductive metabolism; SAR, structure activity relationships; MTD,
maximum tolerated dose; P7.4, octanol/water coefficient at pH 7.4; CT10,
area under concentration-time curve providing 10% surviving fraction; M10,
amount of BTO metabolized for 10% surviving fraction; X½, calculated
penetration half-distance into hypoxic tissue; SF, surviving fraction; LCK,
log cell kill; AUCpred, area under the plasma concentration-time curve
required to give 1 log of cell of hypoxic cells in HT29 tumors; HCD, hypoxic
cytotoxicity differential; LCKpred, model-predicted logs of hypoxic cell kill.
Centre, Faculty of Medical and Health Sciences, The University of
Auckland, Private Bag 92019, Auckland 1142, New Zealand. Phone: 64-
-3737599 ext.
9
86598.
Fax:
64-9-3737502.
E-mail:
m.hay@auckland.ac.nz.
†
Current address: Boehringer Ingelheim Pharma GmbH & Co. KG,
5216 Ingelheim am Rhein, Germany.
5
‡
Current address: OXiGENE Inc., Magdalen Centre, Robert Robinson
Avenue, The Oxford Science Park, Oxford OX4 4GA, United Kingdom.
#
Current address: Dept of Pharmacy and Pharmacology, University of
Bath, Claverton Down, Bath BA2 7AY, United Kingdom.
1
0.1021/jm701037w CCC: $37.00

2007 American Chemical Society
Published on Web 12/06/2007

Hypoxia-SelectiVe 3-Alkyl 1,2,4-Benzotriazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6655
a
the most active compound of this series (the 6-methoxy
Scheme 1
3
-aminoakylamino BTO 2), rather than increased EVT, which
was generally no greater than for TPZ itself. There is thus
potential for further improving therapeutic activity by increasing
EVT of BTOs.
In this study, we test the hypothesis that removing hydrogen
bond donors in the 3-position, by replacement of the 3-amino
or 3-aminoalkyl groups with a 3-alkyl moiety, will provide a
useful increase in EVT. Examples of 3-alkyl BTOs such as 3–5
4
2
have been examined previously, and although some (e.g., 4,
) were active in vivo they were not superior to TPZ. However,
a
5
Reagents: (i) Pd(PPh3)4, stannane, DME, reflux (method A); (ii)
replacement of the strongly electron donating amine with the
weaker alkyl substituent raises the reduction potential; we
therefore hypothesized that the rates of bioreductive metabolism
of such compounds may be too high for adequate EVT. Thus,
in the present study we sought to apply the strategies developed
previously to optimize 3-aminoalkyl analogues, particularly
the use of electron-donating substituents in the 6- or 7-position
to counteract high rates of metabolism expected to be conferred
by 3-alkylaminoalkyl side chains.
CF3CO3H, DCM (method B); (iii) O3, MeOH/DCM, -78 °C; NaBH4,
EtOH, -40 °C; (iv) TMSCH2N2, HBF4, DCM; (v) 9-BBN, THF; NaOH,
H2O2, (vi) Pd(PPh3)4, PhB(OH)2, Cs2CO3, H2O, DME, reflux; (vii) NaOMe,
MeOH.
a
4
0
Scheme 2
Maximizing activity against hypoxic cells in tumors requires
optimizing all aspects of drug action, including cytotoxic
potency and plasma pharmacokinetics, not just EVT. To
integrate all such elements, as in the previous study, we have
used a spatially resolved pharmacokinetic/pharmacodynamic
4
0
a
Reagents: (i) MsCl, Et N, DCM; (ii) amine (method C); (iii) CF CO H,
3
3
3
CF3CO2H, DCM (method B).
(
PK/PD) model to guide identification of synthetic targets and
a
selection of priority compounds for in vivo evaluation. We have
previously validated this model by demonstrating its ability to
predict killing of hypoxic cells in HT29 and SiHa tumors.
Scheme 3
3
7
38
This mathematical model uses EVT parameters measured in
vitro (tissue diffusion coefficient, D, determined in aerobic
MCLs, and the first-order rate constant for bioreductive
metabolism in hypoxic cell suspensions, kmet) to calculate the
drug concentration-time profile at each position in a tumor
microvascular network, using the measured plasma PK as input.
The model then applies the PK/PD (exposure/cell killing)
relationship determined in vitro to calculate the probability of
cell kill at each position within the tumor region defined by the
microvascular network, and the overall drug-induced kill in the
hypoxic subpopulation. Here we apply this approach to a series
of 3-alkyl BTOs 3–28 with solubilizing alkylamine side chains,
linked at either the 3-position or through electron-donating
alkoxy linkers at the 6- and 7-positions. This approach efficiently
identified four new 3-alkyl BTOs with in vivo activity against
hypoxic tumor cells as well as confirming the activity of the
known BTO analogues 4 and 5.
a
Reagents: (i) Pd(PPh3)4, stannane, DME, reflux (method A); (ii)
CF3CO3H, DCM (method B); (iii) O3, MeOH/DCM, -78 °C; NaBH4,
EtOH, -40 °C; (iv) TMSCH2N2, HBF4, DCM.
methoxide gave the 3-alkoxy BTO 38. The BTO 1-oxides 30,
1, and 33–38 were oxidized to the corresponding 1,4-dioxides
–10 using trifluoroperacetic acid in DCM (method B). The
alcohols 33 and 35 were elaborated to alkylamines 39–42 via
the corresponding mesylates with displacement by the appropri-
ate amines (method C, Scheme 2). Selective aromatic N-
oxidation of the 1-oxides 39–42 was achieved using trifluorop-
eracetic acid in the presence of trifluoroacetic acid to protect
the amine side chain as the salt and gave modest yields of 11–14
3
3
(
method B). It was not possible to prepare C2 homologues of
Chemistry
the C3 BTOs 13 and 14 because of the instability of the products
under the reaction conditions. Attempts to convert the corre-
sponding dioxide alcohol 6 to these analogues via mesylation
and displacement with either dimethylamine or piperidine were
unsuccessful with loss of the 4-oxide being observed.
Synthesis. We have previously demonstrated an efficient
synthetic approach to the known BTO analogues 4 and 5 using
a Stille coupling to install the 3-alkyl substituent into a BTO
4
3
1
-oxide. This represented a significant improvement on the
Bamberger approach previously used to generate the 3-alkyl-
44
4
2
Stille coupling of the 6-methyl BTO 3-chloride 43 with the
appropriate stannane gave the 3-ethyl BTO 1-oxide 44 and the
BTO 1-oxide. We extended this methodology to 6- and
-substituted (F, Me, MeO) BTOs and further increased the
efficiency of the Stille coupling through application of micro-
7
3
-allyl-BTO 1-oxide 45 (method A, Scheme 3). Ozonolysis of
4
4
44
45 gave ethanol 46, which was alkylated to give ether 47.
Oxidation of 44, 46, and 47 with trifluoroperacetic acid gave
the corresponding 1,4-dioxides 15–17. Nucleophilic displace-
ment of the fluorides 48 and 49 with various alkoxides gave
6-alkoxy BTO 1-oxides 50–56 and 7-alkoxy BTO 1-oxides
57–59, respectively (method D, Scheme 4). Oxidation with
trifluoroperacetic acid, using trifluoroacetic acid protection of
the amine side chain, where appropriate, gave the BTO 1,4-
wave reaction conditions. Thus, 3-chloride 29 underwent
Stille coupling with various stannanes and Pd(PPh3)4 to give
3
-alkyl 1-oxides 30–32 (method A, Scheme 1). Ozonolysis of
the allyl side chain of 32 followed by a reductive workup gave
the ethanol 33, which was alkylated to give ether 34. Hydrobo-
ration of alkene 32 gave the propanol 35, which was also
alkylated to give ether 36. Suzuki coupling of chloride 29 gave
the 3-phenyl BTO 37, while nucleophilic substitution of 29 with

6
656 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Hay et al.
a
Scheme 4
a
Reagents: (i) NaOMe, MeOH; (ii) CF3CO3H, DCM (method B); (iii) NaH, alcohol, THF, or DMF (method D); (iv) CF3CO3H, CF3CO2H, DCM (method
B); (v) MsCl, Et3N, DCM; morpholine (method C).
Figure 1
dioxides 18–26, and 28 (method B). Morpholide 27 was
prepared from the alcohol 26 via the mesylate.
Physicochemical Measurements. Solubilities of the BTOs
3
–28 were determined in culture medium containing 5% fetal
calf serum, at 22 °C. Octanol/water partition coefficients were
measured at pH 7.4 (P7.4) for TPZ, 3 and a subset of six BTOs
by a low volume shake flask method, with BTO concentrations
in both the octanol and the buffer phases analyzed by HPLC as
4
0,45
previously described.
These values were used to “train”
ACD LogP/LogD prediction software (v. 8.0, Advanced Chem-
istry Development Inc., Toronto, Canada) using a combination
of ACD/LogP System Training and Accuracy Extender. Ap-
parent (macroscopic) pKa values for the side chain were
calculated using ACD pKa prediction software (v. 8.0, Advanced
Chemistry Development Inc., Toronto, Canada).
One Electron Reduction Potential Measurements E(1).
Pulse radiolysis experiments were performed on The University
of Auckland Dynaray 4 (4 MeV) linear accelerator (200 ns pulse
length with a custom-built optical radical detection system). E(1)
values for the compounds were determined in anaerobic aqueous
solutions containing 2-propanol (0.1 M) buffered at pH 7.0 (10
46
mM phosphate) by measuring the equilibrium constant for the
electron transfer between the radical anions of the compounds
and the appropriate viologen or quinone reference standard. Data
were obtained at three concentration ratios at room temperature
(
22 ( 2 °C) and used to calculate the ∆E between the
compounds and references, allowing for ionic strength effects
on the equilibria.
Biological Methods
In vitro Assays. A schematic of the testing algorithm, guided
by the spatially resolved PK/PD model, is shown in Figure 2.
Cytotoxic potency was determined by IC50 assays, using 4 h
drug exposure of HT29 and SiHa cells under aerobic and anoxic
Figure 2. Flowchart of PK/PD model guided drug evaluation of
BTOs.

Hypoxia-SelectiVe 3-Alkyl 1,2,4-Benzotriazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6657

6
658 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Hay et al.

Hypoxia-SelectiVe 3-Alkyl 1,2,4-Benzotriazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6659
(
H2/Pd anaerobic chamber) conditions in 96-well plates as
4
1
described previously. The hypoxic cytotoxicity ratio (HCR)
was calculated as the intraexperiment ratio aerobic IC50/anoxic
IC50; only compounds with HCR > 20 were investigated further.
The relationship between cell killing, drug exposure, and drug
metabolism (i.e., the in vitro PK/PD model) was measured as
3
5,36
previously described
by following the clonogenicity of
stirred single cell suspensions of HT29 cells for 3 h, at a drug
concentration giving approximately one log kill by 1 h, with
monitoring of drug concentrations in the extracellular medium
by HPLC. This concentration-time data was fitted to determine
the apparent first-order rate constant for metabolic consumption,
kmet.
The in vitro PK/PD model providing the best fit to the
36
clonogenic assay data was determined as previously described,
and the CT10 (area under the concentration-time curve provid-
ing 10% surviving fraction) and M10 (amount of BTO metabo-
lized for 10% surviving fraction) were estimated by interpola-
tion. Two models predominated: one in which the log cell kill
was proportional to both the drug concentration and to its
cumulative bioreductive metabolism (C × M), and one in which
log cell kill was proportional to the amount of BTO metabolized
(
M). For a few compounds, a third model provided a better fit,
with log cell kill proportional to the square of the amount of
2
BTO metabolized (M ). The measured parameter values and
best fit PK/PD model are given in Tables 1 and 2, and the
associated error estimates are tabulated in the Supporting
Information.
Tissue diffusion coefficients, D, of 12 BTOs were measured
using HT29 MCLs under 95% O2 to suppress bioreductive
35
metabolism as previously described. These values, along with
3
6,39
previously reported measurements for BTO analogues
and
unpublished data for other BTOs (a total of 73 compounds),
were used to develop a multiple regression model to calculate
40
D for the other analogues as described previously. This model
3
9
extends the reported dependence, for neutral BTOs, of D on
logP and MW by replacing logP with the octanol/water
distribution coefficient at pH 7.4 (logP7.4) and by adding terms
for the numbers of hydrogen bond donors and acceptors. The
calculated and measured values and the associated error
estimates are tabulated in the Supporting Information. The
calculated values for all BTOs are shown in Tables 1 and 2.
There is a good correlation between the values of D measured
in this study and those calculated from the algorithm [using
2
log-transformation: R ) 0.835, N ) 10, p ) 0.0002). A one-
dimensional EVT parameter, the penetration distance into
hypoxic tissue assuming planar geometry (X½), was calculated
from the opposing effects of D and kmet on tissue transport as
40
previously, providing a ready comparison between analogues:
values are given in Tables 1 and 2, and the associated error
estimates are tabulated in the Supporting Information.
Spatially Resolved PK/PD Model. The in vivo 3D PK/PD
37,40
model has been described in detail recently
Briefly, transport
is modeled in the extravascular compartment of a representative
tumor microvascular network by solving the Fick’s Second Law
diffusion-reaction equations using a Green’s function method,
providing a description of the PK at each point in the tissue
region. The transport parameters used in the model are the tissue
diffusion coefficient D (estimated in MCLs) and kmet for
bioreductive drug metabolism under anoxia (scaled to MCL cell
density) determined in vitro as above. Using the homogeneous
PK/PD model established in vitro for each compound, the log
cell kill was predicted at each position in the tumor microregion.
The overall cell kill through the whole region was then

6
660 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Hay et al.
calculated for drug only and for drug plus 20 Gy radiation (using
the reported radiosensitivity parameters for aerobic and hypoxic
3
6
HT29 cells). The difference between these gives the model-
predicted logs of hypoxic cell kill (LCKpred). In addition, the
hypoxic cytotoxicity differential (HCD) is calculated as a
measure of hypoxic selectivity in the tumor:
LCKpred in the hypoxic region (<4 µM O2)
LCKpred in the well oxygenated region (>30 µM O2)
HCD )
(
1)
Compounds were taken forward for evaluation in mice if the
model predicted that the plasma AUC required for 1 log kill of
the hypoxic subpopulation (AUCpred) was less than 20 000
µM·min (i.e., 7-fold higher than for TPZ), with an HCD > 1.
Consideration was also given to the solubility of the analogues
and identification of a suitable formulation. Thus, BTO 18,
which had poor aqueous solubility, was formulated in a blend
of DMSO, PEG, and saline and BTOs 15–17, 19–21, 25, and
Figure 3. Concentrations of TPZ (circles and squares) and BTO 5
(
triangles) as a function of time in the donor (closed symbols) and
receiver (open symbols) compartments during a representative flux
experiment in the absence of cells, Teflon support membranes alone
(
(
A); or with HT29 MCLs (B) under 95% O
C) were normalized against the initial concentration in the donor
; 50 µM). The lines are fits to the diffusion model,
2
at 37 °C. Concentrations
compartment (C
0
2
6 were not tested in vivo despite passing the in vitro filter.
based on the average thickness of each MCL (148 µm for TPZ and
Plasma PK was determined at or near the MTD (below), and
the spatially resolved PK/PD model was run again with the
measured PK parameters to predict hypoxic LCK at this dose
133 µm for 5) as estimated from the flux of the internal standard urea
(
not shown).
(
LCKpred). Only those compounds with LCKpred > 0.4 were
tested for antitumor activity in vivo.
In Vivo Assays. BTO analogues were formulated as shown
in Table 3 as single i.p. doses of BTOs to homozygous nude
nu
mice (CD-1 Foxn 1 ). The maximum tolerated dose (MTD)
was determined using 1.33-fold dose increments and defined
as the highest dose causing no mortality, body weight loss
(
>15%), or other severe morbidity in any of six mice. Plasma
PK of total drug was measured at 75 or 100% of MTD by HPLC
and the noncompartmental PK parameters AUC, Cmax (maxi-
mum drug concentration) and t½ (terminal half-life) were
determined. Plasma protein binding was found to be negligible
4
0
for a series of related BTO analogues; hence, the measured
concentrations were therefore assumed to reflect free drug
concentrations.
Activity against hypoxic cells in HT29 tumor xenografts was
determined by excision assay following treatment of mice with
the test drug immediately after a dose of γ radiation (20 Gy)
large enough to sterilize the oxic cells in the tumor. Tumors
were excised 18 h later, and single cell suspensions prepared
and plated to determine the number of surviving clonogens/g
of tumor. Surviving fractions (SF) were calculated as the ratio
treated/control clonogens/g of tumor, and drug-induced logs of
hypoxic cell kill (LCK) was calculated as
Figure 4. The log P dependence of diffusion coefficients of neutral
BTOs in HT29 MCLs, (D′, corrected to the same M as TPZ as
described previously; ref 39). Open triangles refer to 3-alkyl compounds
lacking hydrogen bond donors. For comparison, values for 3-amino
analogues (filled circles) and reference compounds (open circles) are
redrawn from ref 39.
r
faster through MCLs. This faster flux occurred despite some
loss of 5 due to metabolism even at this high oxygen concentra-
tion (95%), as demonstrated by the greater loss of 5 from the
donor than accounted for by its appearance in the receiver
compartment (Figure 3B). Fitting the concentration-time curves
to a diffusion-reaction model gave an MCL diffusion coefficient,
log
(
hypoxic cell kill
)
) log
(
SF, radiation
)
-
log SF, radiation + drug
(
)
(2)
Results and Discussion
-7
2
-1
In Vitro Studies. The initial set of neutral BTOs 3–10 with
D, of (20.89 ( 0.36) × 10 cm s , approximately 5-fold
higher than TPZ (measured values of D are tabulated in the
Supporting Information). The values of D are plotted in Figure
4 as a function of logP for the five neutral 3-alkyl BTOs lacking
hydrogen-bonding functionality in the side chain for which MCL
experiments have been performed. These show higher tissue
diffusion coefficients at equivalent logP than the neutral 3-amino
3
-alkyl, 3-alkyloxymethyl, and 3-alkylhydroxyl groups were
generally more soluble than TPZ and displayed similar hypoxic
cytotoxicity and selectivity (Table 1). The HCR values for
3
-phenyl BTO 9 and the 3-methoxy BTO 10 were below the
threshold of 20 and were not evaluated further.
The flux of BTO analogues through HT29 MCLs was
3
9
39
determined as previously described. In a representative
example, the flux of TPZ and BTO 5 was measured through
bare Teflon supports (Figure 3A) or through MCLs (Figure 3B).
Appearance of 5 in the receiver (distal) compartment of the
diffusion chamber was the same as for TPZ using Teflon
supports alone, as expected for their similar MW, but was clearly
BTOs reported previously, which are reproduced in Figure 4.
This supports the hypothesis that hydrogen bonding of the 3-NH2
group lowers D, accounting for the improved tissue penetration
of 3-alkyl analogues at equivalent logP. Consistent with this,
the hydroxyl-substituted 6 and 7 also showed low values of D
relative to their logP values (Supporting Information).

Hypoxia-SelectiVe 3-Alkyl 1,2,4-Benzotriazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6661
The one electron reduction potentials, E(1), for BTOs 3–7,
and 9 were considerably higher than that of TPZ, reflecting the
weaker electron-donating nature of the 3-alkyl (σp ) -0.17)
or 3-alkyloxy (σp ranges from -0.06 to -0.12) substituents
compared to the 3-NH2 substituent (σp ) -0.66). This increase
in electron affinity for the neutral 3-alkyl BTOs has a strong
intermediate potency, while all displayed excellent hypoxic
selectivity. The presence of the lipophilic amine group provided
increased diffusion relative to the corresponding hydroxy
analogues (e.g., compare 22 with neutral 21, and 23, 24 with
neutral 19), emphasizing the negative influence of hydrogen-
bond donors on diffusion. The morpholide 22 had a similar kmet
to TPZ resulting in overall improved EVT (X½ ) 64 µm) relative
to TPZ, whereas increases in kmet seen with the stronger bases
(23, 24) offset potential gains in EVT due to increased D. A
similar pattern to that seen for 20–23 was observed in the
7-alkoxy BTOs 25–28. The neutral BTOs 25 and 26 displayed
low solubility, which was increased by the addition of basic
side chains for BTOs 27 and 28. The hypoxic potencies of the
7-substituted BTOs (25–28) were consistently greater than
the corresponding 6-substituted BTOs 20–23 in accord with the
higher E(1) value observed for 27 compared to 22, while all
displayed good hypoxic selectivity. Diffusion coefficients for
influence on the rate constant for hypoxic metabolism, kmet,
2
(
linear regression of E(1) vs log kmet gave r ) 0.86, P < 0.001,
n ) 8). These high rates of metabolism compromised gains in
EVT due to increased diffusion as reflected by the calculated
X½ values. The exceptions to this, BTOs 4 and 5, displayed
4
-fold increases in kmet which were more than compensated for
by ca. 6-fold increases in D, resulting in modestly improved
X½ values relative to TPZ.
In an effort to improve aqueous solubility, we added basic
tertiary amines to BTOs 11–14. These provided excellent
solubility and the BTOs were more potent under hypoxia than
TPZ and had similar hypoxic selectivity. The addition of the
polar amine groups lowered D relative to the 3-alkyloxy BTOs
2
5–28 followed the same pattern as for 20–23, but were lower,
whereas kmet values for 25–28 were consistently higher than
the corresponding values for 20–23. This resulted in lower EVT
for 25–28 than for the 6-substituted analogues 20–23.
5
and 8, but the 3-alkylamino BTOs still showed improved
diffusion relative to TPZ. The reduction potential of the BTO
3 was found to be ca. 30 mV higher than similar neutral BTOs,
1
Modeling. The in vitro PK/PD model parameters, combined
with the estimates of EVT properties, were used in the spatially
resolved PK/PD model to calculate AUCpred (the area under the
plasma concentration-time curve required to give 1 log of cell
kill of hypoxic cells in HT29 tumors), and HCD (the predicted
hypoxic cytotoxicity differential in these tumors) for 3–8. We
presumably as a consequence of the very weak electron-donating
effect of the partially charged side chain (σp ) 0.01). Hypoxic
metabolism was further increased, by up to 11-fold relative to
TPZ, for the alkylamino BTOs 11–14, completely outweighing
the modest improvements in D and compromising the EVT of
these basic BTO analogues.
4
0
have previously shown that TPZ has a modest AUC require-
40,41
ment of 10 300 µM·min and its calculated hypoxic selectivity
Our previous studies
demonstrated the ability of electron-
3
6,37
is consistent with its observed in vivo activity.
For the
donating substituents on the benzo ring of BTO analogues to
lower E(1) values and hypoxic metabolism. The neutral 6-alkyl
or 6-alkyloxy BTOs with 3-ethyl or 3-ethyloxy substituents
compounds of Table 1, only BTOs 4 and 5 had reasonable
AUCpred values (<20 000 µM·min) with an HCD significantly
greater than 1 and were advanced to in vivo evaluation. AUCpred
values were slightly lower than TPZ for the 6-methyl BTOs
1
5–21 displayed increased lipophilicity, which was moderated
by the presence of free hydroxyl groups (16, 19, 21) (Table 2).
Aqueous solubility for 15–21 was low with the exception of
the 6-hydroxyethyloxy BTO 19. The 6-methyl BTOs 15–17 had
similar hypoxic potency and selectivity in HT29 cells to TPZ,
with improved selectivity observed in SiHa cells, while the
1
5–17 and 6-methoxy BTO 18, and slightly higher for the
alkoxy BTOs 19–21, reflecting their lower potency. Increased
HCD values relative to TPZ were calculated for 15–21, with
the largest increase in selectivity expected for the 6-alkoxy BTOs
1
8–21. The hypoxic potency of 22–28 was reflected in lower
6-alkoxy BTOs 18–21 were less potent and had similar hypoxic
AUCpred, while increased in vivo selectivity was predicted for
all but 23 and 28.
selectivity to TPZ. The increased lipophilicity of 15–21
translates to increased diffusion, with the presence of a free
hydroxyl hydrogen-bond donor strongly impairing diffusion (16,
In Vivo Studies. Application of the in vivo filters of AUCpred
(<20 000 µM·min) and HCD (>1) (Figure 2) eliminated all
the unsubstituted 3-alkyl BTOs except for 4 and 5. All of the
6- and 7-substituted BTOs passed the initial thresholds; however,
difficulties with formulation of the neutral BTOs, such as 18,
meant that neutral BTOs with low aqueous solubility and/or
lower hypoxic potency (15–17, 19–21, 25, and 26) were not
advanced to in vivo testing. Thus, three neutral 3-alkyl BTOs
(4, 5, and 18) and four soluble BTOs (22–24 and 27) were tested
in vivo.
19, 21). The increased electron donation from the 6-substituents
is confirmed by the lowered E(1) values for 15 and 18, -418
and -472 mV, respectively. This has a clear impact on the rates
of hypoxic metabolism with 6-methyl BTOs 15–17 having
slightly raised kmet values, and 6-alkoxy BTOs 18–21 having
lightly lower kmet values, relative to TPZ. The reduced rate of
hypoxic metabolism, combined with increases in D, gave up to
3
-fold increases in X½, signaling significantly improved EVT.
The poor solubility of the neutral BTOs 15–21, coupled with
the lower in vivo potency predicted for the alkoxy BTOs 19–21
vide infra), led us to attach solubilizing amine side chains via
-alkoxy substituents on the benzo ring, BTOs 22–24, with the
The 3-alkyl BTOs were generally less toxic in vivo than TPZ,
with the exception of 24. Determination of plasma PK allowed
calculation of LCKpred in combination with radiation and BTOs
with LCKpred > 0.4 were evaluated in the tumor excision assay.
BTO 18, although having exceptional EVT, gave a relatively
low Cmax of 72 µM, which compromised its LCKpred (Table 3).
BTO 24 displayed a very low AUC and Cmax, and the PK/PD
model predicted little hypoxic cell killing in addition to radiation
and was not evaluated further.
(
6
expectation that the electron-withdrawing tertiary amine would
both provide solubility and moderate the electronic effect of
the alkoxy substituent. We also explored neutral and basic side
chains attached at the 7-position, BTOs 25–28. The 7-alkoxy
substituents are expected to have a smaller electronic influence,
lowering E(1) by ca. 40 mV, whereas the 6-alkoxy substituents
4
1
lower E(1) by ca. 100 mV. The presence of the basic side
chains on BTOs 22–24 provided excellent solubility and
increased hypoxic potency relative to the corresponding neutral
BTOs 19–21, with the near neutral morpholide 22 having
Generally, the BTOs showed little single agent activity, in
keeping with the proposed mechanism of action. We have
4
0
-1
previously shown that TPZ, given at 133 µmol kg (75% of
MTD), gave a LCK of 0.59 logs and that 2 when administered

6
662 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Hay et al.
Figure 5. In vivo activity of BTOs against hypoxic cells in HT29 human tumor xenografts following a single i.p. dose 5 min after irradiation
(
RAD, 20 Gy). Hypoxic cell killing was determined as [log(clonogens/g of tumor for RAD only)] - [log(clonogens/g of tumor for RAD + drug)].
#
P values vs RAD: *P < 0.05, **P < 0.01, P < 0.001. A, BTO given at 100% of MTD 5 min. B, BTO given at 75% of MTD.
at 1000 µmol kg- (100% of MTD) had ca. 3-fold higher
activity (LCK, 1.84 ( 0.16) than TPZ against hypoxic cells in
HT29 tumors (Figure 5 and Table 3). The activity of the neutral
BTOs 4 and 5 seen previously in the fractionated irradiation
1
activity against hypoxic tumor cells in vivo. The advantages of
these compounds are illustrated by the morpholide 22 in which
replacement of the 3-amino group of TPZ with a 3-ethyl
provides a higher tissue diffusion coefficient, reflecting the
importance of minimizing hydrogen-bond donor functionality
in designing bioreductive prodrugs able to reach their target
cells. The resulting increase in one-electron reduction potential,
and thus bioreductive metabolism, would preclude efficient
penetration into hypoxic tissue, but is compensated by an
electron-donating 6-alkoxy side chain bearing a weakly basic
morpholino side chain that increases aqueous solubility with
little impact on lipophilicity at neutral pH. Reflecting these
improvements, the lead compound 22 displayed ca. 3-fold
greater log cell kill of hypoxic cells in HT29 tumors than TPZ
and is the subject of ongoing evaluation.
4
2
experiments with murine SCCVII tumors was confirmed as a
single dose in the HT29 model with LCK values of 0.89 and
.80, respectively. The neutral BTO 18 had similar activity to
and 5, although it was given at 75% of the MTD. The three
soluble BTO analogues tested, 22, 23, and 27, displayed
significant hypoxic cell killing in addition to radiation with 22
ca. 3-fold more active than TPZ. BTO 22 is currently being
tested further in combination with radiation against multiple
human tumor xenografts, both as a single dose and in a
fractionated regimen.
0
4
Conclusions
Experimental Section
We have explored the effect of hydrogen-bond donors on
EVT in this class of 3-alkyl BTOs. Replacement of the 3-NH2
of TPZ with small alkyl substituents provided modest increases
in aqueous solubility and increased diffusion sufficiently to offset
increases in hypoxic metabolism due to raised electron affinity.
The addition of hydrogen-bond donors such as hydroxyl groups
severely impaired diffusion while polar amine groups increased
solubility, but decreased diffusion and increased hypoxic
metabolism, thus compromising EVT. Only two BTOs of the
initial set, 4 and 5, were predicted to be active in vivo, and this
was confirmed in the tumor excision assay.
The addition of lipophilic electron-donating 6-substituents
increased diffusion and lowered hypoxic metabolism, contribut-
ing to excellent EVT. The positive effect was reduced by the
addition of hydrogen-bond donors and acceptors. The low
solubility of these neutral BTO analogues 18–21 precluded in
vivo testing of all but 18. BTO 18 was found to be active in
vivo with a relatively low Cmax compromising activity. The
addition of solubilizing amine groups via a 6-alkoxy linker
Chemistry. General experimental details are described in the
Supporting Information. TPZ and BTO 2 were synthesized as
previously described.
Example of Synthetic Methods. (See Supporting Information
for full experimental details of compounds 3–28.)
Preparation of 1-Oxides. General Method A. Pd(PPh ) (0.1
3 4
mmol) was added to a stirred, degassed solution of chloride (2.0
mmol) and stannane (2.4 mmol) in DME (20 mL), and the solution
4
0
was stirred under N at reflux temperature for 16 h. The solvent
2
was evaporated, and the residue was dissolved in DCM (10 mL)
and stirred with saturated aqueous KF solution (10 mL) for 30 min.
The mixture was filtered through Celite, the Celite was washed
with DCM, and the combined organic filtrate washed with water.
The organic fraction was dried, the solvent was evaporated, and
the residue was purified by chromatography, eluting with DCM to
give product, which was, if necessary, further purified by chroma-
tography, eluting with 20% EtOAc/pet. ether, to give the 3-alkyl-
BTO.
Preparation of 1,4-Dioxides. General Method B. Hydrogen
peroxide (70%, 10 mmol) was added dropwise to a stirred solution
of trifluoroacetic anhydride (10 mmol) in DCM (20 mL) at 0 °C.
The mixture was stirred at 0 °C for 5 min, warmed to 20 °C, stirred
for 10 min, and cooled to 5 °C. The mixture was added to a stirred
solution of 1-oxide (1.0 mmol) [and TFA (5.0 mmol) for substrates
with an amine side chain] in DCM (15 mL) at 0 °C, and the mixture
was stirred at 20 °C for 6–16 h. The solution was carefully diluted
with water (20 mL), the mixture was made basic with aqueous
(
BTOs 22–24) provided increased aqueous solubility, but had
a modest negative impact on EVT through lowered diffusion
relative to the neutral analogues 19–21. A similar pattern of
activity was seen with 7-substituted BTOs 25–28, although
hypoxic potency and metabolism was consistently higher and
the diffusion and EVT consistently lower than the corresponding
6
-substituted analogues. Model predictions of activity for the
NH
4
OH solution, and the mixture was stirred for 15 min and then
soluble BTOs 22, 23, and 27 were confirmed with all three more
active in vivo than TPZ at equitoxic doses.
PK/PD-guided screening of BTOs designed to have improved
EVT has assisted us to identify four novel BTO analogues with
extracted with CHCl
3
(5 × 50 mL). The organic fraction was dried,
and the solvent was evaporated. The residue was purified by
chromatography, eluting with a gradient (0–15%) of MeOH/DCM,
to give the 1,4-dioxide.

Hypoxia-SelectiVe 3-Alkyl 1,2,4-Benzotriazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26 6663
Preparation of 1-Oxides. General Method C. MsCl (5.0 mmol)
Endogenous markers of two separate hypoxia response pathways
(
hypoxia inducible factor 2 alpha and carbonic anhydrase 9) are
was added to a stirred solution of alcohol (4.0 mmol) and Et
3
N
associated with radiotherapy failure in head and neck cancer patients
(
5.2 mmol) in dry DCM (50 mL) at 20 °C, and the solution was
recruited in the CHART randomized trial. J. Clin. Oncol. 2006, 24,
stirred for 2 h. The solution was diluted with DCM (50 mL) and
washed with water (2 × 30 mL), brine (50 mL), and the organic
fraction was dried and the solvent was evaporated. The residue was
dissolved in THF (50 mL) and amine (100 mmol) added, and the
solution was heated at reflux temperature for 6 h, then stirred at 20
7
27–735.
(
13) Denny, W. A.; Wilson, W. R.; Hay, M. P. Recent developments in
the design of bioreductive drugs. Br. J. Cancer 1996, 74, S32–S38.
(14) Brown, J. M.; Giaccia, A. J. The unique physiology of solid tumors:
opportunities (and problems) for cancer therapy. Cancer Res. 1998,
5
8, 1408–1416.
°
C for 16 h. The solvent was evaporated and the residue was
(
15) Brown, J. M. SR 4233 (tirapazamine): a new anticancer drug exploiting
hypoxia in solid tumours. Br. J. Cancer 1993, 67, 1163–1170.
partitioned between EtOAc (100 mL) and water (100 mL). The
aqueous fraction was extracted with EtOAc (3 × 50 mL), the
combined organic fraction was dried, and the solvent was evapo-
rated. The residue was purified by chromatography, eluting with a
gradient (0–10%) of MeOH/EtOAc, to give 1-oxide.
(
16) Denny, W. A.; Wilson, W. R. Tirapazamine: a bioreductive anticancer
drug that exploits tumour hypoxia. Expert Opin. InVest. Drugs 2000,
9
, 2889–2901.
(
17) Patterson, A. V.; Saunders, M. P.; Chinje, E. C.; Patterson, L. H.;
Stratford, I. J. Enzymology of tirapazamine metabolism: a review. Anti-
Cancer Drug Des. 1998, 13, 541–573.
(18) Anderson, R. F.; Shinde, S. S.; Hay, M. P.; Gamage, S. A.; Denny,
W. A. Activation of 3-amino-1,2,4-benzotriazine 1,4-dioxide antitumor
agents to oxidizing species following their one-electron reduction.
J. Am. Chem. Soc. 2003, 125, 748–756.
19) Shinde, S. S.; Anderson, R. F.; Hay, M. P.; Gamage, S. A.; Denny,
W. A. Oxidation of 2-deoxyribose by benzotriazinyl radicals of
antitumor 3-amino-1,2,4-benzotriazine 1,4-dioxides. J. Am. Chem. Soc.
Preparation of 1-Oxides. General Method D. Sodium hydride
(
2.0 mmol) was added to a stirred suspension of fluoride (1.6 mmol)
in alcohol (10 mL), and the resulting solution was stirred at 20 °C
for 3 h under N . The solvent was evaporated, and the residue was
2
partitioned between DCM (30 mL) and water (30 mL). The organic
fraction was dried, and the solvent was evaporated. The residue
was purified by column chromatography, eluting with a gradient
(
(
0–5%) of MeOH/DCM, to give the 1-oxide.
2
004, 126, 7853–7864.
Acknowledgment. The authors thank Jane Botting, Dr.
(20) Wang, J.; Biedermann, K. A.; Brown, J. M. Repair of DNA and
chromosome breaks in cells exposed to SR 4233 under hypoxia or to
ionizing radiation. Cancer Res. 1992, 52, 4473–4477.
Maruta Boyd, Alison Hogg, Sisira Kumara, Sarath Liyanage,
and Joanna Sturman for technical assistance. The authors thank
Degussa Peroxide Ltd, Morrinsville, NZ, for the generous gift
of 70% hydrogen peroxide. This work was supported by the
U.S. National Cancer Institute under Grant CA-82566 (M.P.H.,
K.O.H., F.B.P., K.P., W.R.W., and W.A.D.), the Health
Research Council of New Zealand (W.R.W.), the Cancer Society
of New Zealand (B.G.S.), and the Auckland Division of the
Cancer Society of New Zealand (W.A.D.). Further support from
Proacta Therapeutics Ltd and the Australian Institute of Nuclear
Sciences and Engineering is acknowledged.
(
21) Siim, B. G.; van Zijl, P. L.; Brown, J. M. Tirapazamine-induced DNA
damage measured using the comet assay correlates with cytotoxicity
towards hypoxic tumour cells in vitro. Br. J. Cancer 1996, 73, 952–
9
60.
(
22) Siim, B. G.; Menke, D. R.; Dorie, M. J.; Brown, J. M. Tirapazamine-
induced cytotoxicity and DNA damage in transplanted tumors:
relationship to tumor hypoxia. Cancer Res. 1997, 57, 2922–2928.
(23) Siim, B. G.; Pruijn, F. B.; Sturman, J. R.; Hogg, A.; Hay, M. P.; Brown,
J. M.; Wilson, W. R. Selective potentiation of the hypoxic cytotoxicity
of tirapazamine by its 1-N-oxide metabolite SR 4317. Cancer Res.
2
004, 64, 736–742.
(
24) Brown, J. M; Wang, L. H. Tirapazamine: laboratory data relevant to
clinical activity. Anti-Cancer Drug Des. 1998, 13, 529–539.
(25) Marcu, L.; Olver, I. Tirapazamine: From bench to clinical trials. Curr.
Clin. Pharmacol. 2006, 1, 71–79.
Supporting Information Available: Experimental details and
characterization data for synthetic intermediates and BTOs 3–28;
experimental details for the physicochemical and biological meth-
ods; tables of physicochemical and in vitro data with estimates of
errors. This information is available free of charge via the Internet
at http://pubs.acs.org.
(26) Doherty, N.; Hancock, S. L.; Kaye, S.; Coleman, C. N.; Shulman, L.;
Marquez, C.; Mariscal, C.; Rampling, R.; Senan, S.; Roemeling, R. V.
Muscle cramping in phase I clinical trials of tirapazamine (SR 4233)
with and without radiation. Int. J. Radiat. Oncol. Biol. Phys. 1994,
2
9, 379–382.
(
27) Rishin, D.; Peters, L.; Hicks, R.; Hughes, P.; Fisher, R.; Hart, R.;
Sexton, M.; D’Costa, I.; von Roemeling, R. Phase I trial of concurrent
tirapazamine, cisplatin, and radiotherapy in patients with advanced
head and neck cancer. J. Clin. Oncol. 2001, 19, 535–542.
References
(
1) Evans, S. M.; Koch, C. J. Prognostic significance of tumor oxygenation
in humans. Cancer Lett. 2003, 195, 1–16, and references therein.
(
28) Rischin, D.; Peters, L.; Fisher, R.; Macann, A.; Denham, J.; Poulsen,
M.; Jackson, M.; Kenny, L.; Penniment, M.; Corry, J.; Lamb, D.;
McClure, B. Tirapazamine, cisplatin, and radiation versus fluorouracil,
cisplatin, and radiation in patients with locally advanced head and
neck cancer: a randomized phase II trial of the Trans-Tasman Radiation
Oncology Group (TROG 98.02). J. Clin. Oncol. 2005, 23, 79–87.
29) Le, Q.-T.; Taira, A.; Budenz, S.; Dorie, M. J.; Goffinet, D. R.; Fee,
W. E.; Goode, R.; Bloch, D.; Koong, A.; Brown, J. M.; Pinto, H. A.
Mature results from a randomized phase II trial of cisplatin plus
(
(
(
2) Harris, A. L. Hypoxia-a key regulatory factor in tumour growth. Nat.
ReV. Cancer 2002, 2, 38–47.
3) Semenza, G. L. HIF-1 and tumor progression: pathophysiology and
therapeutics. Trends Mol. Med. 2002, 8, S62–S67.
4) Pennachietti, S.; Michieli, P.; Galluzzo, M.; Mazzone, M.; Giordano,
S.; Comoglio, P. M. Hypoxia promotes invasive growth by transcrip-
tional activation of the met protooncogene. Cancer Cell 2003, 3, 347–
(
3
61.
(
5) Cairns, R. A.; Hill, R. P. Acute hypoxia enhances spontaneous lymph
node metastasis in an orthotopic murine model of human cervical
carcinoma. Cancer Res. 2004, 64, 2054–2061.
6) Subarsky, P.; Hill, R. P. The hypoxic tumour microenvironment and
metastatic progression. Clin. Exp. Metastases 2003, 20, 237–250.
7) Durand, R. E. The influence of microenvironmental factors during
cancer therapy. In ViVo 1994, 8, 691–702.
5
-fluouracil and radiotherapy with or without tirapazamine in patients
with respectable stage IV head and neck squamous cell carcinomas.
Cancer 2006, 106, 1940–1949.
(
(
30) Durand, R. E.; Olive, P. L. Physiologic and cytotoxic effects of
tirapazamine in tumor-bearing mice. Radiat. Oncol. InVest. 1997, 5,
(
2
13–219.
(
31) Durand, R. E.; Olive, P. L. Evaluation of bioreductive drugs in multicell
spheroids. Int. J. Radiat. Oncol. Biol. Phys. 1992, 22, 689–692.
32) Hicks, K. O.; Fleming, Y.; Siim, B. G.; Koch, C. J.; Wilson, W. R.
Extravascular diffusion of tirapazamine: effect of metabolic consump-
tion assessed using the multicellular layer model. Int. J. Radiat. Oncol.
Biol. Phys. 1998, 42, 641–649.
(8) Tannock, I. F. Conventional cancer therapy: promise broken or promise
delayed. Lancet 1998, 351 (Suppl 2), 9–16.
9) Brown, J. M.; Wilson, W. R. Exploiting tumor hypoxia in cancer
(
(
treatment. Nat. ReV. Cancer 2004, 4, 437–447.
(
10) Nordsmark, M.; Overgaard, M.; Overgaard, J. Pretreatment oxygen-
ation predicts radiation response in advanced squamous cell carcinoma
of the head and neck. Radiother. Oncol. 1996, 41, 31–39.
11) Fyles, A. W.; Milosevic, M.; Wong, R.; Kavanagh, M. C.; Pintilie,
M.; Sun, A.; Chapman, W.; Levin, W.; Manchul, L.; Keane, T. J.;
Hill, R. P. Oxygenation predicts radiation response and survival in
patients with cervix cancer. Radiother. Oncol. 1998, 48, 149–156.
12) Koukourakis, M. I.; Bentzen, S. M.; Giatromanolaki, A.; Wilson, G. D.;
Daley, F. M.; Saunders, M. I.; Dische, S.; Sivridis, E.; Harris, A. L.
(33) Kyle, A. H.; Minchinton, A. I. Measurement of delivery and
metabolism of tirapazamine to tumour tissue using the multilayered
cell culture model. Cancer Chemother. Pharmacol. 1999, 43, 213–
220.
(34) Baguley, B. C.; Hicks, K. O.; Wilson, W. R. Tumour cell cultures in
drug development. In Anticancer Drug DeVelopment; Baguley, B. C.,
Kerr, D. J., Eds.; Academic Press: San Diego, 2002; pp 269–284.
(
(

6
664 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 26
Hay et al.
(
(
(
35) Hicks, K. O.; Pruijn, F. B.; Sturman, J. R.; Denny, W. A.; Wilson,
W. R. Multicellular resistance to tirapazamine is due to restricted
extravascular transport: a pharmacokinetic/ pharmacodynamic study
in multicellular layers. Cancer Res. 2003, 63, 5970–5977.
36) Hicks, K. O.; Siim, B. G.; Pruijn, F. B.; Wilson, W. R. Oxygen
dependence of the metabolic activation and cytotoxicity of tira-
pazamine: implications for extravascular transport and activity in
tumors. Radiat. Res. 2004, 161, 656–666.
37) Hicks, K. O.; Pruijn, F. B.; Secomb, T. W.; Hay, M. P.; Hsu, R.;
Brown, J. M.; Denny, W. A.; Dewhirst, M. W.; Wilson, W. R. Use of
three-dimensional tissue cultures to model extravascular transport and
predict in vivo activity of hypoxia-targeted anticancer drugs. J. Natl.
Cancer Inst. 2006, 98, 1118–1128.
38) Hicks, K. O.; Myint, H.; Patterson, A. V.; Pruijn, F. B.; Siim, B. G.;
Patel, K.; Wilson, W. R. Oxygen dependence and extravascular
transport of hypoxia-activated prodrugs: comparison of the dinitroben-
zamide mustard PR-104A and tirapazamine. Int. J. Radiat. Oncol. Biol.
Phys. 2007, 69, 560–571.
Published online: November 15, 2007. http://pubs.acs.org/cgi-bin/
asap.cgi/jmcmar/asap/html/jm070670g.html.
(41) Hay, M. P.; Gamage, S. A.; Kovacs, M. S.; Pruijn, F. B.; Anderson,
R. F.; Patterson, A. V.; Wilson, W. R.; Brown, J. M.; Denny, W. A.
Structure-activity relationships of 1,2,4-benzotriazine 1,4-dioxides as
hypoxia-selective analogues of tirapazamine. J. Med. Chem. 2003, 46,
169–182.
(42) Kelson, A. B.; McNamara, J. P.; Pandey, A.; Ryan, K. J.; Dorie, M. J.;
McAfee, P. A.; Menke, D. R.; Brown, J. M.; Tracy, M. 1,2,4-
Benzotriazine 1,4-dioxides. An important class of hypoxic cytotoxins
with antitumor activity. Anti-Cancer Drug Des. 1998, 13, 575–592.
(43) Hay, M. P.; Denny, W. A. A new and versatile synthesis of 3-alkyl-
1,2,4-benzotriazine-1,4-dioxides: preparation of the bioreductive cy-
totoxins SR4895 and SR4941. Tetrahedron Lett. 2002, 43, 9569–9571.
(44) Pchalek, K.; Hay, M. P. Stille coupling reactions in the synthesis of
hypoxia-selective 3-alkyl-1,2,4-benzotriazine 1,4-dioxide anticancer
agents. J. Org. Chem. 2006, 71, 6530–6535.
(
(
45) Siim, B. G.; Hicks, K. O.; Pullen, S. M.; van Zijl, P. L.; Denny, W. A.;
Wilson, W. R. Comparison of aromatic and tertiary amine N-oxides
of acridine DNA intercalators as bioreductive drugs - cytotoxicity,
DNA binding, cellular uptake, and metabolism. Biochem. Pharmacol.
(
39) Pruijn, F. B.; Sturman, J.; Liyanage, S.; Hicks, K. O.; Hay, M. P.;
Wilson, W. R. Extravascular transport of drugs in tumor tissue: Effect
of lipophilicity on diffusion of tirapazamine analogues in multicellular
layer cultures. J. Med. Chem. 2005, 48, 1079–1087.
40) Hay, M. P.; Hicks, K. O.; Pruijn, F. B.; Pchalek, K.; Siim, B. G.;
Wilson, W. R.; Denny, W. A. Pharmacokinetic/pharmacodynamic
model-guided identification of hypoxia-selective 1,2,4-benzotriazine
2
000, 60, 969–978.
(
(
46) Patel, K. B.; Willson, R. L. Semiquinone free radicals and oxygen.
Pulse radiolysis study of one electron equilibria. J. Chem. Soc. Faraday
Trans. 1973, 69, 814–819.
1,4-dioxides with antitumor activity: the role of extravascular transport.
J. Med. Chem. [Online early access]. DOI: 10.1021/jm070670g.
JM701037W