Radiolabeled 5-iodo-3′-O-(17β-succinyl-5α-androstan-3-one) -2′-deoxyuridine and its 5′-monophosphate for imaging and therapy of androgen receptor-positive cancers: Synthesis and biological evaluation

Article abstract of DOI:10.1021/jm9005803

High levels of androgen receptor (AR) are often indicative of recurrent, advanced, or metastatic cancers. These conditions are also characterized by a high proliferative fraction. 5-Radioiodo-3′-O-(17β-succinyl-5α- androstan-3-one)-2′-deoxyuridine 8 and 5

Full text of DOI:10.1021/jm9005803

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
5124 J. Med. Chem. 2009, 52, 5124–5143
DOI: 10.1021/jm9005803
Radiolabeled 5-Iodo-30-O-(17β-succinyl-5r-androstan-3-one)-20-deoxyuridine and
Its 50-Monophosphate for Imaging and Therapy of Androgen Receptor-Positive Cancers: Synthesis
and Biological Evaluation
Zbigniew P. Kortylewicz,* Jessica Nearman, and Janina Baranowska-Kortylewicz*
Department of Radiation Oncology, J. Bruce Henriksen Cancer Research Laboratories, University of Nebraska Medical Center,
986850 Nebraska Medical Center, Omaha, Nebraska 68198-6850
Received May 5, 2009
High levels of androgen receptor (AR) are often indicative of recurrent, advanced, or metastatic cancers.
These conditions are also characterized by a high proliferative fraction. 5-Radioiodo-30-O-(17β-
succinyl-5R-androstan-3-one)-20-deoxyuridine 8 and 5-radioiodo-30-O-(17β-succinyl-5R-androstan-
3-one)-20-deoxyuridin-50-yl monophosphate 13 target AR. They are also degraded intracellularly to
5-radioiodo-20-deoxyuridine 1 and its monophosphate 20, respectively, which can participate in the
DNA synthesis. Both drugs were prepared at the no-carrier-added level. Precursors and methods are
readily adaptable to radiolabeling with various radiohalides suitable for SPECT and PET imaging, as
well as endoradiotherapy. In vitro and in vivo studies confirm the AR-dependent interactions. Both
drugs bind to sex hormone binding globulin. This binding significantly improves their stability in serum.
Biodistribution and imaging studies show preferential uptake and retention of 8 and 13 in ip xenografts
of human ovarian adenocarcinoma cells NIH:OVCAR-3, which overexpress AR. When these drugs are
administered at therapeutic dose levels, a significant tumor growth arrest is observed.
Introduction
One feature common to AR-expressing cancers is their high
S phasefraction.27-34 The survivalandthe timetoprogression
for patients with ovarian cancer are correlated with the tumor
S-phase fraction. Patients with high S-phase fraction tumors
have significantly lower 5-year survival than patients with low
S-phase fraction tumors. Median time to recurrence is 48 and
17 months for low and high S-phase fraction tumor patients,
respectively.28 There is a significant heterogeneity of the mean
S-phase fraction between diploid and aneuploid samples
depending on the tumor site. Diploid lymph node metastases
have the lowest mean S-phase fraction (<7.2%), and the aneu-
ploid lymph node metastases have the highest mean S-phase
fraction (22.3%).29 In prostate cancer, the S-phase fraction is
significantly higher in tumors with high AR density.30,31 The
recurrent prostate tumors, which have the AR amplification,
are highly proliferative and are more often aneuploid com-
pared to tumors with no AR amplification.31 This implies increa-
sed AR-mediated cell proliferation in the recurrent tumor cells.
Drugs described in this study take into account these two
predominant characteristics of the disease in advanced or
recurrent stages and comprise 5R-dihydrotestosterone (DHT)
as the AR-based tumor-seeking moiety and 5-radioiodo-20-
deoxyuridine or its phosphate as the S-phase specific agents to
preferentially target and kill these cancer cells, which are AR-
positive and have a high S-phase fraction. This particular cell
population characterizes tumors that are prone to relapse.29,33,34
When radiolabeled with diagnostic radionuclides, drugs de-
scribed belowwillallowfor the simultaneous evaluation of the
AR status and the S-phase fraction, thereby facilitating tumor
staging and planning of the therapy. To date, 11 new drugs
weredesigned, synthesized, and tested. Twodrugs, which were
identified as good candidates for translational and clinical
studies, are described.
Androgen receptor (ARa) is commonly expressed in many
cancers.1-20 AR is the most frequently detected sex hormone
receptor in breast cancer cells. Its expression is reported
in >70% of all breast cancer cases1-4 and in 45-50% of
patients with estrogen receptor-negative breast cancer.4 The
AR status, not estrogen or progesterone receptor, is predictive
of tumor response tohormonal therapies.21 ARis expressed in
most histological types and stages of prostate cancers includ-
ing primary, metastatic, and hormone refractory malignant
tissues.5-10 High levels of AR predict shorter time to the
biochemical relapse after the androgen deprivation therapy.8
The amplification of AR is associated with the relapsing
disease.9 The AR signaling pathway is critical to the develop-
ment and progression of prostate cancer.10 AR has also been
detected in the majority of ovarian cancers;12-15 however, the
AR status does not appear to be a strong prognostic factor in
ovarian cancer. No definitive correlations between the pre-
sence of AR, blood hormone levels, stage of disease, and
tumor histology have been established22-25 even though the
pathophysiology of this disease supports a strong connection
with androgens.26 Studies into the relevance of AR in various
cancers are impeded because accurate in situ measurements of
the AR expression remain technically challenging.
*To whom correspondence should be addressed. For Z.P.K.: tele-
phone, 402-559-1030; fax, 402-559-9127; e-mail, zkortylewicz@unmc.
edu. For J.B.-K.: telephone, 402-559-8906; fax, 402-559-9127; e-mail,
jbaranow@unmc.edu.
a Abbreviations: AR, androgen receptor; SHBG, sex hormone bind-
ing globulin; DHT, 5R-dihydrotestosterone; HPLC, high-performance
liquid chromatography; ITLC, instant thin layer chromatography, FBS,
fetal bovine serum; PET, positron emission tomography, SPECT, single
photon emission computed tomography; PBS, phosphate buffered
saline; ip, intraperitoneal.
r
pubs.acs.org/jmc
Published on Web 08/04/2009
2009 American Chemical Society

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 16 5125
Scheme 1. Syntheses of Target Compounds 8 and 13a
a (a) DMTrCl (1.1 equiv), Et3N (1.2 equiv), DMAP (0.05 equiv), in dry pyridine, 0 °C to room temp, 6 h; (b) 4-oxopentanoic acid, DCC/DMAP
in CH2Cl2, room temp, 3 h; (c) ZrCl4 (1.1 equiv) in CH3CN, room temp, 20 min; (d) (i) 1H tetrazole (4.5 equiv); (ii) (t-BuO)2PN(i-Pr)2 (1.2 equiv), 0 °C
to room temp, in dry DMF/THF (3:4), ∼4 h (TLC monitoring); (iii) t-BuOOH, 5-6 M solution in n- decane (2.0 equiv), -16 °C to room temp, 1 h;
(e) H2NNH2 H2O, pyridine, CH3COOH, room temp, 2 min; (f) 2, dihydrotestosterone 17β-succinate, DCC/DMAP in CH2Cl2, room temp, or
3
dihydrotestosterone 17β-succinate activated with CDI (1.15 equiv), CH2Cl2, 0 °C to room temp, 40 min; (g) ZrCl4 (1.25 equiv) in CH3CN, room temp,
30 min; (h) Sn2(CH3)6 (1.2 equiv), (Ph3P)Pd(II)Cl2 (0.07 equiv), in dioxane refluxed under N2, ∼3 h (TLC monitoring); (i) (i) Na125I/NaOH (1-10 mCi),
30% H2O2 (5 μL), TFA/CH3CN (0.1% v/v), room temp, 15 min; (ii) HPLC purification; (j) 5, as for step f ; (k) (i) 1H tetrazole (5.0 equiv); (ii) (t-BuO)2-
PN(i-Pr)2 (1.25 equiv), 4 °C to room temp, dry CH3CN, overnight (TLC monitoring); (l) t-BuOOH, 5-6 M solution in n-decane (5.5 equiv), 0 °C to room
temp, 1 h; (m) 9 dried sample, TFA/CH3CN anhydrous, room temp, ∼40 min (TLC monitoring). (n) In preparation of 10: Sn2(n- Bu)6 (1.25 equiv),
(Ph3P)Pd(II)Cl2 (0.1 equiv), EtOAc refluxed under N2, 6 h. In preparation of 11: Sn2(CH3)6 (1.5 equiv), (Ph3P)Pd(II)Cl2 (0.07 equiv), EtOAc refluxed
under N2, 2 h. (o) As for step i; (p) (i) dried sample of 12 (1-10 mCi) under N2, anhydrous CH3CN (200 μL), TFA (30 μL), ∼60 min (HPLC monitoring);
(ii) HPLC purification.
Results
Chemistry. Syntheses of 125I-radioiodinated compounds 8,
uridine. As a result, 5-iodo-O,O0-(di-tert-butyl)-20-deoxy-
uridin-50-yl phosphate 5 was synthesized first and was sub-
sequently reacted with dihydrotestosterone 17β-succinate to
furnish 9 in satisfactory yield of 69%.
13, 17, and 20, based on the non-carrier-added electrophilic
iododestannylation of the corresponding trialkylorganotin
precursors,35,36 are detailed in Schemes 1 and 2. Nonradioac-
tive iodo analogues 5 and 15, as well as 6 and 9 (those con-
taining the androstan-3-one moiety) required in the prepara-
tion of stannanes 7, 10, 11, 16, and 18, were constructed first.
The esterification of dihydrotestosterone 17β-succinate37
with 2 and the subsequent deprotection of the 50-position
on uridine led to 5-iodo-30-O-(17β-succinyl-5R-androstan-3-
one)-20-deoxyuridine 6 isolated in 73% overall yield. The
phosphorylation of uridine 6 by the reactive phosphorami-
dite provided the corresponding phosphotriester 9, however,
in only 11% overall yield. This low yield is most likely caused
by the large size of dihydrotestosterone at the 30-position on
Initially, the starting phosphotriester 5 was prepared by
the 50-O-phosphitylation of 5-iodo-30-O-levulinyl-20-deoxy-
uridine 4 with di-tert-butyl N,N-diisopropylphosphorami-
dite and excess of 1H-tetrazole, followed by one-pot oxida-
tion of phosphite with tert-butyl hydroperoxide, and the
30-O-Lev group deprotection.38 With this approach, the
average overall yield of 5 was 64%. Later, to simplify this
route, the preparation of levulinate 4 was omitted and the
direct phosphorylation of unprotected 5-iodo-20-deoxyuri-
dine (IUdR) 1 was adopted, furnishing a mixture of three
products: the desired phosphotriester 5 (50-regioisomer) in
48% yield, the 30-isomer (34%), and the 30,50-disubstituted

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
5126 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 16
Kortylewicz et al.
uridine (15%). These were separated on a silica gel column
with no difficulties. Both pathways led to phosphate 5 with
practically the same efficiencies when yields of required
protection/deprotection steps are taken into account.
Organotin precursors 7, 11, 16, and 18 were prepared by
the stannylation of iodouridines 5, 6, 9, and 15, using hexa-
methylditin, and were carried out in the presence of palladium-
(II) catalyst. Phosphotriester 9 was also reacted with hexa-n-
butylditin, giving the corresponding tri-n-butyltin derivative
10. The tri-n-butyltin precursor 10, having a much higher
hydrophobicity than stannane 11, eluted from the HPLC
column 23 min later during the purification of 125I-labeled
12, which allowed for a fast and more efficient separation of
the radiolabeled product, especially when larger volumes, up
to 1 mL, of the crude reaction mixture were injected. The
remaining 125I-labeled target compounds 8, 17, and 19 were
sufficiently well resolved from their trimethyltin precursors
7, 16, and 18, allowing the preparation of radioiodinated
no-carrier-added products, with high specific activities,
using a single HPLC purification.
8, 12, 17, and 19, the standard radioiododestannylation
procedure, originally developed for the synthesis of 5-[125I]-
iodo-20-deoxyuridine,43 was followed.
The tert-butyl esters of the (t-BuO)2P(O) group in phos-
photriesters 12 and 19 were cleaved using a mixture of TFA/
CH3CN (10-13%, v/v, 100 μL) in less than 1 h, cleanly
producing the corresponding target radioiodinated phos-
phates 13 and 20 in high radiochemical yield. In both cases,
the progress of hydrolysis was monitored using HPLC (see
Supporting Information pp S39, S40, S58). However, to
achieve the completion of hydrolysis, it was critical to use
anhydrous tert-butyl phosphotriesters 12 and 19. In the
Exploratory radioiododestannylations were performed with
the trimethyltin precursor 7 (120 μg), Na125I (10 μL, 1 mCi),
and four different oxidants. Literature procedures were carried
out with chloramine T,39 Iodogen,40 hydrogen peroxide,41 and
peracetic acid.42 In all reactions, the product was at least
80-95% 8with 5-15% of inorganic radioiodide. Co-injections
of the radiolabeled material with the independently prepared
iodo standard 6 confirmed the identity of radioiodinated 8.
N-Chloro-substituted oxidants led to the measurable
chlorodestannylation of 7 (∼11% by HPLC). The proton
destannylation, observed to some extent in all reactions, was
definitely elevated when peracetic acid was used. Radio-
iododestannylation performed with hydrogen peroxide as
the oxidant consistently gave the best results with an average
radiochemical yield of >85%. For the synthesis of uridines
Figure 1. Binding of compound 13 to sex hormone binding globu-
lin. Samples were diluted with PBS, pH 7.1, to achieve indicated
above concentrations of SHBG. Compound 13 was dissolved in
CH3CN, and 10 μL (on average 5.4 μCi) was added to each SHBG
dilution. Resulting mixtures were vortexed and incubated for 30 min
at ambient temperature. Aliquots were withdrawn, injected, and
analyzed on the size exclusion HPLC column.
Scheme 2. Syntheses of Control Compounds 17 and 20a
a (a) monoethyl succinate (1.25 equiv), DCC/DMAP, CH2Cl2, room temp, 2 h; (b) TFA (90%), tert-butanol, room temp, 2 h (TLC monitoring);
(c) Sn2(CH3)6 (1.5 equiv), (Ph3P)Pd(II)Cl2 (0.015 equiv), in EtOAc refluxed under N2, 40 min (TLC monitoring); (d) (i) Na125I/NaOH (1-10 mCi), 30%
H2O2 (5 μL), TFA/CH3CN (0.1% v/v), room temp, 15 min; (ii) HPLC purification; (e) as for step c, refluxed in dioxane, 3 h; (f) as for step d; (g) (i) dried
sample of 19 (1-10 mCi) under N2, anhydrous CH3CN (200 μL), TFA (30 μL), ∼40 min (HPLC monitoring); (ii) HPLC purification.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 16 5127
Figure 2. HPLC analyses of 8and 13 interactions with purified human SHBG and human (HS), rabbit (RS), and mouse (MS) serum. After 60 min of
incubation, compounds were analyzed on a size-exclusion column (data for 8 are shown in panel A; data for 13 are shown in panel C). Size exclusion
conditions are as follows: solvent A, 0.1 M potassium phosphate buffer and 0.1 M Na2SO4, 1:1 (v/v), pH 6.8; solvent B, CH3CN. Elution parameters
on TSK3000 column are as follows: solvent A for 20 min followed by a linear gradient of solvent B from 0% to 40% over 20 min, held at 40% B for
30 min, flow rate 0.7 mL/min. HPLC analyses of extracts were performed on a reverse phase column (panels B and D for 8 and 13, respectively).
Reverse phase conditions are as follows: solvent A, H2O, and solvent B, CH3CN, both containing 0.07% TFA. Elution conditions are as follows: 100%
solvent A for 10 min, then a linear gradient of solvent B from 0% to 95% over 30 min, held at 95% B for 20 min. Flow rate was 0.8 mL/min. The top
row shows size exclusion and C18 HPLC traces of pure 8(red peak in columns A and B) and pure 13 (purple peak in columns C and D). The second row
shows size exclusion and C18 HPLC traces of pure 1 (blue peak in columns A and B), which is the degradation product of 8, and 20 (orange peak in
columns C and D), which is the degradation product of 13. The green peaks in all panels correspond to 8 and 13 bound to SHBG.
presence of water, the cleavage process always stopped at the
(t-BuO)(OH)P(O)-phosphate diester stage. In the hydrolysis
of dried, anhydrous 12 (10 mCi), three components, the
desired (HO)2P(O)-phosphate 13 (92%), (t-BuO)(HO)P-
(O)-phosphate diester (5%), and (t-BuO)2P(O)-triester 12
(e2%), were consistently formed and separated from the
reaction mixture. The cleavage rates of tert-butyl ester
groups of anhydrous [125I]iodo-20-deoxyuridine-50-mono-
phosphate di-tert-butyl phosphotriester 19 were faster.
Within 35 min of hydrolysis of 19, 10 mCi used in the form
of dried residue, the target [125I]iodo-20-deoxyuridine-50-
monophosphate 20 (96%) was separated alongside mono-
tert-butyl ester of 19 (3%). HPLC analyses indicated no less
than 95% of the intact radiochemically pure product after
24 h of storage. HPLC co-injections of the independently
prepared iodo standards 6, 9, 14, and 15 with the correspond-
ing radioiodinated derivatives 8, 12, 13, and 17 were used to
confirm the identity of radiolabeled compounds. HPLC
retention times of 5-[125I]iodo-20-deoxyuridine-50-monopho-
sphate 20 were identical to the retention time of commer-
cially available 5-iodo-20-deoxyuridine 50-monophosphate.
Binding and Stability Studies. Binding of compounds 8, 13,
17, and 20 to human sex hormone binding globulin (SHBG)
was measured using size-exclusion HPLC and instant thin
layer chromatography (ITLC) methods. Figure 1 shows a
typical binding isotherm for 13 derived from the HPLC data.
Kd for this compound was estimated at 17.2 ( 4.2 nM and
Bmax at 4.5 ( 0.4 nM. Interactions of all compounds with
human, rabbit, and mouse serum were also analyzed. Data
for 8 and 13 are shown in Figures 2 and 3. The nonspecific
control compounds 17 and 20 are in the Supporting Informa-
tion (S53, S54, S61, S62). Additionally, to confirm the
specific binding to SHBG and the lack of binding to albumin,
all radioactive drugs were incubated with albumin alone,
PBS, and human SHBG mixed with albumin, and these
mixtures were analyzed on the size exclusion column
(Supporting Information, pp S25, S44, S45, S48).
Compounds 8 and 13 bind to pure SHBG isolated from
human serum. Both compounds also bind to SHBG in rabbit
and human serum (Figures 1, 2, and 3). This binding confers
a significant degree of stability, as indicated by the presence
of intact 8 and 13 in extracts prepared from the incubation