The magic of small structure differences in a sodium-glucose cotransporter drug discovery project—14C-labelled drug candidates in a key-differentiating study

Article abstract of DOI:10.1002/jlcr.3869

We describe the dramatic differences in the synthesis and physiological and pharmacokinetical profiling of two sodium-glucose cotransporter (SGLT) drug candidates AVE2268 and AVE8887 with very similar chemical structures. It is a classic example of how a radioactive study was able to spare resources in preclinical development prior to entering a costly clinical program. It also demonstrated that radioactive compounds can be used to study differences between two very similar compounds in vivo.

Full text of DOI:10.1002/jlcr.3869

Received: 20 May 2020
DOI: 10.1002/jlcr.3869
Revised: 17 June 2020
Accepted: 3 July 2020
S P E C I A L I S S U E A R T I C L E
The magic of small structure differences in a sodium-
glucose cotransporter drug discovery project—14C-labelled
drug candidates in a key-differentiating study
Manfred Schudok1 | Heiner Glombik2 | Volker Derdau2
1Research and Development, DMPK,
We describe the dramatic differences in the synthesis and physiological and
pharmacokinetical profiling of two sodium-glucose cotransporter (SGLT) drug
candidates AVE2268 and AVE8887 with very similar chemical structures. It is
a classic example of how a radioactive study was able to spare resources in
preclinical development prior to entering a costly clinical program. It also
demonstrated that radioactive compounds can be used to study differences
between two very similar compounds in vivo.
Sanofi-Aventis Germany Deutschland
GmbH, Frankfurt, Germany
2Research and Development, Integrated
Drug Discovery, Sanofi-Aventis Germany
GmbH, Frankfurt, Germany
Correspondence
Volker Derdau, Research and
Development, Integrated Drug Discovery,
Sanofi-Aventis Deutschland GmbH,
Frankfurt, Germany.
K E Y W O R D S
Email: volker.derdau@sanofi.com
SGLT, diabetes, 14C synthesis, pharmacological profiling
1 | INTRODUCTION
These were AVE2268 (1a) and AVE8887 (1b), differing
by a trifluoromethoxy (AVE8887) instead of a methoxy
Sanofi has developed several compounds known as inhib-
itors of sodium-glucose cotransporters (SGLT-1/2).1
These membrane proteins play an important role in
maintaining glucose equilibrium in the human body. It is
known that the inhibition of SGLT 1/2 transporters
decreases glucose blood levels either by preventing
absorption from the intestine (SGLT 1) or by inhibiting
reabsorption from the urine in the kidneys (SGLT 2).2
The concept was successfully proven in clinical trials.
SGLT 2 inhibitors as canaglifozin,3,4 dapaglifozin,5 and
empaglifozin6 are already on the market for the treat-
ment of type 2 diabetes from 2012 onward reached
blockbuster status. Even today, the development of
small-molecule inhibitors of SGLT is of interest for the
pharmaceutical industry, as several SGLTs like
sotagliflozin (Zynquista®)7 have recently (2019) entered
the market in Europe.
group (AVE2268).8 In animal models, both compounds
reduced the intestinal absorption of glucose (less pro-
nounced) and at the same time increased renal excretion
of glucose. In the course of drug development, the candi-
date's pharmacokinetic (PK) properties and the absorp-
tion, distribution, metabolism, and elimination (ADME)9
characteristics were evaluated in vitro, then in animals10
and finally in humans.11 In order to keep track of the
drug molecules and their metabolites throughout the
body and in excreta, the administration of radiolabelled
drugs was considered essential.12, 13
In conjunction with the planned development pro-
gram for AVE2268 1a and AVE8887 1b (Figure 1),
both14C-labelled and stable isotopically labelled
isotopologues of these new drug candidates were
synthesized.14–17 Both compounds conform to the
Lipinsky rule14 of 5, and due to the trifluoromethyl-group
in AVE8887 was more lipophilic (logD6.8 = 2.8 vs. 2.0 for
AVE2268), with lower aqueous solubility (0.25
vs. 1.59 mg/ml), similar solubility in FeSSIF (2.5
vs. 2.3 mg/ml) and higher human plasma protein binding
(98% vs. 91%). The compounds inhibited Na+-dependent
glucose transporters, with a selectivity of at least 6 for
2 | DISCUSSION
Two candidates were developed by Sanofi as SGLT-2
inhibitors for the treatment of type II diabetes mellitus.
J Label Compd Radiopharm. 2020;1–4.
wileyonlinelibrary.com/journal/jlcr
© 2020 John Wiley & Sons, Ltd.
1

2
SCHUDOK ET AL.
Corrected by dose, radioactivity plasma levels obtained
were similar both after IV (1.5 mg/kg) and PO (5 mg/kg)
administration. The half-life was very long, much longer
than expected (approximately 160 and 140 h, respec-
tively), and even after 35 days (!), excretion was not
complete after oral administration. Thus, after 7 days
(allowed time in metabolism cages for the dogs), recov-
ery of radioactivity was only 64% after oral and 62% after
intravenous administration, half by renal and half by
faecal excretion. In this context be of interest residual
radioactivity in the dogs with prolonged excretion cau-
sed some issues required the decontamination of their
kennels 6 weeks later.
In comparison AVE2268, dosed at 1 mg/kg IV and
5 mg/kg PO, showed a high recovery of approximately
94% after IV and even 96% after PO dosing in male beagle
dogs (Figure 3, Table 2). Here, excretion was complete
after 1 week and mainly via urine (nearly 90% of excreted
radioactivity). The terminal half-life was calculated as
11 h after IV and 15 h after oral dosing, thus well suited
for once-daily dosing. The oral bioavailability was calcu-
lated to be 87%, and the development of the compound
was continued until phase II where it was stopped due to
other reasons.
Interestingly, the great difference of both compounds
1a, 1b had already been realized during the syntheses of
the two 14C-labelled derivatives before. The synthesis of
AVE2268 1a was adapted from the medicinal chemistry
laboratories without challenges, and an overall yield of
50% in five steps was obtained. In detail, starting from
commercially available 3-methoxy thiophene 2 and
[14CO]anisic acid chloride [14C]-3a, a regioselective
Friedel-Crafts acylation followed by methyl ether cleav-
age gave the hydroxyl-thiophene derivative [14C]-5a.
Subsequent alkylation with acetobromo-alpha-glucose
FIGURE 1 Structures of AVE2268 and AVE8887
SGLT-2 compared to SGLT-1, depending on the
in vitro model (AVE2268: IC50(hSGLT-2) = 10 nM;
IC50(hSGLT-1) = 8.2 μM), with almost no effect on
GLUT4. Both compounds powerfully and dose-
dependently inhibited glucose reabsorption in the kid-
neys of healthy male NMRI-mice, with AVE8887 being
slightly more potent and significantly longer acting than
AVE2268. In a 3-week study in db/db mice, the com-
pounds (30 and 100 mg/kg po given with the feed)
lowered basal blood glucose, improved oral glucose toler-
ance, and decreased HbA1c, both being equally effective.
In male Zucker Diabetic Fatty rats, AVE2268 and
AVE8887 (30 and 100 mg/kg given po for 6 weeks) dose-
dependently lowered basal blood glucose and improved
the impaired oral glucose tolerance. Urinary glucose
secretion was tremendously increased (proof of princi-
ple), and both compounds prevented the development of
diabetic syndromes in ZDF rats as compared to controls
(proof of concept). For this reason, both compounds were
selected to be drug candidates and were developed in
parallel until then.
However, in the early preclinical phase of the
program, surprisingly, AVE8887 showed a totally unex-
pected behaviour in a combined radiokinetic-mass bal-
ance study in male beagle dogs (Figure 2, Table 1).
FIGURE 2 Mass balance
study and plasma concentration
in male beagle dogs with
AVE8887-14C

SCHUDOK ET AL.
3
TABLE 1 Data of a radiokinetic-mass balance study in male beagle dogs with AVE8887-14C
Mass Balance in % of Dose Eliminated
P.O.
Dog 1
27.1
37.6
0.7
I.V.
Dog 2
29.0
37.0
0.7
Dog 3
29.4
29.0
2.4
Dog 4
30.3
30.5
0.6
Dog 5
27.9
30.8
0.7
Dog 6
39.3
25.5
1.0
Urine
Faeces
Cage washing
Total
65.4
66.7
60.9
61.4
59.3
65.7
FIGURE 3 Mass balance study and plasma
concentration in male beagle dogs with
AVE2268-14C
TABLE 2 Data of a radiokinetic-mass balance study in male beagle dogs with AVE2268-14C
Mass Balance in % of Dose Eliminated
P.O.
Dog 1
90.6
2.7
I.V.
Dog 4
87.7
0.7
Dog 2
88.4
4.9
Dog 3
89.7
2.8
Dog 5
87.0
2.5
Dog 6
86.2
2.2
Urine
Faeces
Cage washing
Total
3.5
4.0
3.9
6.8
3.8
4.4
96.8
97.2
96.4
95.2
93.4
92.8
under phase transfer conditions, carbonyl reduction with
NaCNBH3/TMSCl and final basic acyl deprotection
afforded the desired compound [14C]-1a (see Scheme 1,
for experimental details, see other studies.14,15
In contrast, the 14C-synthesis of [14C]-1b was full of
challenges and hurdles (Scheme 2). Already in the first
step, the acylation turned out to deliver just black
decomposition products, and therefore, a new pathway
is needed to be established. Starting from [14C]
trifluoromethoxy benzoic acid [14C]-6, the corresponding
Weinreb amide [14C]-7 was formed under Appel condi-
tions. Addition of the lithiated thiophene (butyllithium
was added to the thiophene in refluxing ether) to an ice-
cold solution of the Weinreb amide [14C]-7 yielded [14C]-8
in 76%. Luckily, the rest of the synthesis showed no
SCHEME 1 Synthesis of [14C]-AVE2268 1a

4
SCHUDOK ET AL.
effects of sodium-glucose co-transporter-2 inhibitors. Diabetes
Obes Metab. 2019;21(Suppl. 2):9-18.
3. Jakher H, Chang TI, Tan M, Mahaffey KW. Canagliflozin
review—safety and efficacy profile in patients with T2DM.
Diabetes Metab Syndr Obes. 2019;12:209-215. https://doi.org/10.
2147/DMSO.S184437
4. Lin R, Hoerr DC, Weaner LE, Salter R. Syntheses of isotope-
labeled SGLT2 inhibitor canagliflozin (JNJ-28431754).
J Labelled Comp Radiopharm. 2017;60(13):616-623.
5. Dhillon S. Dapagliflozin: a review in type 2 diabetes. Drugs.
2019;79(10):1135-1146.
01148-3
https://doi.org/10.1007/s40265-019-
6. Scott LJ. Empagliflozin: a review of its use in patients with type
2 diabetes mellitus. Drugs. 2014;74(15):1769-1784. https://doi.
org/10.1007/s40265-014-0298-1
7. Nuffer W, Williams B, Trujillo JM. A review of sotagliflozin for
use in type 1 diabetes. Ther Adv Endocrinol Metab. 2019;10:
1-12. https://doi.org/10.1177/2042018819890527
8. Glombik, H, Frick W, Heuer H, Kramer W, Brummerhop H,
Plettenburg O. Novel thiophenylglycoside derivatives, methods
for production thereof, medicaments comprising said com-
pounds and use thereof; WO 2004;007517.
9. Caldwell J, Gardner I, Swales N. An introduction to drug
disposition: the basic principles of absorption, distribution,
metabolism, and excretion. Toxicol Pathol. 1995;23(2):102-112.
10. Roffey SJ, Obach RS, Gedge JI, Smith DA. What is the objective
of the mass balance study? A retrospective analysis of data in
animal and human excretion studies employing radiolabeled
drugs. Drug Metab Rev. 2007;39:17-43.
SCHEME 2 Synthesis of [14C]-AVE8887 1b
significant deviation from the synthesis of [14C]-1a
reported above. The colourless compound [14C]-1b was
isolated after six reaction steps with an overall yield of
45%. Nevertheless, the change of reactivity in the acyla-
tion step could have made us more suspicious how signifi-
cant the electronic changes in the molecule have been.
11. Vogel HG, Hock FJ, Maas J, Mayer D (Eds). Drug Discovery
and Evaluation: Safety and Pharmacokinetic Assays. Springer
Verlag; 2006.
3 | CONCLUSION
12. Marathe PH, Shyu WC, Humphreys WG. The use of radio-
labeled compounds for ADME studies in discovery and explor-
atory development. Curr Pharm Des. 2004;10:2991-3008.
13. Dalvie D. Recent advances in the applications of radioisotopes
in drug metabolism, toxicology and pharmacokinetics. Curr
Pharm Des. 2000;6:1009-1028.
This is a classic example how a radioactive study was able
to spare resources in preclinical development prior to
entering a costly clinical program. It also demonstrated
that radioactive compounds can be used to study differ-
ences between two very similar compounds in vivo and
that there is a lot of value in comparing different lead
series in early radioactive PK studies to increase the prob-
ability of success in the lead to precandidate phase of
drug research programs.
14. Derdau V, Bierer L, Kossenjans M. Method for producing thio-
phene glycoside derivatives. USPTO 20080207882.
15. Derdau V, Fey T, Atzrodt J. Synthesis of isotopically labelled
SGLT inhibitors and their metabolites. Tetrahedron. 2010;66:
1472-1482. https://doi.org/10.1016/j.tet.2009.12.003 7
16. Atzrodt J, Derdau V, Holla W, Sandvoss M. The synthesis of
selected phase II metabolites—O-glucuronides and sulfates of
drug development candidates. ARKIVOC. 2012;3:257-278.
https://doi.org/10.3998/ark.5550190.0013.319
17. Shultz MD. Two decades under the influence of the rule of five
and the changing properties of approved oral drugs. J Med
Chem. 2019;62(4):1701-1714.
CONFLICT OF INTEREST
All authors are or have been employees of Sanofi
Germany.
ORCID
Volker Derdau https://orcid.org/0000-0002-3767-643X
How to cite this article: Schudok M, Glombik H,
Derdau V. The magic of small structure differences
in a sodium-glucose cotransporter drug discovery
project—14C-labelled drug candidates in a key-
differentiating study. J Label Compd Radiopharm.
2020;1–4. https://doi.org/10.1002/jlcr.3869
REFERENCES
1. Whalen K, Miller S, Onge ES. The role of sodium-glucose co-
transporter 2 inhibitors in the treatment of type diabetes. Clin
Ther. 2015;37(6):1150-1166. https://doi.org/10.1016/j.clinthera.
2015.03.004
2. Brown E, Rajeev SP, Cuthbertson DJ, Wilding JPH. A review of
the mechanism of action, metabolic profile and haemodynamic