Kevin Michael Foote, J. Willem M. Nissink, Thomas M. McGuire, Paul Turner, Sylvie Guichard, James T. Yates, Alan Lau, Kevin Blades, Dan Heathcote, Rajesh Odedra, Gary Wilkinson, Zena Wilson, Christine Wood, and Philip J Jewsbury
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 22 Oct 2018
Downloaded from http://pubs.acs.org on October 23, 2018

Just Acceis published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036

Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
6 Discovery and characterization of AZD6738, a potent inhibitor of ataxia telangiectasia mutated and rad3 related (ATR) kinase
11
12
13 with application as an anti-cancer agent
18 Kevin M. Footea#, J. Willem M. Nissink a*, Thomas McGuirea, Paul Turnera, Sylvie
21 Guichardb£, James W. T. Yatesc, Alan Laub, Kevin Bladesa$, Dan Heathcoted, Rajesh Odedrad^,
22
23 Gary Wilkinsona&, Zena Wilsonb, Christine M. Wood a, and Philip J Jewsburya
29 a Chemistry, Oncology, IMED Biotech Unit, AstraZeneca, Cambridge Science Park,
32 Milton Rd, Milton, Cambridge, CB4 0WG, UK

35 b Bioscience, Oncology, IMED Biotech Unit, AstraZeneca, Chesterford Research Park,
37
38 Little Chesterford, Cambridge, CB10 1XL, UK
42 c DMPK, Oncology, IMED Biotech Unit, AstraZeneca, Chesterford Research Park,
43
44 Little Chesterford, Cambridge, CB10 1XL, UK
46
47
48 d Discovery Sciences, IMED Biotech Unit, AstraZeneca, Cambridge Science Park, 310 Milton
51 Rd, Milton, Cambridge CB4 0WG , UK

3 # Current affiliation: Pharmaron, Drug Discovery Services Europe, Hertford Road,
5
6 Hoddesdon, Hertfordshire, EN11 9BU , UK

10 £ Current affiliation: Forma Therapeutics, 500 Arsenal Street, Watertown, MA
11
12 02472, Boston, USA

16 $ Current affiliation: AMR Centre Ltd, 19B70, Mereside Alderley Park, Alderley Edge, SK10
22 & Current affiliation: Bayer, Müllerstraße 178, 13353, Berlin, Germany

26 ^ Current affiliation: Evotec, 114 Innovation Drive, Milton Park, Abingdon Oxfordshire
27
28 OX14 4RZ, UK

33 ABSTRACT

37 The kinase ataxia telangiectasia mutated and rad3 related (ATR) is a key regulator of the
38
39 DNA-damage response and the apical kinase which orchestrates the cellular processes that
40
41
42 repair stalled replication forks (replication stress) and associated DNA double-strand breaks.
43
44
45 Inhibition of repair pathways mediated by ATR in a context where alternative pathways are
46
47 less active is expected to aid clinical response by increasing replication stress. Here we describe
48
49
50 the development of the clinical candidate 2 (AZD6738), a potent and selective sulfoximine
51
52 morpholino-pyrimidine ATR inhibitor with excellent preclinical physicochemical and
54
55 pharmacokinetic (PK) characteristics. Compound 2 was developed improving aqueous

3 solubility and eliminating CYP3A4 time-dependent inhibition starting from the earlier
5
6 described inhibitor 1 (AZ20). The clinical candidate 2 has favorable human PK suitable for
7
8
9 once or twice daily dosing and achieves biologically effective exposure at moderate doses.
10
11 Compound 2 is currently being tested in multiple Phase I/II trials as an anti-cancer agent.

18 INTRODUCTION

21 Human cells are constantly exposed to DNA-damage events as a result of environmental and
22
23 endogenous factors. In order to suppress genomic instability, an integrated group of biological
24
25
26 pathways collectively called the DNA-damage response (DDR), has evolved to recognize,
27
28
29 signal, and promote the repair of damaged DNA.1, 2 DNA-damage leads to cell death if
30
31 sufficiently high and left unrepaired and is the concept behind DDR inhibition for cancer
32
33
34 therapy. Tumor cells are sensitized to DDR based therapies through a combination of relatively
35
36 rapid proliferation and DDR pathways that may already be functionally compromised. Ataxia
38
39 telangiectasia and Rad3-related (ATR) is a serine/threonine-protein kinase belonging to the
40
41
42 phosphatidylinositol 3-kinase-related kinase (PIKK) family of proteins and is a key regulator
43
44 of DNA replication stress response (RSR) and DNA-damage activated checkpoints.3, 4
46
47 Replication stress, a hallmark of cancer,5 may occur in tumors through oncogene drivers or
48
49
50 induced exogenously through treatment with DNA-damaging drugs or ionizing radiation (IR).
51
52 Persistent replication stress leads to DNA breaks which if left unresolved are highly toxic to
53
54
55 cells. In recent years potent and selective inhibitors of ATR (Scheme 1) have been developed

1
2
3 from orthogonal chemical series demonstrating preclinical in vivo proof of concept. These
5
6 pivotal compounds and studies have been extensively reviewed,6-9 and reveal synthetic
7
8
9 lethality of ATR inhibitors on tumors with p53-mutations or Ataxia telangiectasia mutated
10
11 (ATM) loss-of-function,10, 11 as well as synergy in combination with a broad range of
12
13
14 replication stress inducing chemotherapy agents such as platinums,12 ionizing radiation,13, 14
15
16 and with novel agents such as the PARP inhibitor olaparib.15

22 Scheme 1. ATR inhibitors; 2, 3 and 4 are undergoing clinical testing
HN O N

S NH
34 NH2 O N
35 HNN

2; AZD6738

41 3; Berzosertib (M-6620 / VX-970)

4; BAY 1895344

47 We have previously described a series of potent and selective ATR inhibitors, exemplified
48
49
50 by 1 (AZ20), from the sulfonylmethyl morpholino-pyrimidine series.16 Compound 1, and close
51
52 analogues,17 were shown to inhibit the growth of ATM-deficient xenograft models at well
54
55 tolerated doses. However, we did not consider compound 1 of sufficient quality for further
56

1
2
3 development due to low aqueous solubility and high-risk for drug-drug interactions (DDI)
5
6 resulting from Cytochrome P450 3A4 (CYP3A4) time-dependent inhibition (TDI). In this
7
8
9 report we describe our further studies to identify ATR inhibitors with the requisite properties
10
11 suitable for clinical development that led to the discovery of 2 (AZD6738). The sulfoximine
12
13
14 morpholino-pyrimidine 2, along with the aminopyrazine 3 (berzosertib, M-6620 / VX-970),
15
16
17 originating from Vertex and licensed to Merck KGaA, and most recently the naphthyridine 4
18
19 (BAY 1895344) from Bayer,18 have entered human studies. These compounds are being
20
21
22 explored in early-phase clinical trials as single-agents and in combination with standard of care
23
24 (SOC) and novel agents.8, 9
26
27
28
29
30 RESULTS
31
32 Compounds 1, 5, 6, 8, 9 and 10 (Table 1) and intermediates 39, 46 – 49, 70 were prepared as
33
34
35 described previously.16 Compounds 7, 17 and 18 were prepared as shown in Scheme 2 and
36
37
38 compounds 11 – 16 were prepared as shown in Scheme 3. Intermediates 43 – 45 were prepared
39
40 starting from the dichloropyrimidine 39. Suzuki coupling with 2-cyclohexenyl-4,4,5,5-
41
42
43 tetramethyl-1,3,2-dioxaborolane led to 40 while SNAr reaction with 8-oxa-3-
44
45 azabicyclo[3.2.1]octane and 3-oxa-8-azabicyclo[3.2.1]octane afforded 41 and 42 respectively
47
48 with no evidence of substitution at the 2-position. Cyclopropanation with 1,2-dibromoethane
49
50
51 and strong base afforded 43 – 45. The 2-arylpyrimidine test compounds were synthesized by
52
53 Suzuki coupling between the 2-chloropyrimidine substrates 43 – 48 and the corresponding

1
2
3 boronic acid or esters which were either purchased or prepared from the corresponding aryl
5
6 bromides using literature methods.19
7
8
9 1-H-benzimidazole test compounds 19 – 30 (Table 4) were made as shown in Scheme 4 and
10
11 5. Cyclopropanation of 49 led to 50 where reaction with 3(R)-methyl morpholine followed by
12
13
14 sodium tungstate catalysed oxidation of the sulfide proceeded well to afford 51 without over-
15
16 oxidation of the pyrimidine ring. Reaction with 1H-benzo[d]imidazol-2-amine or N-methyl-
18
19 1H-benzo[d]imidazol-2-amine led directly to compounds 19 and 21 respectively; compound
20
21
22 19 was derivatized to the N-acetyl 20. Benzimidazoles 22 – 30 were made starting from the 2-
23
24 chloropyrimidine intermediate 47 (Scheme 5). Buchwald-Hartwig coupling with the
26
27 appropriate substituted 2-nitroaniline utilising the xantphos ligand system afforded
28
29
30 intermediates 52 – 60. Reduction of nitro to amino was achieved under indium catalyzed
31
32 transfer hydrogenation conditions or with zinc in acetic acid to give the corresponding anilines
33
34
35 61 – 69 which were then cyclized using cyanogen bromide to afford compounds 22 –
38 The iodobenzyl compound 70 was used as the starting point for the synthesis of sulfoxides
39
40 31, 32 and sulfoximines 2, 33 – 36 (Scheme 6). Displacement of the iodide in 70 with
41
42
43 sodiumthiomethoxide gave sulfide 71 in high yield which was then oxidized to the
44
45 corresponding sulfoxide R/S-72 with sodium metaperiodate. Compounds were either made as
47
48 a mixture of diastereoisomers and separated by chiral chromatography or prepared startingS N Cl

40 R1 = [ O O

41 R1 = [ N

42 R1 = O
[ N

b

R1

Cl43 R1 = [ O O

44 R1 = [ N45 R1 = O

[ N
36 a Reagents: (a) compound 40: (Ph3P)4Pd, 2-(3,6-dihydro-2H-pyran-4-yl)-4,4,5,5-
37
38 tetramethyl-1,3,2-dioxaborolane, Cs2CO3, 1,4-dioxane, water, rt; compounds 41 and 42: 8-oxa-
39 3-azabicyclo[3.2.1]octane or 3-oxa-8-azabicyclo[3.2.1]octane respectively, Et3N, DCM, rt; (b)
40 compound 43: 1,2-dibromoethane, NaH, DMF, 0 °C → rt; compounds 44 and 45: 1,2-
41 dibromoethane, 50% NaOH (aq.), tetraoctylammonium bromide, DCM, rt; (c) compound 7:
43 (Ph3P)2PdCl2, 1H-indol-4-yl boronic acid, Na2CO3 (aq.), 4:1 DME:water, microwave, 110 °C;
44 compounds 17 and 18: (Ph3P)4Pd, 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1H-
45 pyrrolo[2,3-c]pyridine, Na2CO3 (aq.), 1,4-dioxane, 95 °C.
50 Schem14 a Reagents: (a) compound 11: (Ph3P)4Pd, 1H-pyrrolo[2,3-b]pyridine-4-ylboronic acid,

15 Na2CO3 (aq.), 1,4-dioxane, 90 °C; compound 12: 1H-benzo[d]imidazol-2-amine, Na2CO3,
16
17 DMA, 160 °C, microwave; compound 13: (Ph3P)2PdCl2, 1H-pyrrolo[2,3-b]pyridine-4-
18 ylboronic acid, Na2CO3 (aq.), 4:1 DME:water, 110 °C, microwave; compound 14: 2-
19 dicyclohexylphosphino-2′,4′,6′-triisopropylbiphenyl, bis(dibenzylideneacetone)palladium(0),
20
21 4-bromo-1H-pyrrolo[2,3-c]pyridine, KOAc, bis(pinacolato)diboron, dioxane, 100°C followed
22 by compound 46, (Ph3P)4Pd, Na2CO3 (aq.), 100 °C; compound 15: 1,1′-
23 bis(diphenylphosphino)ferrocenedichloropalladium, 4-bromo-1H-pyrrolo[2,3-c]pyridine,
24 KOAc, bis(pinacolato)diboron, dioxane, 95 °C followed by compound 47, (Ph P) Pd, Na CO
25 3 4 2 3
26 (aq.), 95 °C; compound 16: (Ph3P)4Pd, Na2CO3 (aq.), 4-(4,4,5,5-tetramethyl-1,3,2-
27 dioxaborolan-2-yl)-1H-pyrrolo[2,3-c]pyridine, dioxane, 95 °C.
60 ACS Paragon Plus Environment

18 S
19 N N
20 N
21 HN
22 R1

19 R1 = H 51
20 R1 = Ac
21 R1 = Me

31 a Reagents: (a) 1,2-dibromoethane, 50% NaOH (aq), tetraoctylammonium bromide, toluene,
32 60 °C; (b) (R)-3-methylmorpholine, DIPEA, 1,4-dioxane, 80 °C; (c) NaO4W·2H2O, Bu4NHSO4,
33 EtOAc, H2O2, 0 °C → rt; (d) compound 19: 1H-benzo[d]imidazol-2-amine, Cs2CO3, DMA, 110
35 °C, microwave; compound 21: N-methyl-1H-benzo[d]imidazol-2-amine, Cs2CO3, DMA, 90 °C;
36 (e) DMAP, Ac2O, 90 °C.

52 R1 = 3-F

53 R1 = 4-F
54 R1 = 5-F
5 55 R1 = 6-F
6 4 56 R1 = 3-Cl
R1 57 R1 = 3-OMe
1 2

58 R1 = 4-Cl

59 R1 = 4-CN
60 R1 = 4-OMe

61 R1 = 3-F

62 R1 = 4-F
63 R1 = 5-F
64 R1 = 6-F

19 22 – 30

O O N 6

4 65 R1 = 3-Cl
R1 66 R1 = 3-OMe
20 S N

67 R1 = 4-Cl
68 R1 = 4-CN
69 R1 = 4-OMe

26 a Reagents: (a) Substituted 2-nitroaniline, Pd(OAc) , xantphos, Cs CO , 1,4-dioxane, 80 °C,
27 2 2 3
28 microwave; (b) Zn, AcOH, rt or In, NH4Cl (aq.), EtOH, reflux; (c) Cyanogen bromide,

30 a Reagents: (a) NaSMe, DMF, rt; (b) NaIO4, EtOAc, MeOH, H2O, rt; (c) 1,2-dibromoethane,
32 50% NaOH (aq.), tetraoctylammonium bromide, 2-Me-THF, 60 °C; (d) X-Phos 2nd gen.
33 precatalyst, 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1H-pyrrolo[2,3-b]pyridine,
34 Cs2CO3, 1,4-dioxane:H2O (4:1), 90 °C; (e) trifluoroacetamide, iodobenzene diacetate, Rh(OAc)2
36 dimer, MgO, iso-propylacetate, 80 °C; then 7M NH3 in MeOH, rt; (f) NaOH (50% aq.), 1,2-
37 dibromoethane, tetra-octylammonium bromide, mTHF, rt; (g) compound 33: (Ph3P)2PdCl2,
38 2M Na2CO3, 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1-tosyl-1H-pyrrolo[2,3-
39
40 b]pyridine, DME:H2O (4:1), 90 °C then 2M NaOH (aq.), 50 °C; compound 34: 1H-pyrrolo[2,3-
41 b]pyridin-4-ylboronic acid, (Ph3P)2PdCl2, 2M Na2CO3, DME:H2O (4:1), 90 °C; compound 35
42 and 36: Cs2CO3, N-methyl-1H-benzo[d]imidazol-2-amine, DMA, 80 °C.

49
50 Compound 1 is a potent ATR inhibitor with excellent kinase selectivity and free exposure
51
52 and is a useful tool compound to explore ATR pharmacology in vivo.16 However, compound 1
53
54
55 was also found to be a time-dependent inhibitor of CYP3A4 and to suffer from low aqueous

1
2
3 solubility. Inhibition of CYP3A4, particularly mechanism based inhibition, is a concern for
5
6 clinical DDI,22-24 whereas low solubility limits the maximum absorbable dose (Dabs).25 These
7
8
9 properties increase risk of failure in clinical development and compound 1 was therefore not
10
11 considered a suitable candidate for clinical-enabling studies. Eliminating CYP3A4 TDI and
12
13
14 achieving high aqueous solubility at the same time maintaining high ATR potency, excellent
15
16
17 specificity and the attractive pharmacokinetic properties exhibited by 1, were the key
18
19 medicinal-chemistry design goals in the optimization phase.
20
21
22 CYP TDI has been observed as a common feature in kinase inhibitors with CYP3A4 being
23
24 the most commonly inhibited isoform.26 CYP TDI is associated with the formation of covalent
26
27 (or reversible-covalent) adducts to heme or protein following metabolic activation.23, 24 In
28
29
30 addition to the risk of DDIs, formation of reactive metabolites is a causative factor for
31
32 idiosyncratic drug toxicity.23, 27 The risk and impact of clinical DDI will be determined by
33
34
35 overall drug disposition, dose, regimen and target patient population. While compound 1
36
37
38 clearly demonstrated TDI of CYP3A4 when incubated at 10 µM in human microsomes,16 we
39
40 did not fully characterize this activity or model in detail the human PK and predicted clinical
41
42
43 dose to understand the magnitude of the expected clinical DDI. We anticipated ATR inhibitors
44
45 would be combined with cytotoxic and targeted drugs in the clinic. As DDI arising from
47
48 inhibition of CYP3A4 would complicate co-dosing of such agents and many likely co-
49
50
51 medicants,28 we set out to remove this undesirable activity.
52
53 The indole and morpholine groups were thought particularly vulnerable to metabolic
55
56 activation. The susceptibility of cyclic tertiary amines such as morpholine to -carbon

1
2
3 oxidation, generating reactive iminium ion intermediates is well described.29-31 Indoles are
5
6 known to undergo ring hydroxylation particularly at C-5 and/or C-6 positions;32 these species
7
8
9 could arise via reactive epoxides and could also lead to quinone-like reactive intermediates
10
11 following additional bioactivation. Oxidation of indole has also been observed at C-2 or C-3,
12
13
14 presumably via the corresponding epoxide, and further oxidation and/or oxidative ring
15
16
17 cleavage can lead to anthranilic acid products. The putative metabolic vulnerability of
18
19 morpholine and indole presented us with a potentially insoluble problem as our previous work
20
21
22 clearly demonstrated the importance of both groups to ATR potency.16
23
24 Reactive metabolites formed from bioactivation are generally electrophilic in character and
26
27 highly unstable. Trapping experiments can be used to detect and characterize metabolites
28
29
30 whereby a compound is incubated in human liver microsomal preparations and any reactive
31
32 metabolites generated are trapped by specific added nucleophiles. Orthogonal nucleophiles are
33
34
35 used to trap the different electrophilic species arising from bioactivation of chemical
36
37
38 substrates. The nucleophiles commonly used are glutathione (GSH), a soft nucleophile efficient
39
40 at trapping soft electrophiles such as epoxides, cyanide to trap iminium species arising from
41
42
43 oxidation of tertiary amines and methoxyamine to effectively trap aldehyde products as Schiff
44
45 bases.33 These screens provide valuable mechanistic information to support rational medicinal
47
48 chemistry design. When 1 was incubated with human liver microsomes, adducts with GSH
49
50
51 but not cyanide were detected. Whilst the mechanisms of bioactivation and TDI may not
52
53 necessarily overlay, the formation of GSH adducts implicates the indole group, likely through
55
56 ring oxidation.
3 We had systematically explored the structure activity relationships (SAR) in the
5
6 morpholino-pyrimidine pharmacophore, varying each of the substituents on the pyrimidine
7
8
9 core. These compounds now allowed facile investigation into the molecular features in
11 responsible for CYP3A4 TDI, independent of ATR potency.41 aUncertainty (95%confidence) for pIC50 measurements is 0.38 (2.4-fold) based on an average

42
43 of two repeat occasions per compound. b Uncertainty (95% confidence) for pIC50
44 measurements is 0.7 (4.6-fold) based on at least two repeat measurements per compound
45 (median number of repeats across tables is 3). The data presented is from a variation of the
46 ATR cell assay described in Ref. 16(see experimental section). A good correlation is observed
48 between ATR cell assay versions; N=41 compounds from morpholino-pyrimidine series,
49 correlation: 0.94, mean difference: -0.084 pIC50. *Compound 1 IC50 = 0.050 µM in the earlier
50 version reported.16 c LogD7.4 assay: lipophilicity was determined using the ‘shake-flask’
52 method. Plated aliquots of sample are dried down and octanol and water are added. Sample
53 content of the phases is determined by LC/MS/MS after stirring, equilibration, and separation
54 of phases by centrifuge. Full description of the assay can be found in the Supplementary
56 Materials. An excellent correlation is observed between versions of the LogD7.4 methods; N=23

1
2
3 compounds from the morpholino-pyrimidine series, correlation: 0.96, mean difference: -0.075.
5 #Compound 1 LogD7.4 = 2.7 using the earlier methodology (Ref. 16). ND = not determined. d
6 Mean value (N≥2) unless otherwise stated. Compounds were pre-incubated at 10 μM with
7 human liver microsomes (1 mg/mL) with and without NADPH (5 mM) for 30 min at 37 °C
8
9 followed by 15 min incubation with 10 μM midazolam; analysis of 1-hydroxymidazolam was
10 performed using liquid chromatography-tandem mass spectrometry.34 No activity detected vs.
11 control for 1A2, 2C19, 2C9, and 2D6. † Result for compound 12 was just above background
12
13 level in test 1 and below background level (<11%) in test 2; result for test 1 shown.
14
15
16
17 The unsubstituted and 3(R)-methyl substituted morpholines (compounds 5 and 1
18
19 respectively) both show clear CYP3A4 TDI activity when incubated at 10 µM in human liver
21
22 microsomes using a standard liquid chromatography-tandem mass spectrometric endpoint
23
24
25 (Table 1).34 Introducing structural architecture to eliminate reactive moieties resulting from
26
27 oxidation on the morpholine ring, for example by modification of the methylene adjacent to
28
29
30 the oxygen atom using the bridged morpholine 6 or removal of the nitrogen heteroatom by
31
32 substitution of morpholine for the 3,6-dihydro-2H-pyran 7, did not eliminate CYP3A4 TDI
34
35 activity compared with morpholine. The dihydropyran and bridged morpholines, both of
36
37
38 which have been described as morpholine isosteres in mTOR inhibitor series,35, 36 are
39
40 equivalent in ATR potency to unsubstituted morpholine but at the price of higher
42
43 lipophilicity. The impact of blocking and/or deactivating substituents on the indole ring was
44
45
46 investigated. Simple ring substituents that retain ATR potency, for example the 6-fluoro indole
47
48 8, did not reduce CYP3A4 TDI. In contrast, addition of a polar and deactivating group such as
49
50
51 acetamido 9 and replacement of the indole, for example with benzimidazole 10, which has
52
53
54 equivalent lipophilicity to the indole 5, led to undetectable CYP3A4 TDI activity. In both cases

1
2
3 ATR potency was also significantly reduced but these results supported the notion that the
5
6 indole ring was the likely key contributor to CYP3A4 TDI. We discovered the 7-azaindole
7
8
9 (1H-pyrrolo[2,3-b]pyridine) 11 retains the ATR potency of the indole while effectively
10
11 eliminating CYP3A4 TDI. In addition, through a wider campaign to identify ATR-active
12
13
14 indole isosteres, the 2-amino-N1-substituted benzimidazole 12 was found to have no or weak
15
16
17 detectable 3A4 TDI. This variant was also found to possess superior cellular potency without
18
19 increasing lipophilicity compared with the indole 5. Unknown to us at the time, Safina et al.37
20
21
22 had developed a series of PI3K-specific morpholinopyrimidines substituted with 4-indole
23
24 that were also found to be potent CYP3A4 TDIs and further determined that the 4-indole
26
27 group was associated with this activity. In contrast to our own findings, replacement of the
28
29 indole with 7-azaindole in the PI3K series did not attenuate CYP3A4 TDI. The apparent
31
32 conflicting results are a reminder that metabolic activation is complex, and SAR may not be
33
34
35 simplified to contributions of individual functional groups. It is whole-molecule structure and
36
37
38 properties that determine metabolic fate. However, in our efforts towards identifying an ATR
39
40 inhibitor candidate for human studies, both the 7-azaindole 11 and 2-aminobenzimidazole 12
41
42
43 became productive leads for optimization.
44
45 In the 7-azaindole series, the SAR was found to mirror that determined for indole with
47
48 cellular potency increasing for the corresponding 3-(R) methyl morpholine 13 (Table 2).
49
50
51 Pleasingly we could not detect CYP3A4 TDI activity for this compound. Measured
52
53 lipophilicity for 13 is unchanged in comparison to 1 and unsurprisingly solubility remains low.
0.005 0.085 1.8 5.3 130 <15 0.002 0.012 2.1 5.8 108 <20

ND 0.36 2.2 4.2 16ND 0.086 2.2 4.9 >1510 ND ND 0.26 2.1 4.5 98 ND
43 a,b For assay uncertainty, see footnotes in Table 1 c Lipophilicity ligand efficiency (LLE): ATR
45 cell pIC50 – LogD7.4. d Solid material was agitated in 0.1 M pH 7.4 phosphate buffer for 24 h,
46 double centrifuged, and the supernatant analyzed for compound concentration by
47
48 LC−UV−MS. Crystallinity assessed by polarized light microscopy of remaining solid. Full
49 description is provided in Supplementary Materials. Lowest solubility given in the table where
50 multiple measurements were taken: N≥3 for compounds 1, 13, 15; N=1 for compounds 14, 16,
51 17, 18. ND: not determined.
3 Substitution with the 6-azaindole (1H-pyrrolo[2,3-c]pyridine) isomer surprisingly resulted in
5
6 compounds with much improved cellular potency compared with the corresponding indole
7
8
9 and 7-azaindole. Moreover the 6-azaindole series combines improved potency with reduced
10
11 lipophilicity and improved aqueous solubility (compounds 14 – 18, Table 2; note: the very high
12
13
14 solubility result for compound 17 is most likely an outlier that we speculate is due to low
crystallinity) while also having no detectable CYP3A4 TDI. It is interesting to observe a
18
19 seemingly small structural change delivering such a significant effect on both potency and
20
21
22 physicochemical properties.38 Directionality of the hydrogen bond acceptor nitrogen in the
23
24 azaindole isomers is changed relative to the morpholine hinge binder offering the potential to
26
27 impact key interactions to the protein. Recent structures of ATR-mimicking PI3K mutants,39
28
29
30 and an ATR cryo-EM,40 structure suggest that the difference in potency between the
31
32 azaindoles isomers can be explained by their interaction with Asp2335, potentially mediated
33
34
35 by a water molecule. 6-Azaindole is also considerably more basic than 7-azaindole (pKa 7.9
36
37
38 and 4.6 respectively41) resulting in an expected change in ionization state at physiological pH
39
40 and concomitant effects on LogD7.4 and solubility. The unsubstituted morpholine 7-azaindole
41
42
43 14 demonstrated excellent ATR cellular potency (IC50 <100 nM) and again the established
44
45 morpholine SAR translated, with a significant increase in potency observed when 3-(R)
47
48 methyl morpholine was employed over the 3-(S) methyl isomer (compounds 15 and
51 respectively). The bridged morpholines, compounds 17 and 18, retained potency compared to
52
53 morpholine 14 and the reduced lipophilicity provided by the 6-azaindole group led to
55
56 compounds with a good overall balance of properties. It is noticeable from the compounds

3 shown in Table 2 that the hitherto excellent ATR enzyme to cell correlation breaks down. We
5
6 had high confidence in the cellular assay measuring inhibition of CHK1 phosphorylation, a
7
8
9 direct substrate of ATR, in response to a DNA-damage stimulus. Presumably, with the more
10
11 potent compounds, the IC50 is below the detection threshold (tight binding limit) of the

14 enzyme assay. Therefore, at this stage of the optimization phase the cellular assay was
15
16
17 primarily used to drive chemistry in conjunction with lipophilicity and physicochemical
18
19 driven properties of aqueous solubility, permeability and metabolic stability. The 3-(R)-methyl

22 morpholine 6-azaindole 15 stood out in combining excellent cellular potency and moderate
23
24 lipophilicity (LLE = 5.8) with improved aqueous solubility and no measureable CYP3A4 TDI
26
27 and represented a significant advancement over the indole lead 1.
28
29
30 The benzimidazole head group utilized in compound 12, substituted at the 2-position with
31
32 amino or alkyl substituents on a morpholino-pyrimidine core was described in a series of anti-
33
34
35 tumor agents with PI3K activity from Zenyaku Kogyo.42 We discovered the novel
36
37
38 methylsulfonylmethyl morpholino-pyrimidine 12 to possess excellent ATR potency (Table 1)
39
40 albeit with some class 1 PI3K inhibitory activity (e.g. PI3K IC50 = 0.24 µM). Substitution on
41
42
43 the morpholine hinge binder has been shown to affect PI3K activity,16, 36 thus the 3(R)-methyl
44
45 morpholine 2-aminobenzimidazole (compound 19, Table 3) became a key compound to make
47
48 that was subsequently shown to possess greatly improved ATR potency and selectivity over
49
50
51 PI3K (ATR cell IC50 = 0.015 µM, PI3K IC50 = 9 µM).

56 Table 3. Benzimidazole C2 SAR
LLE Solubi CYP3A4 %

24 a,b For assay uncertainty, see footnotes in Table 1 c Lowest solubility result is given in the
25 table where multiple measurements were taken: N=2 for compound 21; N=1 for compounds
26 19, 20. † Result for compound 19 was just above background level in test 1 and below
27
28 background level (<17%) in test 2; result shown for test 1.
29
30
31
32 The potency improvement relative to indole is achieved without increasing lipophilicity;
33
34
35 compound 19 shows borderline CYP3A4 TDI and aqueous solubility is not significantly
36
37
38 improved compared to indole 1. Substitution on the amino group with acetyl (compound 20)
39
40 or methyl (compound 21) were well tolerated with no loss of cellular potency and in the case
41
42
43 of the N-acetyl 20, led to higher LLE. Moreover, we did not detect CYP3A4 TDI for either of
44
45 the substituted 2-amino-benzimidazole compounds. While a relatively modest improvement
47
48 in aqueous solubility was perhaps evident, particularly for N-methyl 21, aqueous solubility was
51 not robustly improved compared with indole.
52
53 Substitution on the benzimidazole aryl ring was explored for effects on potency and to
55
56 address a theoretical concern that the naked aryl ring might be susceptible to metabolic

1
2
3 instability. The result of systematic substitution on the aryl ring of the benzimidazole group
5
6 with fluorine is shown in Table 4.ATR cell

38 a,b See footnotes a,b in Table 1 for assay uncertainty; repeat measurements: N=2 for
39
40 compound 19; N=1 for compounds 22 – 30.
41
42
43
44 A 5-F substituent (compound 23) retains the same level of ATR potency compared with
47 unsubstituted benzimidazole 19. While 4-F and 6-F (compounds 22 and 24 respectively) retain
48
49 a degree of ATR potency the 7-F analogue (compound 25) is significantly less potent. A broader
50
51
52 range of substituents in positions 4- and 5- (exemplified by compounds 26 – 30, Table 4) were
53
54

3 then made to probe into the protein further but for all these examples reduced ATR potency
5
6 was seen.
9 In the solid state, methylsulfone compounds such as 1 display a centrosymmetric
10
11 methylsulfone to methylsulfone contact (in addition to ring-ring stacking and a hydrogen
12
13
14 bonding network between indole N-H and sulfone oxygen) associated with high melting
15
16 points and low solubility.16 Our SAR study of the indole series demonstrated that specific
18
19 changes around the methyl sulfonyl moiety (e.g. addition of charged substituents) could
20
21
22 improve physicochemical properties, but these changes ultimately led to reduced potency
23
24 and/or attenuation of other fundamental properties such as permeability leading to low oral
26
27 exposure. However, we had yet to attempt specific changes to the sulfone group, for example
28
29
30 replacement with sulfoxide or sulfoximine whereby one of the oxygen atoms is replaced with
31
32 nitrogen. We hypothesized that such changes could disrupt the observed solid-state contacts
33
34
35 albeit at the complication of introducing additional chirality and with uncertain

40 result is given in the table where multiple measurements were taken: N≥2 for compound 13,
41 2, 33, 35; N=1 for compounds 31, 32, 34, 36.
42
43
44
45
46 The diastereomeric sulfoxides in the 7-azaindole series, compounds 31 and 32 (Table 5), were
47
48 found to retain ATR potency compared with the corresponding sulfone 13 with no change in
50
51 measured lipophilicity. These sulfoxides were also found to have high aqueous solubility
52
53
54 though it proved challenging to generate stable crystalline forms and therefore the apparent
provement compared with the corresponding sulfones was treated with caution. The
5
6 sulfoxides are highly permeable and have high exposure from oral doses in rodents. However,
7
8
9 the sulfoxide group is prone to oxidation in vivo and a significant level of the corresponding
10
11 sulfone was observed in rodent PK studies; for this reason, despite other properties being
12
13
14 generally attractive, we discounted the sulfoxides from further progression. The sulfoximines
15
16
17 (compounds 2, 33 – 36 Table 5) achieved both of the key aims we set at the start of the lead
18
19 optimization campaign. The R-stereochemistry of the sulfoximine 2 was initially inferred from
20
21
22 the stereochemistry of the sulfoxide precursor, with imination to the sulfoximine proceeding
23
24 with retention of configuration, and later confirmed from an X-ray structure
26
27 Interestingly, hydrogen bonds are observed only between the azaindole substituents of
28
29
30 adjacent molecules for 2, whereas the indole NH donor in the structure of 1 does not form a
31
32 hydrogen bond. The centrosymmetric packing characteristic for methylsulfones as seen for 1
s prevented in the structure of 2 by introduction of the sulfoximine.. Kinome selectivity depiction for compound 2. Inhibition data (%) shown for a6 compound test concentration of 1 M. PI3K isoforms indicated as α, β, γ, δ. Data in

7
8
9 Supplementary Materials, Table T1.
16 Table 7. PI3K, PIKK-family selectivity and growth inhibitory potency for preclinical candidate
17
18 shortlist compared with lead 1
35 a See footnote of Table 1. b Standard error of mean (SEM) pIC50 measurement is ≤0.13. c
36 Inhibition of AKT pSer473 in MDA-MB-468 cells. d Inhibition of p70S6K pSer235/236 in
38 MDAMB-468 cells. e Inhibition of pAKT T308 in BT-474 cells. f Inhibition of ATM Ser1981 in
39 HT-29 cells following IR treatment. g Inhibition of DNA-PK pSer2056 in HT-29 cells following
40 IR treatment. h MTS (tetrazolium dye) assay with 72 h continuous exposure to compounds.
42
43
44
45 The 2-aminobenzimidazole 21 and 6-azaindole sulfone 15 also display excellent selectivity
46
47
48 over all kinase classes but have a degree of promiscuity versus lipid kinases. It is interesting to
49
50 compare kinase selectivity with the aminopyrazine 3 which exhibits a wholly different profile,
52
53 something that is not unexpected given that it belongs to a different structural class (see
54
55
56 Supplementary Materials S1). The aminopyrazine 3 has been reported to have high ATR

1
2
3 specificity particularly over the closely related mTOR.48 The morpholinopyrimidine inhibitors
5
6 were also screened specifically for inhibition of the related targets mTOR, PI3K, ATM and
7
8
9 DNA-PK. These compounds all show moderate inhibitory potency, relative to ATR, against
10
11 mTOR in an enzyme assay and some activity in cell assays reading out mTORC1 (inhibition of
12
13
14 AKT pSer473) and mTORC2 (inhibition of p70S6K pSer235/236). This is unsurprising as the
15
16
17 PI3K and PIKK kinases are known to have similar binding sites, with some compounds such
18
19 as NVP-BEZ235,6 exhibiting a promiscuous pan-PI3K and -PIKK profile. In terms of protein
20
21
22 sequence similarity, mTOR and ATM are nearest neighbors of ATR, yet, a margin of activity
23
24 was observed in all cases including mTOR, relative to inhibition of ATR-dependent kinase
26
27 signaling. Moreover, no activity could be detected for PI3K, ATM or DNA-PK in cell-based
28
29
30 systems. These combined data suggest the optimized compounds are unlikely to have activity
31
32 against other PI3K/PIKK signaling pathways at relevant doses. LoVo are MRE11A-mutant
33
34
35 (MRE11A is key component of the ATM signaling and DNA double-strand break (DSB) repair
36
37
38 pathway) colorectal adenocarcinoma cells which are sensitive to ATR inhibitors.16 Compound
39
40 1 was shown to induce S-phase arrest, an increase in γH2AX over time and caspase-3 activation
41
42
43 and cell death.49 HT29 colorectal adenocarcinoma cells are classified as MRE11A and ATM-
44
45 proficient expressing high levels of total ATM protein without an ATM pathway defect and
47
48 therefore expected to be relatively insensitive to selective ATR inhibition. As can be seen in
49
50
51 Table 7, the morpholinopyrimidine ATR inhibitors show greater growth inhibition in LoVo
52
53 compared with HT29 in support of this general hypothesis. Across a broader cell panel, ATR
55
56 inhibitors from structurally orthogonal series show an inhibition profile that is distinct from

1
2
3 PI3K and PIKK-family inhibitors, further supporting a cellular mode of action arising from
5
6 selective ATR inhibition
Colon and gastric tumor cell line responses for ATR inhibitors 1, 2, 3 compared with:
26
27
28 NVPBEZ-235 (labelled pPI3/mTORi)6 (mTOR, PI3K, ATR, ATM, DNA-PK), AZD818650
29
30 (labelled PI3Kβ+δ) and AZD883551 (labelled PI3K α+δ). Data shown is normalised as pGI50
32
33 minus mean pGI50 across the panel to correct for the influence of absolute potency.
34
35
36 Hierarchical clustering of profiles is shown on the right.
42 The compounds were next characterized for tumor growth inhibition (TGI) in vivo.
43
44
45 Compounds were first administered at their maximum well-tolerated daily dose by oral gavage
46
47 to female nude mice bearing human LoVo colorectal adenocarcinoma xenografts. Mouse
48
49
50 tolerance of the morpholinopyrimidine ATR inhibitors was found to be variable. Indole
51
52 sulfone 1 and the sulfoximines 2 and 35 were well tolerated at 50 mg/kg once daily (QD)
54
55 whereas the 6-azaindole sulfone 15 and 2-methylaminobenzimidazole sulfone 21 were

1
2
3 tolerated at a maximum daily dose of 25 mg/kg QD. From a mechanistic standpoint and as a
5
6 monotherapy, we expected continuous exposure would be required to drive efficacy.16 A broad
7
8
9 relationship can be seen between the observed tolerance, efficacy and compound exposure in
10
11 mouse relative to potency.
12
13
14
15
16 Table 8. Monotherapy in vivo tumor growth inhibition (TGI) in human LoVo colorectal
18
19adenocarcinoma xenografts.
Schedule TGIa (p-value T-test vs.
vehicle control)

49 a Female nude mice bearing established human LoVo xenografts were dosed orally with
51 compound at the indicated dose and schedule. ns = not significant
52
53
54
55
56
3 Table 9. Plasma concentration at 8 hours following multiple doses in nude mouse at compound
5
6 maximum well tolerated dose.
9 Mean 8 h plasma

10 Compound Dose
11 (mg/kg)
12Dosing
frequency
concentration, µM (Std Dev)a
Free plasma
LoVo multiple

29 a Female nude mice bearing established human LoVo colorectal adenocarcinoma xenografts

30 were dosed orally with compounds either once-daily (QD) [compound 1: averaged data from
31 3 {N=10, mean=4.2 µM (Std Dev=1.3)}, 4 {N=10, 3.1 (1.1)} and 14 {N=10, 3.2 (0.77)} consecutive
32
33 doses; compound 2: 75 mg/kg: averaged data from 4 doses, 50 mg/kg: averaged data from 4
34 {N=5, 2.1 (0.65)}, and 14 {N=10, 2.3 (0.49)}, consecutive doses; compound 15: 4 consecutive
35 doses (N=10 independent samples); compound 21: 4 consecutive doses (N=5 independent
36 samples); compound 35: 14 consecutive doses (N=10 independent samples)] or twice-daily for
38 8 consecutive doses. Plasma was sampled at 8 hours following the last dose.
31  Exposure of ATR inhibitors 1, 2, 15, 21, and 35 is correlated to tumor growth
32
33
34 inhibition (TGI). The observed plasma concentrations following a single dose of each
35
36 compound were multiplied by the compound specific in vitro measured free fraction and
38
39 divided by in vitro GI50 to give fold free concentration above GI50. The time above in vitro
40
41
42 LoVo GI50 is plotted against LoVo xenograft tumor growth inhibition in vivo. A logarithmic
43
44 trendline (Log.(All)) best-fit curve is shown for all compounds.
45
46
47
48
49
50 All compounds other than compound 35, achieved TGI >50% in the LoVo model when dosed
52
53 once daily over the course of the study (Table 8). The comparative lack of efficacy for
54
55
56 compound 35 can be explained by the lowest free plasma concentration multiple over LoVo

1
2
3 GI50 measured at 8 h of all the compounds tested (Table 9). The 7-azaindole sulfoximine 2
5
6 showed the greatest TGI, equivalent or greater at 50 mg/kg across multiple experiments, to the
7
8
9 indole sulfone lead 1 and this is associated with free plasma concentration at 8 h in excess of
10
11 the LoVo GI50. The total plasma concentration of the 6-azaindolesulfone 15 was found to be
12
13
14 the lowest of the compounds tested but the combination of high unbound fraction with high
15
16
17 potency results in a high free plasma multiple over LoVo GI50. In contrast, benzimidazole
18
19 sulfone 21 shows high total drug plasma concentration and high potency but has a relatively
20
21
22 high bound fraction, particularly in comparison to compounds 2 and 15, and this leads to a free
23
24 plasma multiple over LoVo GI50 in a similar range to the other compounds. Twice-daily dosing
26
27 was investigated for compounds 1, 2 and 15 in an attempt to achieve longer exposure and drive
28
29
30 a greater tumor response. When the maximum well-tolerated daily single dose was split (dosed
31
32 8 hour apart), neither compound 15 or 1 showed greater anti-tumor activity. For indole sulfone
33
34
35 1, this may be explained by a relatively flat PK profile in the mouse negating the impact of
36
37
38 twice daily dosing (BD).16 The 6-azaindole sulfone 15 has a relatively short half-life in mouse
39
40 and we expected BD dosing would lead to greater efficacy. However, the efficacy achieved for
41
42
43 compound 15 dosed BD was indistinguishable from the higher single dose. In contrast, 7-
44
45 azaindole sulfoximine 2 achieved near complete TGI in the LoVo xenograft model when
47
48 administered at a dose of 25 mg/kg BD. The time each day that free concentrations in plasma
49
50
51 were above the in vitro LoVo GI50 was estimated using the observed plasma concentrations
52
53 following a single dose of each compound. This duration exhibits a saturating relationship
55
56 with tumor growth inhibition . The relationship is consistent across compounds

1
2
3 regardless of differences in pharmacokinetic properties, including different terminal half-lives
5
6 in the mouse. This analysis broadly correlates with the cover seen in earlier experiments, with
7
8
9 15 showing the largest difference. A clear dose-response could be demonstrated for 2 in LoVo
10
11 (Table 8 and , top graph) and this was compared with HT29 ( bottom graph).
12
13
14 Compound 2 delivers significant TGI in LoVo at doses as low as 25 mg/kg QD or 12.5 mg/kg
15
16
17 BD. A dose of 75 mg/kg QD leads to regression in the LoVo model albeit tolerability is
18
19 borderline with 4 of 10 animals in the group terminated in accord with study protocol due to
20
21
22 bodyweight loss greater than 15%. In the remaining six animals a maximum bodyweight loss
23
24 of 9% was observed. Therefore 75 mg/kg, while formerly tolerated, was not considered to be
26
27 a well-tolerated dose. The observed in vitro sensitivity (Table 7) translated in vivo with no
28
29
30 significant anti-tumor efficacy observed for 2 in HT29 using doses and schedules which are
31
32 highly active in LoVo xenografts ( bottom graph). γH2AX is a sensitive marker for
33
34
35 DNA damage and a useful marker to study ATR inhibition. Increases in γH2AX reflect the
36
37
38 time- dependent accumulation of collapsed replication forks, which only occur in actively
39
40 replicating cells during S-phase of the cell cycle), and replication-associated DSBs.52, 53 In LoVo
41
42
43 xenografts, the magnitude and maintenance of γH2AX over 24 hours is obtained in a dose-
44
45 dependent manner after repeat daily dosing with the sulfoximine 2  and this is
47
48 associated with the greater anti-tumor effect observed for this compound. We observe an
49
50
51 indirect relationship between plasma PK and tumor PD based on γH2AX induction, with
52
53 signals being sustained beyond 24 hours despite plasma concentrations predicted to be below
55
56 detectable levels at this timepoint (LOQ 0.09 μM data not shown). While we need sufficient

1
2
3 levels and duration of cover to induce DNA breaks and γH2AX, once these have formed it may
5
6 take many hours for the damage to be repaired, or cells to die, and the γH2AX signal to
7
8
9 dissipate. Persistence of γH2AX signal over time (after the damage insult) is indicative of
10
11 unrepaired DSBs and/or DNA repair inhibition and is observed even after breaks have been
12
13
14 repaired.52
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49 . In vivo tumor growth inhibition (TGI) for compound 2. Top graph: female nude mice
50
51 bearing established human LoVo (MRE11A mutant/ATM deficient) colorectal
53
54 adenocarcinoma xenografts were dosed orally with either vehicle (♦) or 2 at 10 mg/kg once

1
2
3 daily (