Optimization of a Series of Bivalent Triazolopyridazine Based

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Optimisation of a series of bivalent triazolopyridazine based bromodomain and extraterminal inhibitors: the discovery of (3R)-4-[2-[4-[1-(3-methoxy-[1,2,4]triazolo[4,3-b]pyridazin-6-yl)-4piperidyl]phenoxy]ethyl]-1,3-dimethyl-piperazin-2-one (AZD5153) Robert H Bradbury, Rowena Callis, Gregory R. Carr, Huawei Chen, Edwin Clark, Lyman Feron, Steven C. Glossop, Mark A. Graham, Maureen Hattersley, Chris Jones, Scott G. Lamont, Gilles Ouvry, Anil Patel, Joe Patel, Alfred A. Rabow, Craig A. Roberts, Stephen Stokes, Natalie Stratton, Graeme E. Walker, Lara Ward, David Whalley, David Whittaker, Gail Wrigley, and Michael J. Waring J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00070 • Publication Date (Web): 15 Aug 2016 Downloaded from http://pubs.acs.org on August 16, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Medicinal Chemistry is 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.

Page 1 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Optimisation of a series of bivalent triazolopyridazine based bromodomain and extraterminal inhibitors: the discovery of (3R)-4[2-[4-[1-(3-methoxy-[1,2,4]triazolo[4,3-b]pyridazin-6-yl)-4piperidyl]phenoxy]ethyl]-1,3-dimethyl-piperazin-2-one (AZD5153)

Robert H. Bradbury1, Rowena Callis1, Gregory R. Carr1, Huawei Chen2, Edwin Clark2, Lyman Feron1, Steve Glossop1, Mark A. Graham1, Maureen Hattersley2, Chris Jones1, Scott G. Lamont1, Gilles Ouvry3, Anil Patel1, Joe Patel2, Alfred A. Rabow1, Craig A. Roberts1, Stephen Stokes1, Natalie Stratton1, Graeme E. Walker1, Lara Ward1, David Whalley1, David Whittaker1, Gail Wrigley1 and Michael J. Waring4*.

1

AstraZeneca, Mereside, Alderley Park, Macclesfield, Cheshire, SK10 4TG, UK.

2

AstraZeneca, Gatehouse Park, Waltham, MA 02451, USA.

3

AstraZeneca, Chemin du Moulin de Vrilly, 51100 Reims, France.

4

Northern Institute for Cancer Research, School of Chemistry, Newcastle University, Bedson

Building, Newcastle upon Tyne, NE1 7RU, UK.

Abstract Here we report the discovery and optimization of a series of bivalent bromodomain and extraterminal inhibitors. Starting with the observation of BRD4 activity of compounds from a previous programme, the compounds were optimised for BRD4 potency and physical properties. The optimised compound from this campaign exhibited excellent pharmacokinetic profile and exhibited high potency in vitro and in vivo effecting c-Myc downregulation and tumour growth inhibition in xenograft studies. This compound was selected as the development candidate AZD5153. The series showed enhanced potency as a result of bivalent binding and a clear correlation between BRD4 activity and cellular potency.

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

Introduction The bromodomains (BRD) are a family of proteins responsible for binding acetylated lysine residues of proteins. The BRDs are rapidly emerging as a target class amenable to pharmacological intervention with small molecules1. The bromodomain and extraterminal (BET) sub-family are perhaps the most widely explored of the class2. Initially identified as a result of phenotypic screening for modulation of ApoA1, subsequent target identification in tandem with optimisation has revealed multiple small molecule inhibitors of BET bromodomains, which have progressed to preclinical and clinical evaluation (Figure 1)3-5. The BET family consists of four proteins, termed BRD2, BRD3, BRD4 and BRDT. Each of these proteins contains two separate bromodomains. BRD4 in particular has been lauded as a potential drug target6 although it is worth remembering that there is a high degree of homology within this class and the majority of compounds, particularly those studied most extensively in biological elucidation, do not exhibit selectivity within the class. The BET family regulate gene expression through binding of acetylated lysine residues on histone proteins followed by activation of transcription elongation driven by RNA-PolII. BET inhibitors inhibit this gene transcription and, as a consequence, exhibit growth inhibition in a number of haematological and solid tumour models. This action is believed, at least in part, to be due to downregulation of critical oncogenes such as c-Myc, leading to great interest in BET inhibitors in the oncology field and a number are currently progressing in early clinical trials. From the initial ApoA1 screen, pioneering work at GlaxoSmithkline led to the discovery of the triazolo-benzodiazepine 1 (iBET-762, Figure 1)3,4, which has become one of the leading clinical BET inhibitors. Work carried out by Bradner and the Structural Genomics Consortium identified the structurally-related compound 2 (JQ1)5. 1 and 2 have become key standard compounds in the field with many studies using these compounds as probes to understand BET biology. More recently, OTX015, 37, the corresponding p-hydroxyanilide derivative of 2 has been reported to show encouraging clinical results in AML patients8. We have recently described the discovery that compounds from the androgen receptor (AR) downregulator (DR) programme9,10 were found to be potent BET inhibitors that are demonstrated to


Page 2 of 60

Page 3 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


act through an in-cis bivalent binding mode in which a single inhibitor molecule simultaneously engages both bromodomains of a single protein via two acetyl lysine mimicking moieties11. This activity is postulated to explain the effects of the compounds observed in AR and estrogen receptor (ER) DR assays as well as their effects on c-Myc levels. Here we describe the medicinal chemistry programme that led to the identification of these compounds and their optimization, leading to the discovery of a clinical candidate bivalent BET inhibitor.

1 i-BET762

2 JQ1

3 OTX-015

Figure 1

Results and Discussion The campaign of optimisation was initiated with the objective of improving the potency of the previous clinical AR downregulator AZD3514, 4, and the corresponding unsaturated analogue 5 whilst maintaining their excellent ADMET properties9. These compounds exhibited modest cellular AR DR potencies (pIC50 5.8 and 6.5 for 4 and 5 respectively) and subsequent chemistry was focused on cellular AR DR potency to guide the chemistry campaign. The BET inhibitory activity of the compounds and resulting effects on ER and c-Myc DR were discovered during the course of this work11. Later chemistry focused on the BRD4 potency and ER DR data to develop SAR and was generated retrospectively on the earlier compounds. The AR DR, ER DR and c-Myc DR data (measured in LNCaP, MCF7 and MM1.S cell lines respectively) are tightly correlated and can be used reasonably interchangeably to interpret the SAR. The data from these assays are also predictive of growth inhibitory effects measured in MM1.S cells (see later and Supporting Information for more detail). In order to assess the effect of the bivalent binding, an assay employing a protein construct containing both bromodomains was required. Hence, a fluorescence polarisation (FP) assay for full



1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 60

tandem domain BRD4 protein (residues 44-460) was developed using a fluorescently labelled small molecule probe derived from 2. It should be noted throughout, that whilst data for BRD4 are quoted, no selectivity of these molecules within the BET family is anticipated due to their close homology.

5

4, AZD3514

AR downregulation pIC50 6.5

AR downregulation pIC50 5.8

Figure 2

The synthetic route that could be used to prepare analogues with modifications to the acylpiperazine is exemplified with the synthesis of 5 (Scheme 1). SNAr addition of 4-(4-hydroxyphenyl)piperidine to chlorotriazolopyridazine afforded the corresponding phenol (Scheme 1). The phenol was alkylated with ethylene carbonate to give the ethyl alcohol, which was derivatised as the mesylate and this was used to alkylate the required amine, in this case acetylpiperazine, to give 5 in 35% overall yield for the four steps. Compounds 6 and 7 were prepared in an analogous manner.

Scheme 1 – Synthesis of 5. Compounds 6 and 7 were prepared in an analogous manner.

ii

i

iii

iv

5

(i) DIPEA, MeCN, rt, 66%. (ii) Ethylene carbonate, K2CO3, DMF, 80 °C, 72%. (iii) Mesyl chloride, Et3N, DCM, 0 °C, 100%. (iv) N-acetylpiperazine, DIPEA, DMA, 110 °C, 73%.


Page 5 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


Alternatively, the amine substituent could be introduced before the triazolopyridazine allowing more effcient synthesis of compounds with different triazolopyridazine substituents. This is exemplified by the synthesis of 8 (Scheme 2). BOC protection of 4-(4-hydroxyphenyl)piperidine was followed by alkylation of the phenol with 1-bromo-2-chloroethane to give the alkyl halide. This underwent displacement with R-1,3-dimethylpiperazin-2-one to afford the BOC protected intermediate. Removal of the BOC group with ethereal HCl afforded the piperazine, which could be arylated with a variety of substituted chlorotriazolopyridazines, in this case 6-chloro-3-(trifluoromethyl)[1,2,4]triazolo[4,3-b]pyridazine to give the final compound 8 in five steps and 52% overall yield. Compounds 9 to 20 were prepared in a similar fashion.

Scheme 2 – Synthesis of 8. Compounds 9 to 20 were prepared in a similar fashion.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

i

Page 6 of 60

ii

iii

iv

v

8

(i) Et3N, BOC2O, DCM, 0 °C, 88%. (ii) 1-Bromo-2-chloroethane, K2CO3, acetone, 90 °C, 97%. (iii) DIPEA, KI, DMA, 110 °C, 90%. (iv) HCl, Et2O, quant. (v) DIPEA, DMA, 80 °C, 68%.

Analogues with variations to the ethoxy linker of the piperazinone were prepared by initial alkylation of the piperazinone by reductive amination with an appropriate ketone, exemplified with N- BOC azetidine-3-one (Scheme 3). Removal of the BOC group with trifluoroacetic acid gave the intermediate amine. Arylation of 4-(4-bromophenyl)piperidine with 6-methoxy-3-(trifluoromethyl)[1,2,4]triazolo[4,3-b]pyridazine gave the arylbromide, which was coupled with the amine under Buchwald-Hartwig conditions to give the final compound 22 (10% overall yield for the lowest yielding sequence). Compounds 23 and 24 were prepared in an analogous manner.

Scheme 3 – Synthesis of 22. Compounds 9 to 20 were prepared in a similar fashion.

ii

i

iv

22

iii

(i) Et3N, NaBH(OAc)3, DCM, 56%. (ii) TFA, DCM, quant. (iii) DIPEA, DMF, 80 °C, 34%. (iv) Ruphos palladium(II) phenethylamine chloride, dicyclohexyl(2',6'-diisopropoxy-[1,1'-biphenyl]2-yl)phosphine, Cs2CO3, toluene, 90 °C, 28%.


Page 7 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


In exploring the structure activity relationships in relation to the potential bromodomain inhibitory activity, we considered that both the triazolopyridazine and acylpiperazine moieties of the molecule could act as acetyl lysine mimetics (subsequently, we discovered that in fact both do, resulting in the in cis binding mode and accounting for the SAR we observed)11. Modifications to both of these regions were designed with consideration to the overlay of the compounds with acetyl lysine in the binding site. Changing the acyl piperazine of 5 for N-methylpiperazin-2-one (6) resulted in retained potency (AR DR pIC50 6.5 for both compounds) (Table 1). The isomeric piperazin-3-one 7 was slightly less potent (ER DR pIC50 6.0 compared to 6.3 for 6). 3-Methyl substitution of the piperazin2-one, 8, led to an increase in ER DR potency (pIC50 7.6 compared to 6.3 for 6). In this series, the piperazinones had improved solubility but also increased metabolism, as measured by human hepatocyte turnover, compared with the acylpiperazine 5. BRD4 assay data reveleaved that compounds 5 to 7 had weak BRD4 potency (pIC50 range 5.6 - 6.7) and that introduction of the piperazinone 3-methyl substituent led to increased BRD4 activity (pIC50 8.1) consistent with the increase in the cellular potency.

Table 1 – Effects of structural modifications of the acylpiperazine

Compound number

R

BRD

AR

ER

Myc

MM.

LogD

Solubili

Hu

4 FP

DR

DR

DR

1S

7.4

ty / µM

heps

pIC

pIC50

a

pKi

pIC50

a

a

a

pGI50

Clint /

a

mLmi

50

n1

×10-6

cells



1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 60

5

6.2

6.5

6.3

3.2

5.4

12

6

6.7

6.5

6.3

3.0

65

18

7

5.6

6.0

3.1

52

26

8

8.1

3.3

63

70

a

7.4

7.6

8.0

8.3

Measurements reported are the mean of at least three determinations. The standard deviation is not

shown, as all three assays were found to be highly reproducible on repeat testing (see Experimental Section for more detail).


Page 9 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


Next, the effect of changing the triazolopyridazine trifluoromethyl substituent of 8 was explored (Table 2). Removal of the substituent (9) led to a loss of BRD4 and cellular potency. Replacement of the trifluoromethyl group with methyl, chloro, thiomethyl, methoxy and dimethylamino substituents (10-14 respectively) resulted in retained activity. Of these compounds, the methoxy derivative 13 had the lowest logD7.4 value and showed very high solubility and a significantly improved human hepatocyte stability, resulting in a clear improvement in overall profile. Hence, the methoxy substituent appeared optimal and was retained in subsequent analogues.

Table 2 - Effect of changing the triazolopyridazine trifluoromethyl substituent

Compound

R

number

BRD4

AR

ER

Myc

MM.1S

pKia

DR

DR

DR

pGI50a

pIC50a

pIC50a

pIC50a

LogD7.4

Solubilit

Hu

y / µM

heps Clint / mLmin1

×10-6

cells

8

-CF3

8.1

7.4

7.6

8.0

9

-H

< 6.0

5.8

4.2

63

70

1.9

1000

180

8.3

2.4

670

7.2

7.7

3.1

63

24

2.9

660

21

2.3

>1600

1600




4.4

Page 11 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


Further exploration of changes to the piperazinone group (Table 3) showed that the chirality of the 3methyl substituent was important, with the S-enantiomer 15 showing a loss of potency (1.6 log units relative to 13), suggesting that a specific interaction was formed with the methyl group. The 3,5dimethyl-1-acetylpiperazine analogue 16 showed an increase in potency relative to piperazinone 13 (0.5 log units vs BRD4 but 1.2 units in ER DR. It is possible that this apparent discrepancy arises due to the potency limit in the FP assay. Unfortunately, this change also resulted in a decrease in human hepatocyte stability. Again, the methyl groups appeared to be important for potency, for instance, the corresponding 2,6-dimethyl isomer 16 showed a loss of potency (1.4 log units in ER DR) relative to 15. Further methylation of the piperazinone, for example 18, led to similar potency to 13 but with a marked decrease in human hepatocyte stability. These data suggested that balancing potency and metabolic stability in these compounds was very challenging and the (R)-1,3-dimethylpiperazin-2-one group appeared to strike the best balance.

Table 3 - Further exploration of changes to the piperazinone group

Compound number

R

BRD4

ER

Myc

MM1S LogD7.

Solubi

Hu

pKia

DR

DR

pGI50a

lity /

heps

pIC50a

pIC50a

µM

Clint /

4

mLmin -1

×10-6

cells



1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

13

8.3

7.6

7.9

15

6.9

6.5

7.1

16

8.8

8.8

>8.7

17

8.3

7.4

18

8.1

7.7

a

7.9

8.3

>8.7

8.2

Page 12 of 60

2.3

>1600

1600

8.8

2.4

>1000

5.1

O

N N

19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

20

8.2

7.7

21

8.5

7.6

22

7.2

23

24

a

Page 14 of 60

2.8

470

12

2.5

570

8.7

8.3

8.8



Solving the X-ray crystal structure of 13 in complex with the first bromodomain of BRD4 [BRD4(1)] revealed the formation of a dimer as previously observed11, in which both the triazolopyridazine and N-methylpiperazinone engage the acetyl lysine sites of different bromodomains (Figure 2). In the first bromodomain, critical hydrogen bonds were observed between the triazolopyridazine nitrogen-1 and Asn140 with the nitrogen-2 and the methoxy substituent acting as the acetyl lysine mimic. In the second bromodomain, the N-methylpiperazinone methyl and carbonyl groups acted as the acetyl lysine mimic with the carbonyl also engaging Asn140. The 3-methyl group filled a small lipophilic hole, explaining the increase in potency arising from its introduction. Whilst this structure is homodimeric, showing the interaction of both ends of the molecule with BRD4(1), given the close


Page 15 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


homology between the first and the second bromodomains, it is reasonable to anticipate that similar interactions would be made with each bromodomain in an in-cis complex with full length BRD4. This is fully consistent with the observed SAR and explains, for instance, the increase in potency for 16 relative to 13, 17 and 18 since the additional methyl group of 16 is able to occupy a vacant lipophilic cavity in the binding site. In this model, the diminished potency of 22 and 23 relative to the other exemplified linkers could be explained by the linear constraints of the linkers of these molecules requiring an energetically less favoured orientation of the two bromodomains in order to achieve bivalent binding.

Figure 2 – X-ray crystal structure of 13 in complex with BRD4(1) showing the formation of a dimeric structure with two bromodomains. b) Engagement of the triazolopyridazine of 13 with the first bromodomain (blue ribbon) acetyl lysine binding site. c) Engagement of the N-methylpiperazinone of 13 with the second bromodomain (red ribbon) acetyl lysine site.

To provide further evidence of the bivalent binding mode, compounds were tested against the N140A and N433A mutant forms of BRD4 in the FP assay and the pKi values obtained were compared to the wild type protein (Table 5). N140 in BD1 and N433 in BD2 of BRD4 have been shown to be important in the binding of potent BRD ligands12. Mutation of these residues was expected to provide insights into the binding mechanism. Monovalent ligand 1 showed consistent Ki values across the wild type and both mutant proteins as would be expected. Compounds 5, 6 and 7, which have the potential for bivalent binding but have an unoptimised second site binding moiety showed a decrease in affinity for the N140A mutant but an increase in affinity for the N433A mutant in relative to wildtype. This suggests that the triazolopyridazine moiety binds to BD1 and the acetyl piperazine or



1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

piperazinone group binds to BD2 although this has not been definitively established. The reason for the increase in affinity for the N433A mutant relative to wild-type for these compounds is unclear. It may suggest that these the unoptimised second site binding groups employed can interact more strongly with mutated BD2 than for wild type, indeed, these changes would be expected to disrupt the critical interaction with the asparagine residue. Compounds 8 – 24, which are all optimised bivalent compounds consistent show a decrease in affinity of approximately 1 order of magnitude for both mutant proteins relative to wild type. These structure activity relationships provide further evidence of the bivalent binding mechanism of the series as a whole.

Table 5 – pKi values for BD1 and BD2 mutant proteins compared to wild type Compound

WT FP pKia

number

N140A

N433A

pKia

pKia

1

7.3

7.3

7.4

5

6.2

5.5

6.6

6

6.7

6.0

6.5

7

5.6

5.1

6.4

8

8.1

7.2

6.4

9

< 6.0

5.6

5.2

10

7.8

6.7

6.6

11

7.9

6.5

6.7

12

7.9

6.9

6.8

13

8.3

6.8

6.9

14

7.9

6.5

6.5

15

6.9

6.0

6.6

16

8.8

7.8

7.3

17

8.3

6.8

7.3

18

8.1

6.6

6.9


Page 16 of 60

Page 17 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


a

19

8.4

7.2

7.1

20

8.2

7.0

7.2

21

8.5

6.9

7.0

22

7.2

6.4

6.7

23

7.3

6.5

6.8

24

8.0

7.0

6.8





1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60

The wild type pKi values were shown to accurately predict the cellular potencies such as ER DR (Figure 3). This provides confidence that the observed pharmacology is BRD4 driven, although we would expect the compounds to have similar affinities for other members of the BET family due to their close homology and the lack of selectivity within the BET family for 13 in selectivity screens (see later). Moreover, the observation of this correlation and lack of drop off between in vitro BRD4 inhibition and cellular potency, in contrast to monovalent inhibitors, which show significantly lower cellular potency11, suggests that the bivalent mechanism is operative in a cellular environment.

Figure 3 - Correlation of BRD4 WT pKi vs ER Downregulation. Bivalent triazolopyridazine are shown with black dots. Line Fit: ER DR pIC50 = 0.046 + 0.93*[BRD4 WT FP pKi], r2 = 0.84, n=20

The compounds with the best balance of potency and in vitro ADME properties (favorable solubility, Caco-2 Papp and metabolic stability) and attractive pre-clinical pharmacokinetics were selected for further profiling (Table 6). A preliminary human dose prediction13 was calculated based on achieving


Page 18 of 60

Page 19 of 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 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 50 51 52 53 54 55 56 57 58 59 60


24 hour free cover above a cellular GI50 (potency in MM.1S cells) from a twice-daily dosing schedule (BID, 12 hours apart). The maximum absorbable dose (MAD) and human Fabs were estimated in GISim14 using measured Caco-2 Papp, solubility and pKa data. Several compounds had low predicted doses ( 100 fold). Compound 13 was also attractive in that it had its in vivo plasma clearance appeared well predicted from in vitro data across species, providing greater certainty in the human clearance and plasma half-life predictions. 13 also had improved metabolic stability in hepatic microsomal fractions compared to other examples (data not shown). As a result, 13 was selected for further development and given the internal identifier AZD5153.

Table 6 - Tabulated PK data. Preclinical PK parameters were derived from plasma concentration-time data using non-compartmental analysis. Property

13

16

24

21

14

MM.1S GI50 / µM

0.005

0.002

0.002

0.005

0.003

logD / basic pKa

2.3 / 4.8

2.5 / 6.2

1.7 / 5.9

2.5 / 6.3

2.4 / -

Solubility, pH7.4 / µM

>1650

>1000

>1000

570

>1681

Caco-2 Papp / nms-1

400

300

460

280

-

4 / 24 / 23

6 / 16 / 19

5 / 27 / 32

2 / 24 / 25

4 / 9 / 16

5 /

Optimization of a Series of Bivalent Triazolopyridazine Based

Recommend Documents