Richard M. Eglen ELRIG, Manchester, UK September 7-8, 2011

High throughput screening and assay development – fit for purpose? Richard M. Eglen ELRIG, Manchester, UK September 7-8, 2011

Agenda • Novel drug target trends • Single target vs. cellular networks • Target based vs. phenotypic screening • HTS assay development • Parallels & differences with diagnostic assays • Emerging trends • Predicting the future • Assay development • …still ‘fit for purpose’?

Novel trends in drug discovery targets • 1996 - Drews & Ryser • 483 drug targets exploited • 5 - 10,000 targets in human genome • (prevailing view was 300,000 genes in genome) • 2002 - Hopkins & Groom • 120 drug targets for marketed small molecule drugs • 399 targets druggable • 10% of human genes pharmacologically tractable • 2002 - Human genome sequenced • 30,000 genes; 3,000 linked to disease; 600 – 1500 druggable targets • 2006 - Imming et al • 218 drug targets • 2006 - Overington et al • 324 pharmacological targets assigned to 1065 pharmacological agents Rask-Andersen et al., 2011 Nature revs drug Disc. 10, 579.

Novel drug targets – emerging trends • 435 drug targets in human genome… • …modulated by 989 unique drugs… • …via 2,242 drug-target interactions. Rask-Andersen et al., 2011 Nature Revs Drug Disc. 10, 579.

New drugs affecting human genome targets– structural classes Rask-Andersen et al., 2011 Nature Revs Drug Disc. 10, 579.

Targeting drug networks …not targetse.g. PDEs Rask-Andersen et al., 2011 Nature Revs Drug Disc. 10, 579.

Targeting drug networks e.g. EDGR/PDGR Rask-Andersen et al., 2011 Nature Revs Drug Disc. 10, 579.

Smaller networks suggest both novel targets and mechanisms for intervention Rask-Andersen et al., 2011 Nature Revs Drug Disc. 10, 579.

Novel drug targets – emerging trends (1982 – 2010) • A constant rate of NPI introduction …but not in line with increases in R&D investments • Most drugs approved were acting at previously exploited targets • Older drugs were most ‘connected’ in terms of drug networks • Newer drugs (2005 – 2010) directed at smaller novel networks One drug/disease, one target or… One drug/disease, target network?

Target based vs. network based screening– the rise and rise of phenotypic screening? • Target based drug discovery • Accelerated by the human genome sequencing revolution & advances in structure based analysis, in vitro and in silico • Target identification & validation a ‘sine qua non’ • …how successful has this approach been? 2011 - Swinney and Anthony • 1998 - 2008; 259 NPIs approved • 75 had novel mechanisms of action • 67% small molecules; 33% Biologics • 28 by phenotypic screening; 17 by target based screening Swinney and Anthony 2011 Nature Revs Drug Disc. 10, 507.

Discovering first-in-class NPIs Swinney and Anthony 2011 Nature Revs Drug Disc. 10, 507.

Pharmacology of novel NPIs– approx. 50% target are enzymes Swinney and Anthony 2011 Nature Revs Drug Disc. 10, 507.

NPIs – first in class vs. followers Swinney and Anthony 2011 Nature Revs Drug Disc. 10, 507.

NPIs according to mode of discovery First in class: Novel mechanisms of action (MOAs) Follower drugs: Established MOAs Swinney and Anthony 2011 Nature Revs Drug Disc. 10, 507.

Target based vs. phenotypic drug screening • Phenotypic screening more successful that historically realized • Can give rise to more successful first in class NPIs • Robust MOA knowledge is required to optimize back ups • Fast follower NPIs often target based as a result. • But… • Robust knowledge of MOAs yields new drug approaches • allosterism, kinetics, complex binding phenomena etc. • Generally, critical for Biologic-based drug design • 33% of first in class drugs were Biologics e.g. mAbs • How well do phenotypic assays translate to human disease? • How fit for purpose are current assay development & screening? Swinney and Anthony 2011 Nature Revs Drug Disc. 10, 507.

ADMET HTS Target based HTS HTS concept DMSO + synthetic cpmds 96 well plates HTS Centralized Full file screening HTS ADMET P450, CACO2 binding 96 well Recent advances Miniaturization Nanotechnology Academic entry NIH roadmap 1997 2000 2002- 1989 1995 1996 1984 1986 1987 Natural products screening Automation 10,000 assays/wk Pre Candidate tech 90 cmpds/wk HT LC-MS Cytotox MTT Applied Biotech/Screening 7,200 cmpds/wk 20 concurrent HTS Cell based + biochemical RT-PCR 180 cmpds/wk HT LC-MS Cytotox MTT 2880 cmpds/wk All data recorded 96 pipettors + harvesters Evolution of HTS e.g. Pfizer 360 cmpds/wk HT LC-MS Cytotox MTT after Pereira and Williams 2007, Br J Pharmacol 207, 152

Evolution of HTS assay strategies Coupled protein readouts Isolated membrane studies Isolated protein activity Cell based phenotypic approaches Inglese et al 2007 Nature Chem Biol 3 466.

Moving from bench top to HTS assays Inglese et al 2007 Nature Chem Biol 3 466.

Assay technologies: HTS & Dx compared

Emerging trends in HTS… In vivo (animal) screening In vitro (biochemical) screening In vivo (immortalized cells) screening Phenomenological Low throughput Disease relevance? Target based High throughput Disease relevance - low Target based High throughput Disease relevance? In vivo (ES & iPS cells) screening In vivo (3D tissue assemblies) screening In vivo (primary cells) screening Phenotypically based Throughput? Disease relevance - high Phenotypically based High throughput Disease relevance - high Phenotypically based Low throughput Disease relevance - high

Converging trends in HTS … Imaging & phenotypic Microfluidic Label-free Disease relevant cells

Questions… • Adopting diagnostic (Dx) assay platforms, automation and detection systems has provided a strong basis for HTS assay development. • Drug network, cell based and phenotypic screening approaches are being more widely adopted. • Will the historical parallels of Dx technologies providing HTS assay formats hold going forward? • How fit for purpose are classical HTS assays for the next generation of drug targets?

Assay technologies: HTS & Dx compared

Example 1 - stepping away from the microtiter plate • Microfluidics • Networks of channels, 10-100nm in diameter • Biochemical & cell based rapidly being developed for diagnostics, notably PCR systems • e.g. Droplet based microfluidics • overcomes mixing issues, laminar flow issues • Allows physiologically relevant cell-based assays (Chapman, 2004) • e.g. Assays with primary human cells Clausell-Tormos et al. 2008, Chem & Biol 15, 427.

Example 2 - Single cell flow cytometry and immune profiling • Cells stained with epitope-specific antibodies conjugated to transition element isotope reporters, each with a different mass. • Cells nebulized into single-cell droplets, and an elemental mass spectrum is acquired for each. • The integrated elemental reporter signals for each cell can then be analyzed by using flow cytometry. Bendall et al. 2011, Science 332, 687.

Example 3 - Cell trapping & sequential array cytometry • Hydrodynamic cell trapping for exchange of solutions and imaging. • (a) Three-dimensional hydrodynamic cell traps were created in massive arrays • (b) Cell traps are raised to allow fluid streamlines to pass beneath them, dragging in cells. No external forces other than the fluid driving force are needed. • (c) Hydrodynamically trapped cells can have fluid solutions exchanged around them, allowing for sequential staining and imaging of a constant set of cells. Gossett et al. 2010, Ann Biomed Eng 39, 1328.

Hand held Dx devices … implications for HTS? “In this paper, we present a mobile application that automatically quantifies immunoassay test data on a smart phone. The speed and accuracy demonstrated by the application suggest that cell-phone based analysis could aid disease diagnosis at the point of care”. Dell et al., 2011NSDR’11, June 28, 2011, Bethesda, USA. Chin et al., 2011 Nat Med DOI 10.1038/nm.2408

HTS contributions in Pharma drug development Macarron et al 2011 Nature Revs Drug Disc. 10, 188.

Recently approved drugs with HTS origins Macarron et al 2011 Nature Revs Drug Disc. 10, 188.

Common & incorrect HTS myths • Data is poor quality • Expensive & time consuming • Anti intellectual and irrational • Fails to find leads for many targets • Why are so few drugs from HTS? • Poorly validated targets • Non physiological screens • Limited informatics • Unpredictable ADME&T Misplaced and naïve expectations! HTS …views from the front line Fit for purpose: …something that is ‘fit for purpose’ is good enough to do the job it was designed to do… Webster Macarron et al 2011 Nature Revs Drug Disc. 10, 188.

Assay development & HTS – still ‘fit for purpose’? • HTS is historically a successful and well integrated activity in drug discovery… • …adapting many technologies initially developed for the in vitro diagnostic (IVD) industry. • As novel target types - and target networks - are validated, it is likely that new assay technologies (non microtiter plate based?) will need to be adopted. • The impact of phenotypic screening may have been underestimated…this fact, plus near universal adoption of cellular assays…suggests HTS assay development will no longer mirror technologies developed for IVDs. Next generation HTS technologies? – what, when & where?

Questions?

Richard M. Eglen ELRIG, Manchester, UK September 7-8, 2011