Pharmacodynamics-Immunogenicity Approach for Bioanalysis of Biological Drugs

An integrated pharmacokinetics- pharmacodynamics-immunogenicity approach for bioanalysis of biological drugs

Andrea Kiessling – Principal Scientist PK/PD Bioanalytics II Preclinical Safety / Biologics Safety & Disposition EBF 3rd Open Conference, 2 December 2010 Outline of the presentation

 PK-PD models for monoclonal antibodies - ligand binding models for soluble and cell surface targets - typical PK-PD behavior

 Integrated bioanalytical strategy: PK-PD and immunogenicity (IG) - ELISA formats to characterize PK and PD (soluble and cell surface targets) - impact of immunogenicity

 Examples - Example 1: soluble target; effect of immunogenicity on PK and PD - Example 2: cell surface target; detection of neutralizing immunogenicity from PK assay format - Example 3: cell surface target; effect of PK on PD

 Summary

2 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 A simple mAb-ligand PK-PD model

input ligand

mAb “A” + ligand “B” mAb – ligand complex “C”

dose elimination elimination elimination mAb ligand mAb – ligand complex

3 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 mAb-ligand PK-PD binding model(s)

Soluble target: Cellular target: mAb-ligand complex “tends” to mAb-ligand complex “tends” to take on the elimination take on the elimination characteristics of the mAb characteristics of the ligand - accumulation of “total” - TMDD apparent (PD marker) (inactive) ligand (PD marker)

Lowe PJ et al: On setting the first dose in man: Quantitating biotherapeutic drug- target binding through PK and PD models Basic & Clin Pharmacology & Toxicology 2009; 106: 195-209

4 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010

“Typical” PK-PD behavior – cellular target anti-CD11a mAb – Raptiva (efalizumab)

mAb complex ligand

increasing dose: change in IgG kinetics

increasing dose: increases duration of effect

Joshi et al An overview of the pharmacokinetics and pharmacodynamics of efalizumab: a monoclonal antibody approved for use in psoriasis J Clin Pharmacol 2006; 46: 10-20

5 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 “Typical” PK-PD behavior – soluble target anti-IL1 mAb – Ilaris (canakinumab) simulation 0.1, 0.3, 1, 3 and 10 mg/kg

mAb complex ligand

increasing dose: increasing dose: no change in IgG kinetics increases duration of effect

10 mg/kg 0.1 mg/kg 3 mg/kg 0.3 mg/kg 1 mg/kg 1 mg/kg 0.3 mg/kg 3 mg/kg 0.1 mg/kg 10 mg/kg

6 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Outline of the presentation

 PK-PD models for monoclonal antibodies - ligand binding models for soluble and cell surface targets - typical PK-PD behavior

 Integrated bioanalytical strategy: PK-PD and immunogenicity (IG) - ELISA formats to characterize PK and PD (soluble and cell surface targets) - impact of immunogenicity

 Summary

7 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 8 Bioanalytical strategy strategy Bioanalytical

| | 3 EBF rd

Open Conference| Open Andrea Conference| BIAcore Plate based Bridging format Protein Protein G format Kiessling

Protein G drug Anti drug Anti labelleddrug | 02 December 02 December | 2010 - -

drug antibody drug antibody

–

Immunogenicity (IG) Immunogenicity Sandwich format Direct Direct format

drug Anti drug Anti labelledsecondaryantibody - -

drug antibody

Bioanalytical strategy – immunogenicity (IG)

Initial screening assays for immunogenicity do not differentiate non-neutralizing vs neutralizing immunogenicity

Immunogenicity to Fc region Neutralising immunogenicity

mAb (drug) – labeled

AHA

mAb (drug)

Biacore Bridging ELISA Biacore Bridging ELISA

increase in clearance of mAb increase in clearance of mAb no effect on target binding decrease in capacity for target binding

influence on PK and/or PD assays?

9 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Bioanalytical strategy – pharmacokinetics (PK)

Free-”bioactive” mAb

labeled drug

anti-human IgG labeled target drug drug drug anti-id or or anti-id or target or target target Sandwich ELISA Competitive ELISA Bridging assay

Total mAb

anti-human IgG

drug target anti-human IgG

Sandwich ELISA

10 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Bioanalytical strategy – PK Impact of neutralizing (nIG) and non-neutralizing (non-nIG) immunogenicity

neutralizing AHA

non-neutralizing AHA anti-human IgG

mAb (drug) nIG and non-nIG may interfere in the PK assay; detected as decrease in exposure ligand (short PK t½) sandwich ELISA

mAb (drug) ligand nIG may intefere in the PK assay; mAb (drug) detected as either an increase or ligand ligand decrease in exposure (see example later) competitive ELISA bridging ELISA

anti-human IgG non-nIG may interfere in the total PK mAb (drug) assay (pre-clinical); although nIG should not interfere in this assay

sandwich ELISA 11 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Bioanalytical strategy – PD (soluble ligand) total ligand (target capture) free ligand

ligand –endogenous anti ligand Ab2 mAb (drug) ligand – endogenous anti ligand Ab4 mAb (drug) binding protein binding protein + +   anti ligand Ab1 anti ligand Ab3

sandwich ELISA sandwich ELISA

Technically challenging Technically very challenging . capture and detection reagents for . detection of free ligand ligand are selected which bind in (decreasing) in the presence of presence of mAb (drug) and increasing concentrations of total endogenous binding protein ligand (if relevant) . specificity and sensitivity (LLOQ)

12 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Bioanalytical strategy – PD (soluble ligand) Impact of neutralizing (nIG) and non-neutralizing (non-nIG) immunogenicity total ligand (target capture) free ligand

ligand –endogenous anti ligand Ab2 mAb (drug) ligand – endogenous anti ligand Ab mAb (drug) binding protein binding protein + +   anti ligand Ab1 anti ligand Ab3

sandwich ELISA sandwich ELISA

Immunogenicity to mAb (drug) should not directly interfere with PD assay format for either total or free ligand per se if appropriate reagents can be identified (i.e. different binding epitope(s) to drug) However: nIG may lead to decreased ligand capture and therefore a decrease in total ligand but also directly interfere with the PD assay if the drug is used for detection (see example) more rapid clearance of mAb (drug) from nIG and/or non-nIG will also lead to decreased ligand capture and a more rapid return to baseline free ligand

13 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Outline of the presentation

 PK-PD models for monoclonal antibodies - ligand binding models for soluble and cell surface targets - typical PK-PD behavior

 Integrated bioanalytical strategy: PK-PD and immunogenicity (IG) - ELISA formats to characterize PK and PD (soluble target) - impact of immunogenicity

 Summary

14 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010

Example 1: background

 fully human IgG1 mAb; high affinity against soluble target  target ligand can be measured in the systemic circulation at baseline  high expression, high turnover?  mAb acts as a “capture system” increase in mAb-ligand complex (detected in serum) is a “biomarker” for suppression of free ligand in interstitial space via a PK/PD model

15 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 1: pre-clinical bioanalytical strategy

neutralizing AHA

non-neutralizing AHA

PK assay: total mAb PD assay: total target (target capture)

TMB substrate goat anti-mouse IgG

HRP mouse anti-human IgG anti-human IgG

mAb (drug) – add saturating amount

mAb (drug) ligand

anti-ligand Ab anti-human IgG

sandwich ELISA sandwich ELISA

16 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010

Example 1: PK-PD data (4wk DRF study cynomolgus monkey)

50 mg/kg mAb - days 0, 7, 14, 21 50 mg/kg mAb - days 0, 7, 14, 21

100000 100000 Total mAb Total mAb (t½ 4 days) 10000 10000

1000 1000

100 100 Concentration (nM) Concentration Concentration (nM) Concentration Total ligand Total ligand 10 10

1 1 0 7 14 21 28 0 7 14 21 28 Time (days) Time (days)

NB One animal excluded from plot ... evidence of IG from PK and PD data after the third dose; confirmed on Biacore

17 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 1: Conclusion

In this example, mAb(drug)-ligand complex appears to be eliminated more rapidly than typical human IgG in cynomolgus monkey

Immunogenicity apparent in both, PK and PD behavior No information about neutralizing capacity of the immune response; although PK-PD suggest that pre-dose on day 14 less target was captured as compared to the other two animals  slight indication for neutralizing anti-drug antibodies at least at the beginning of the response

18 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 2: background

 Fully human IgG1 mAb; high affinity against cell surface target  At low concentrations, mAb (drug) is expected to be cleared by target mediated disposition (TMDD)

 No PD (receptor occupancy) as target cells present only at low level in serum

19 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 2: pre-clinical bioanalytical strategy

neutralizing AHA

non-neutralizing AHA

PK assay format 1: free “bioactive” mAb Immunogenicity assay:

mAb (drug) AHA mAb (drug) ligand competitive ELISA Biacore

PK assay format 2: free “bioactive” mAb

ligand

mAb (drug) ligand bridging ELISA

20 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 2: PK and IG data (cynomolgus monkey)

Three doses 100 mg/kg i.v. days 0, 7, 14 (n=3 animals) 10000 400 PK assay 350

1000 ligand

/ml] 300

[RU]

[ mAb (drug) 250 100

Untis ligand 200 bridging ELISA 10

150

Response

100 Rel

Serum concentration Serum 1 50 Immunogenicity 0 0 0 20 40 60 80 100 120 AHA Days

mAb (drug)

Biacore

21 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 2: PK and IG data (cynomolgus monkey)

Single dose 10 mg/kg i.v. n=3 animals PK Immunogenicity

300 1400 1200

1000 200 800

600 [µg/mL] 100 400

200

anti-humaninterleukin mAb Serum concentration Serum

0 [RU] Response Unit rel. 0 0 10 20 30 40 50 0 10 20 30 40 50 Time [day] -200 Time [day]

2001 2002 2003 2001 2002 2003

AHA mAb (drug)

ligand mAb (drug)

competitive ELISA Biacore

22 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 2: influence of IG on competitive ELISA

drug ADA

Low mAb concentration High mAb concentration Low mAb concentration No immunogenicity Low AHA response High AHA response Preferential binding of labeled Preferential binding of mAb AHA blocks binding of mAb and mAb to ELISA plate labelled mAB to ELISA plate to ELISA plate

High signal = Low signal = Low signal = low mAb concentration high mAb concentration high mAb concentration

CORRECT CORRECT INCORRECT

23 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010

Example 2: Conclusion

Immunogenicity is apparent in both the competitive and bridging ELISA PK assay formats

The bridging ELISA format does not allow to differentiate between TMDD, nIG or increased clearance due to non-nIG

The competitive format allows identification of nIG by giving false positive results at low drug concentration in presence of strong immune response

24 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 3: background

 Fully human IgG1 mAb; high affinity against cell surface target  At low concentrations, mAb (drug) is expected to be cleared by target mediated disposition (TMDD)

 No receptor occupancy available as target is expressed in solid tissues  PD is analyzed by a physiological response readout

25 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 3: pre-clinical bioanalytical strategy

PK assay format 1: free “bioactive” mAb Immunogenicity assay:

ligand

mAb (drug) AHA mAb (drug) ligand bridging ELISA Biacore Blue arrows: interference with neutralizing AHA

PD: analysis of physiological response

26 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 3: PK and PD data (rat)

Based on data of a two-dose (day 0 and day 7) DRF study in rat

ligand

mAb (drug) ligand

bridging ELISA Serum concentration Serum

PD read out PDread 0

BodyweightChange (g) -10 0 10 20 30 40 50 60 70 80 90 100 TIME (Days) 27 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Example 3: Conclusion

Immunogenicity cannot explain the typical PK found in all animals showing an infliction point at a serum concentration of 10 µg/mL where clearance increases

The increased clearance is most likely due to target-mediated disposition when the concentration in serum drops below that needed to saturate an receptor mediated elimination mechanism

The direct connection between the saturation of target mediated disposition and a sustained PD effect likely reflects that (nearly) complete receptor occupancy is required to drive the PD

28 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Outline of the presentation

 PK-PD models for monoclonal antibodies - ligand binding models for soluble and cell surface targets - typical PK-PD behavior

 Integrated bioanalytical strategy: PK-PD and immunogenicity (IG) - ELISA formats to characterize PK and PD (soluble and cell surface targets) - impact of immunogenicity

Summary

29 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Draft immunogenicity addendum ICH S6:

Immunogenicity assessments are conducted to assist in the interpretation of the study results and design of subsequent studies. Such analysis in non-clinical animal studies are not relevant in terms of predicting potential immunogenicity in humans. Measurement of anti-drug antibodies (ADA) in non-clinical studies is not routinely required if there is evidence of sustained pharmacodynamic activity and no unexpected changes in the pharmaco- / toxicokinetics of the test article during the dosing or recovery phase, and/or no evidence of immune mediated reactions (immune complex related, vasculitis, anaphylaxis etc), ADA testing is typically not warranted. However, it is difficult to predict whether such analyses will be needed prior to completion of the in-life phase of the study; therefore it is often useful to obtain appropriate samples during the course of the study, which may be subsequently analysed to aid in interpretation of the study results

30 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Draft immunogenicity addendum ICH S6:

When study results suggest potential for immunogenicity, antibody detection assays should be conducted to evaluate the presence of ADAs. When ADAs are detected it is important to address neutralising activity, but not titre because of the poor correlation with neutralising potential. Although assessment of neutralising activity can be addressed using a separate neutralising antibody assay, other approaches can also be valid and may be more robust, such as monitoring of a PD marker, use of an ex-vivo bioactivity assay or the combination of assay formats for PK-PD which address neutralising potential.

31 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 Summary

. An integrated design of bioanalytical assays and interpretation strategy maximizes information about PK, PD and IG interrelationship

. An integrated bioanalytical strategy should take into account: - nature of the target; eg cell surface, soluble - target expression and target turnover - influence of IG on analytical formats for PK and PD assays

. The need for specific neutralizing assays may be obviated by the use of alternative markers of functional activity

32 | EBF 3rd Open Conference| Andrea Kiessling | 02 December 2010 THANK YOU for your attention

Acknowledgements: Sebastian Spindeldreher Peter Lloyd Jennifer Sims Andrew Warren Annette Zaar Sherri Dudal Kerstin Kentsch Ulf Klein

Phil Lowe

33 | CHI Immunogenicity Summit 2010 | Sebastian Spindeldreher | 19 October 2010