HOS Characterization, Stability, and Comparability of ADCs ......Common antibody-drug conjugation strategies 3 Conjugation strategy Typical locations Typical DAR * distribution Semi

HOS Characterization, Stability, and

Comparability of ADCs

– Biophysical Analyses

Yin Luo*, Sharon Polleck, Lucy Liu

Analytical R&D, Biotherapeutic Pharmaceutical Sciences

Pfizer, Andover, MA

CASSS HOS 13-April-2015

What is an ADC Covalent conjugate of cytotoxic drug (payload) with mAb against

selected antigen on the surface of target cells

What are the advantages of ADC − increase efficacy

− decrease toxicity (lower dosing of toxic drug)

Why ADCs have these advantages ‒ enhanced drug specificity – mAb binding to target antigens

‒ superior Mode of Action – drug cytotoxicity within target cells

Linker and Payload

mAb

Antibody Drug Conjugate (ADC)

2

Target

cell

Cell death

mAb binding ADC internalization Cytotoxicity

Target

cell

mAb binding

Cell death

ADCC

CDC

Cytotoxicity

mAb cytotoxicity MOA ADC cytotoxicity MOA

Fc effector

functions

DNA / scaffold

destruction

Common antibody-drug conjugation strategies

3

Conjugation

strategy Typical locations

Typical DAR*

distribution

Semi site

specific

Cysteine (inter-chain S-S partial

reduction)

IgG1 IgG4

Lysine

Site

specific

Cysteine (engineered)

Glutamine tag (transglutaminase)

*Drug Antibody Ratio

4

Heterogeneity of inter-chain cysteine conjugation

Interchain S-S

partial reduction

Thioether

conjugation

Possible locations DAR

2

4

6

8

mAb

Challenges of ADC as biotherapeutics

CMC – complex manufacturing

– insufficient conjugations

– lot to lot reproducibility

– challenging analytics

Potency – uncontrolled payload loading

– maleimide exchange with HSA

– PK liability

– variable potency

Toxicity – linker instability resulting in premature drug release

– aggregation

– first pass metabolism in liver

5

Boswell, CA, R et al. Bioconjugate Chemistry 2011, 22, 1994 - 2004

Junutula, R et al. Nat Biotechol. 2008, 26, 925 – 932.

Junutula, R et al. Clinical Caner Res. 2010, 16, 4769 – 4778.

Analytical challenges

Heterogeneity (multiple payload

conjugation)

Payload hydrophobicity

Linker/payload stability

Aggregation propensity of ADC

Impact of conjugation on HOS

Relation of HOS to ADC function

Unconjugated species

etc

Foci of this

presentation

HOS: higher order structure

Protein higher order structure (HOS) and biophysical

methods for (low-resolution) characterizations

Far-UV CD Near-UV CD SEC-Multi Angle Light Scattering

(SEC-MALS)

FTIR Intrinsic fluorescence Analytical Ultracentrifugation (AUC)

Differential scanning

calorimetry (DSC) Field Flow Fractionation (FFF) - MALS

Dynamic light scattering (DLS)

Modified from

Figure 3-23

Lehninger “Principles

of Biochemistry”

5th Edition

HOS

Common

methods Common

methods

Case studies – Molecular information

Lysine conjugation 6 mAb3 (IgG4)

ADC3

Linker/Payload Conjugation

Partial reduction/

cysteine conjugation

Partial reduction/

cysteine conjugation

Average

DAR*

4

4 mAb1 (IgG1)

mAb ADC

mAb2 (IgG1)

ADC1

ADC2

MW ~1660

*Drug Antibody Ratio

Case studies –

Aggregation propensity of ADC in comparison

with mAb by SEC-MALS and AUC

8

Pfizer Confidential │ 9

Aggregation in mAb1 and ADC1 by SEC-MALS

0.E+00

1.E-05

2.E-05

3.E-05

4.E-05

5.E-05

6.E-05

0

50000

100000

150000

200000

250000

300000

350000

400000

450000

500000

5 6 7 8 9 10 11 12 13 14 15

Dif

fere

nti

al R

efra

ctiv

e In

dex

Mo

lar

Ma

ss (g

/mo

l)

Time (Minutes)

0.E+00

1.E-05

2.E-05

3.E-05

4.E-05

5.E-05

6.E-05

7.E-05

0

50000

100000

150000

200000

250000

300000

350000

400000

450000

500000

5 7 9 11 13 15

Dif

fere

nti

al R

efra

ctiv

e In

dex

Mo

lar

Ma

ss (g

/mo

l)

Time (Minutes)

mAb1 ADC1

Weight average

molar mass

Weight average

molar mass

SEC-MALS reported dimer as the main component of the HMMS of both mAb1

and ADC1 in similar quantities. ADC3 & mAb3 show similar profiles

Sample Monomer Dimer

Weight average

molar mass % Weight average

molar mass %

mAb1 148 99.0 295 1.0

ADC1 152 98.7 318 1.3

Aggregation in mAb1 and ADC1 by AUC

Pfizer Confidential │ 10

0

0.5

1

1.5

2

2.5

4 6 8 10 12 14 16 18 20

-0.02

0

0.02

0.04

0.06

0.08

0.1

4 6 8 10 12 14 16 18 20

C (

S2

0,w

)

S(20,w)

C (

S2

0,w

)

S(20,w)

HMMS

Monomer

0

0.5

1

1.5

2

2.5

4 6 8 10 12 14 16 18 20

-0.02

0

0.02

0.04

0.06

0.08

0.1

4 6 8 10 12 14 16 18 20

S(20,w)

C (

S2

0,w

)

C (

S2

0,w

)

S(20,w)

Monomer

HMMS

The mAb and ADC were run

in the SE-HPLC mobile phase

mAb1 ADC1

Sample Monomer (%) Dimer (%)

mAb1 >99

Aggregation in mAb2 and ADC2 by SEC-MALS

mAb

Sample Monomer “Other” Dimer

Weight average

molar mass % Weight average

molar mass % Weight average

molar mass %

mAb2 147 99.6 NA NA 311 0.3

ADC2 156 96.2 234 1.7 328 2.0

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0

50000

100000

150000

200000

250000

300000

350000

400000

450000

8 10 12 14 16 18 20 22 24

UV

Ab

sora

nce

@ 2

80

nm

Mo

lar

Ma

ss (D

a)

Time (Minutes)

Dimer 328 kDa

Monomer 156 kDa

“Other” 234 kDa

ADC2 DAR = 4.2

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0

50000

100000

150000

200000

250000

300000

350000

400000

450000

8 10 12 14 16 18 20 22 24

UV

Ab

sorb

an

ce @

28

0 n

m

Mo

lar

Ma

ss (D

a)

Time (Minutes)

SEC-MALS reported dimer and “other” species with mass between monomer and dimer

Pfizer Confidential │ 12

mAb2

0.0

1.0

2.0

3.0

4.0

5.0

6.0

4 5 6 7 8 9 10 11 12 13 14 15

C (

s)

S

0.00

0.05

0.10

0.15

0.20

0.25

4 5 6 7 8 9 10 11 12 13 14

C (

s)

S

Monomer

Dimer

ADC2

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4 5 6 7 8 9 10 11 12 13 14 15

C (

s)

S

0.0

0.1

0.2

0.3

0.4

0.5

4 5 6 7 8 9 10 11 12 13 14 15

C (

s)

S

Monomer

Dimer

“Other”

The mAb and ADC were run in the SE-HPLC mobile phase

Aggregation in mAb2 and ADC2 by AUC

AUC confirmed low levels of dimer and “other” species, but did not detect larger aggregates

Sample Monomer (%) “Other” (%) Dimer (%)

mAb2 98.6 NA 1.4

ADC2 92.1 5.1 2.8

Aggregation in ADC by SEC-MALS

Exemplary ADC early development materials

0

25000

50000

75000

100000

125000

150000

175000

200000

225000

250000

275000

300000

325000

350000

375000

400000

425000

450000

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

11 12 13 14 15 16 17 18 19 20 21

Mol

ar M

ass

(g/m

ol)

Abso

rban

ce (A

U)

Time (Minutes)

UV

Molar Mass

13

8 k

15

0 k

17

7 k

29

0 k

35

4 k

SEC-MALS can provide average molar mass for SE-HPLC fractions to

characterize heterogeneous materials and help process development

Summary of aggregation characterization for ADC

SEC-MALS is useful to characterize/confirm the resolution of a SE-

HPLC method via the in-line measurement of average molar mass for

any given fraction (assay and process development support)

AUC is useful to characterize the size distribution of a sample using a

matrix-free separation mode, but results depend on fitting parameters

The two methods provide size information by different principles; each

has pros and cons, no “gold standard”

The case studies shown in this work indicate

– The multiple-site payload conjugations did not significantly increase

the aggregation propensity of the ADCs compared to the mAbs

– Low level new species smaller than dimer was observed, no new

species larger than dimer was above the detection limit

– The heterogeneity in conjugation products may be controlled by

process improvements

14

Case studies –

Characterization of ADC higher order structure

(HOS) using CD, fluorescence and DSC

15

Far-UV CD: ADC1 vs mAb1, ADC2 vs mAb2

16

De

(M-1

cm

-1)

Wavelength (nm)

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

195 205 215 225 235 245 255

hu6M024

PF-06647020

mAb2

ADC2

ADC1 vs mAb1 ADC2 vs mAb2

The multi-site payload conjugations did not alter the secondary structure of the mAbs

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

190 200 210 220 230 240 250

Δε

(M-1

cm-1

)

Wavelength (nm)

PF-06263507PF-06263507

huA1

mAb1

ADC1

17

Near-UV CD: ADC1 vs mAb1, ADC2 vs mAb2

-100

-80

-60

-40

-20

0

20

250 260 270 280 290 300 310 320 330 340 350De

(M

-1cm

-1)

Wavelength (nm)

mAb2

ADC2

ADC1 vs mAb1 ADC2 vs mAb2

The multi-site payload conjugations did

not alter the tertiary packing of the mAb1

in ADC1

Are the spectral differences indicative of

structural change in mAb2 by the payload

conjugation?

-40

-30

-20

-10

0

10

20

250 270 290 310 330 350

Δε

(M-1

cm-1

)

Wavelength (nm)

PF-06263507

huA1

mAb1

ADC1

18

Near-UV CD: Payload absorption in ADC2

-100

-80

-60

-40

-20

0

20

250 260 270 280 290 300 310 320 330 340 350

De

(M

-1cm

-1)

Wavelength (nm)

-95

-75

-55

-35

-15

5

250 260 270 280 290 300 310 320 330 340 350

De

(M

-1cm

-1)

Wavelength (nm)

mAb2+payload

ADC2

ADC2 vs (mAb2 + linker/payload)

The multi-site payload conjugations did not alter the tertiary structure of the mAbs

Linker/payload may also absorb, needs to be taken into account

ADC2 vs mAb2

mAb2

ADC2

Wavelength (nm)

250 275 300 325 350

-40

-30

-20

-10

0

10

20

Inotuzumab

F75626

20 °C

De

(M-1

cm-1

)

A

Near-UV CD: ADC3 vs mAb3

19

Wavelength (nm)

250 275 300 325 350

-200

-100

0

100

200

300 Inotuzumabozogamicin A

Inotuzumab

ozogamicin B

Inotuzumab

ozogamicin C

Inotuzumab

F75626

20 °CDe

(M-1

cm-1

)

B

mAb3 ADC3 vs mAb3

ADC3 (DAR=7.3)

ADC3 (DAR=6.6)

ADC3 (DAR=4.6)

mAb3

Linker/payload absorbance can dominate the spectrum, rendering the method

not suitable for structural characterization for the mAb in the ADC

Alternative method: Fluorescence spectroscopy

20

Fluorescence: ADC3 vs mAb3 Payload 3 CD and UV absorbance

Wavelength (nm)

350 400 450 500 550

Flu

ores

cen

ce I

nte

nsi

ty0

200

400

600

800

1000

inotuzumab

F75626

inotuzumab

ozogamicin

2007B0088

inotuzumab

ozogamicin

H87906

20 °C

A

mAb3

ADC3

mAb3 + payload 3

The linker/payload absorbance can also

affect fluorescence emission spectra

Giorgio, at al. (2005) Bioorganic

& Medicinal Chemistry 13:5072

excitation

wavelengths

Fluorescence spectra: ADC3 vs mAb3

21

Wavelength (nm)

350 400 450 500 550F

luo

resc

ence

In

ten

sity

0

200

400

600

800

1000

inotuzumab

F75626 in buffer

inotuzumab

ozogamicin

H87906 in buffer

inotuzumab

F75626 in 6M GdmCl

inotuzumab

ozogamicin

H87906 in 6M GdmCl

20 °C

Wavelength (nm)

350 400 450 500 550

Flu

ores

cen

ce I

nte

nsi

ty

0

200

400

600

800

1000

inotuzumab

F75626

inotuzumab

ozogamicin

2007B0088

inotuzumab

ozogamicin

H87906

20 °C

A mAb3

ADC3

mAb3 + payload 3

Native (formulation) ADC3 & mAb3 (peak normalized)

Denatured (6M GdmCl) ADC3 & mAb3

ADC3 vs mAb3 Native vs denatured

The fluorescence spectrum of ADC is superimposable with the peak-normalized

spectrum of the mAb, indicating the overall shape of the mAb is not altered in ADC

Using near-UV CD as a screening tool - a caveat

22

mAb1

Be cautious when interpreting the

apparent trends

-30

-25

-20

-15

-10

-5

0

0 1 2 3 4 5 6

Δe

(M

-1cm

-1) @

272 n

m

GuanidineHCl (M)

[GdmCl (M)]

PBS

mAb1 in GdmCl

@272 nm

-40

-30

-20

-10

0

10

20

260 280 300 320 340

Ele mAb 0MEle 1M mAbEle mAb 3MEle mAb 6M

A

0

272 nm

1 M

3 M

6 M

De

(M

-1cm

-1)

Wavelength (nm)

mAb1

6 M GdmCl

3 M GdmCl

FB (pH 6)

1 M GdmCl

6 M GdmCl

-40

-30

-20

-10

0

10

20

250 260 270 280 290 300 310 320

De

(M

-1cm

-1)

Wavelength (nm)

FB (pH 6)

FB + 150mM NaCl

PBS (pH 7)

Pfizer Confidential │ 23

Thermal stability: DSC profiles

0

40

80

120

160

30 40 50 60 70 80 90

Temperature (°C)

Cp

(kca

l\m

ole

\°C

)

ADC

huA1 mAb1 (IgG1)

ADC1 (Cys conj)

-5

20

45

70

95

120

145

35 45 55 65 75 85 95

Cp

(k

cal/

mo

le/

C)

Temperature ( C)

ADC1 and mAb1 exhibit different thermal

stability profiles; all Tm’s are >60ºC

Tm1 of ADC2 is lower than Tm1 of

mAb1; all Tm’s are >60ºC

mAb2 (IgG1)

ADC2 (Cys conj)

c 24

Thermal stability: DSC profiles

Temperature (°C)

30 40 50 60 70 80 90

0

50

100

inotuzumabozogamicin2007B0088

inotuzumabozogamicinH87906

inotuzumab F75626

Cp

(kcal·m

ole-

1 ·°C

-1)

ADC3 and mAb3 exhibit different thermal stability

profiles; all Tm’s are >60ºC

mAb3 (IgG4)

ADC3 (Lys conj)

Interpretations – structural perspectives

In general ─ Spectroscopic methods are non-destructive methods to assess the

HOS of proteins, reporting mainly the overall shape

─ DSC is a destructive method, applying heat to unfold proteins; it

reflects mainly the strength of side chain interactions in folded proteins

─ Proteins with similar shapes may have different side chain interaction

energies, manifest in different stabilities

25

For Pfizer ADC1, ADC2 and ADC3

─ The CD and fluorescence spectra of the ADCs are similar to that of their respective

mAbs, indicating that the protein conformations are not significantly altered by the

multi-payload conjugations

─ The DSC profiles of the ADCs are not the same as that of their respective mAbs,

indicating changes in amino acid side chain interactions due to the payload

conjugation and/or the conjugation process

─ The payload, IgG type, conjugation chemistry and conditions, payload locations,

etc., may affect the structural properties of the conjugates

Egg vs egg-

shaped rock

https://www.google.com/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&uact=8&ved=0CAcQjRw&url=https://www.etsy.com/market/granite_stone_egg&ei=T8jvVPrcHoqrggSSvoD4Dg&bvm=bv.86956481,d.eXY&psig=AFQjCNHpCxATudO74xOBuO_cYfV9OdAKJA&ust=1425086832511191http://www.google.com/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&uact=8&ved=0CAcQjRw&url=http://fytastock.deviantart.com/art/Perfect-Egg-5812482&ei=6MrvVLaQE8GLNr3AgPgM&psig=AFQjCNGNopFA8V-_0STlw_soFjcqK2FXNw&ust=1425087112866299

Interpretations – relation to function and safety

In general ─ Cytotoxic immune responses induced by mAb require the antigen-binding and the

Fc effector functions, therefore, require intact structure of the entire mAb

─ Cytotoxicity induced by ADC requires mAb binding to the antigen, therefore,

requires intact structure of the antigen-binding region

─ Once internalized, the payloads cause the cell death; therefore, the structure of

mAb other than the antigen-binding region is not critical to the function of ADC

─ However, a mAb in ADC with significantly altered structure may pose quality and

immunogenicity (anti-drug antibodies) issues for the ADC as biotherapeutics

26

Target

cell

Cell death

mAb binding ADC internalization Cytotoxicity

Target

cell

mAb binding

Cell death

ADCC

CDC

Cytotoxicity

mAb cytotoxicity MOA ADC cytotoxicity MOA

Fc effector

functions

DNA / scaffold

destruction

For Pfizer ADC1, ADC2 and ADC3

─ The antigen-binding of each ADC is comparable to the respective mAb,

indicating the first step of the ADC function is not compromised by conjugation

─ The overall shapes (conformation) of the mAbs are not altered by the

conjugations, suggesting little or no new surface is exposed; therefore, little or

no change in the immunogenicity of the protein (anti-drug antibodies)

─ The thermal stability may be affected by the conjugation; however, all Tm’s of

these ADCs are >60ºC, suggesting the HOS is stable under the storage

conditions and at the body temperature

─ These 3 ADCs should have the expected functionalities, with low safety

concerns from the HOS perspective

27

mAb1

ADC1

Antigen-binding ELISA for the ADCs vs the mAbs

mAb3

ADC3

mAb2

ADC2

Interpretations – relation to function and safety

Wavelength (nm)

350 400 450 500 550

0

200

400

600

Flu

ores

cen

ce I

nte

nsi

ty

Wavelength (nm)

200 210 220 230 240 250

-1

0

1

2

De

(M-1

cm-1

)

Biophysical methods for product comparability

28

-100

-80

-60

-40

-20

0

20

250 260 270 280 290 300 310 320 330 340 350

De

(M-1

cm-1

)

Wavelength (nm)

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

195 205 215 225 235 245 255

De

(M-1

cm-1

)

Wavelength (nm)

30 40 50 60 70 80 90

0

20

40

60

80

100

120

140

Demo1 A

Demo 1 B

Demo 1 C

GMP0 A

GMP0 B

GMP0 C

RM ADC A

RM ADC B

RM ADC C

Cp

(kca

l/m

ole

/oC

)

Temperature (oC)

Batch 1

Batch 2

Batch 3

Batch 1

Batch 2

Batch 3

Batch 1

Batch 2

Batch 3

ADC2 ADC2 ADC2

Far-UV CD Near-UV CD DSC

Far-UV CD Fluorescence

Batch 1

Batch 2

Batch 3

ADC3

Batch 1

Batch 2

Batch 3

ADC3

Acknowledgements

Lucas Wafer

Cliff Entrican

Marek Kloczewiak

Zhaojiang Lu

Sharada Sant

Peter Richard

Jamie Lee

Tom Lerch

Jim Zobel

29

Jason Starkey

Jim Mo

Heyi Li

Scott Allen

Olga Friese

Jason Rouse

Meg Ruesch

c 30

DSC profiles for IgGs

Garber and Demarest (2007) Biohem Biophys Res Commun. 355:751-7