Process Analytical Technology (PAT) for Bioprocessing · PDF fileProcess Analytical Technology (PAT) for Bioprocessing LCDR Cyrus Agarabi, PharmD, PhD, MBA Division of Biotechnology

Process Analytical Technology

(PAT) for Bioprocessing

LCDR Cyrus Agarabi, PharmD, PhD, MBA

Division of Biotechnology Review and Research-II

OBP/OPQ/CDER

June 8, 2016

*Disclaimer: The views expressed are presented by the authors for consideration. They do not

necessarily reflect current or future FDA policy or official views of the US Government.

Outline

Background

Bioprocessing & mAb Overview

PAT Research Examples and Opportunities

2

Office of

Program and

Regulatory

Operations

Office of Policy

for

Pharmaceutical

Quality

Office of

Biotechnology

Products

Office of New

Drug Products

Office of

Lifecycle Drug

Products

Office of

Testing and

Research

Office of

Surveillance

Office of

Process and

Facilities

Office of Pharmaceutical

Quality (OPQ)

Center for Drug Evaluation and

Research (CDER)

Food and Drug Administration

(FDA)

Why Regulatory Research?

• Outreach- Internal/External Stakeholders

• Discuss regulatory research to support FDA

guidance documents (PAT, QbD, etc)

• Reviewer Training

• Internal/external training on bioprocessing

equipment and analytics

• Support Biotech Regulatory Decisions

• Support policy development

• Support biotech review decisions 4

PAT Guidance Released September 29, 2004

Scientific principles and tools – Process Understanding

– PAT Tools

– Risk-Based Approach

– Integrated Approach

Regulatory strategy accommodating innovation

– Training

– Lab research

www.fda.gov/cder/gmp

How can this be applied to biotech?

5

The Essence of PAT Product quality is monitored and controlled during the

manufacturing process.

Process decisions are based on assessments of material

attributes.

Forward-feed of incoming material

In-process monitoring & control

Critical product attributes measured/assessed

Instantaneously (on-line, in-line, at-line) or

Before decision point (near at-line)

• With as large a window as feasible 6

7 *For illustrative purposes only- Data generated by Google Trends Based on searches for “Process Analytical Technology”

Google search interest for “Process Analytical

Technology” (2004-Current)

• Where do we go from here?

• How can PAT play an important role in new initiatives?

PAT Guidance Released

0

50

100

What is an Emerging Technology? Technology with which the FDA has limited review or inspection experience New (or not previously applied to pharmaceutical applications) to support more robust, predictable, and/or cost-effective processes or novel products Innovative or novel products, manufacturing processes, or analytical technology subject to CMC review Examples: •Continuous manufacturing of drug substances and drug products

•“On-demand” manufacturing of drug products

•Aseptic filling closed system

•3-D printed tablets

•New container and closure systems for injectable products

Emerging Technology Team (ETT)

A small cross functional team with representation from all relevant CDER review and inspection programs Encourage and support the adoption of innovative technology to modernize pharmaceutical development and manufacturing Two tracks:

• Stand alone innovative or novel product, manufacturing process, or analytical technology

• Existing or planned submission(s) with the above

• [email protected]

• Amendment to FDA Broad Agency Announcement for the Advance

Research and Development of Regulatory Science and Innovation

highlights Continuous Manufacturing as an area of interest

• Enabling Technologies for Continuous Manufacturing – Continuous processing equipment (e.g., crystallizers, coaters, and viral clearance)

– Enhanced in-line process analytical technologies

– Integrated data management and plant-wide control systems

– Process modeling and simulation

– Advanced process control (e.g., feedback, feedforward or plant-wide control)

• Continuous Manufacturing Innovation – Synthetic processes that would benefit from flow processing; syntheses that could be affected through a

reduced number of steps or that would not be feasible by batch production

– Modular or plug and play type equipment with re-usable or flexible, interchangeable parts that allows the

development of platform technologies for drug substance, drug product or end-to-end continuous processes.

– Non-column based chromatography and alternative purification techniques (e.g. continuous precipitation)

– Continuous processes for homogeneous production of final dosage forms (e.g., strip film manufacturing

system, injection molding, and printing).

– Alternatives to inherently batch unit operations (e.g. viral and sterile filtration)

– Process control systems with improved user interface (e.g. GUI) and capability for integration with new unit

operations and ancillary equipment, with reduced need for programmer hours

BARDA-FDA Continuous Manufacturing Innovations Initiative

Bioprocessing &

Monoclonal Antibody

Overview

12

Overview of Bioprocessing of Drug

Substance Protein A clarify concentrate

Cation

Exchange

Filter

Sterilize Anion

Exchange

TFF UFDF

Drug

Substance

centrifuge

Characterization Upstream

Downstream

seed

Culture

WASTE

LN2

Production culture (100’s- 10,000’s L bioreactor)

Thaw WCB

Expansion of cells (pre-culture)

13

Bioreactors (Upstream)

14

• Stirred Tank- (Traditional) • Tank with a motor driven

impeller or agitator

• Ports for probes & sampling

• Cleaned & steamed in place

• Simple scale-up

• Most common equipment

• Disposable Systems • Plastic bags or vessels

equipped with disposable plastic liners

• No cleaning required • Easy to scale-up & cost-

effective • Can be equipped with

optical sensor patches

Attributes & Combinatorics

• 2 x 6 x 6 x 4 x (10+5) x 2 = 8460 • (8460)2≈ 75 million

pyro-E

D

D

D

G

G

K

O

O

O K

pyro-E O

D

G

G

D

O

D

O

Courtesy of S. Kozlowski

• Pyro-Glu (2)

• Deamidation (3x2)

• Methionine oxidation (3x2)

• Glycation (2x2)

• High mannose,

Fucosylation G0, G1, G1,

G2 (10)

• Sialylation (+5)

• C-term Lys (2)

15

Aseptic autosampling of parallel

bioreactors

Integrated on-line analysis

Retains samples at (4oC) for

at-line and offline analysis

Integrated Bioprocessing System

5L Parallel Bioreactors (6)

Integrated analysis: pH, DO, Temp

Controls: Agitation, Acid/Base,

Heating/Cooling,

Gasses (O2, N2, CO2, Air),

Automated process adjustments

Automated Control &

Feedback In-Line

Real-time monitoring

of glucose/lactate,

Dialysis (no loss in

reactor volume),

Self-calibrating

Automated Glucose Feeding

Cell Counter /

Bioanalyzer

Chemistry & Gas, VCD, Cell

count, Osmometer, IgG module

On-Line Analysis

2D HPLC

Analysis Amino Acids

(media), Conc.,

Size Exclusion,

Impurities,

Affinity

UPLC

Analysis Conc., Size

Exclusion,

Impurities, Affinity

Future: Mass

Spectrometry

integration

At-Line Analysis

O2/CO2 Monitoring

Humidity/pressure compensation,

Real-time measurement of metabolic

processes (OUR, CER, RQ)

Reactor

Autosampler

w/ OPC

Server

ToF-MS

Analysis Comprehensive

antibody

characterization,

Glycosylation profile

Near Infrared

Real-time

monitoring of

products

chemical and

physical

properties (i.e.

homogeneity

screening)

Biomass Monitoring

Capacitance, Real-time viable cell density

16

In House Model Cell Lines Murine suspension hybridoma

- monoclonal IgG3k

• Serum free commercially

available media

Processing experience:

• Spinners (0.15-4L)

• Bioreactors 1L, 4L, 5L, 7L

• Air Wheel (15L)

Culture Length ~120 hrs

Observed Yields- 25-175mg/L

17

From Dwek et al., Annu. Rev. Immunol.2007,25:21

Chinese Hamster Ovary

(CHO) Cell (DG-44)

- monoclonal, chimeric IgG1

IgG1 binds murine Neisseria

meningitidis outer capsule

Limited small scale experience:

Shake flasks and AMBR

Source: http://www.laboratory-journal.com/science/life-sciences-biotechnologie/antibody-therapeutics

PAT Research Examples

and Opportunities

18

PAT Case Study 1: Nutrients

• Nutrient Content in Bioreactors can affect:

Cell Viability & Yield (Titer)

Glycosylation & Glycation

• Optimized and Controlled Feed Strategies

Improve CQAs and Yield

• Platform Approach – “One-Size doesn’t always fit all”

• PAT’s Role

↑ Scope, monitoring, and control of in-process nutrients, to

pull the correct “lever” to improve CQA’s

Yuk et. Al. (2011) Biotech. & Bioeng., Vol. 108, PP2600-10 19

Standard Approaches to Media

Composition Analysis • Current Standard- Automated Analyzers

Ion Specific Electrodes (ISE)

• Ammonia, Sodium, Potassium

Amperometric Electrodes- Immobilized

Enzymes

• Glucose, Lactate, Glutamine, Glutamate

Trypan Blue and Digital Imaging • VCD, Cell Count, Cell Size

www.novabiomedical.com

20

• Pro:

Minimal sample prep and operator expertise, automated, high throughput

Capable of integration with bioreactors and auto samplers

• Con:

Limited scope of analyte information

Real-time Nutrient Analysis

• BioPAT TRACE

• Glucose/Lactate

Direct measurement

Internal calibration

Orthogonal offset

• Dialysis mode

Cycles sample through in situ membrane

21

Nutrients Overview:

Feeding Strategy

22

Batch Fed-Batch Fed-Batch

Feed Type -- 1.5 mmol/L Gln 1.5 mmol/L Gln +

1X NEAA

Feed Volume -- 60 mL 60 mL

Feed Timing

(Automated)

-- 48, 72 h 48, 72 h

0

0.5

1

1.5

2

2.5

3

3.5

4

0 24 48 72 96 120

g/L

Time (hrs)

Batch 5 GlucosePU1 Trace

PU1 Nova

PU2 Trace

PU2 Nova

PU4 Trace

PU4 Nova

Nutrients Overview - Trace

Gln

Batch

Gln+NEAA

Batch mode consumed the least glucose.

GLN + NEAA consumed most glucose

23

Representative Viable Cell Density

24

Specific Productivity Batch 5

0.00E+00

1.00E+04

2.00E+04

3.00E+04

4.00E+04

5.00E+04

6.00E+04

Batch Gln Gln+NEAA

Spec

ific

Pro

du

ctiv

ity

(Qp

),

pg/

cells

/day

Feeding Strategies

25

26

Glycan Analysis

Feeding Strategy Overview

• Fed-batch cultures performed better

Increased viable cell density

Increased antibody production

Potential changes in glycan patterns

• On-line glucose analysis

Trends followed off-line analysis

Increased data density

Potential tool for feeding decisions

27

PAT Case Study 2*: VCD & Process Mode

ABER Biomass Probe:

•Dielectric Spectroscopy

technology

•Tiny capacitors under the

influence of an electric field

• Build up of charge→

capacitance measured

•Cells with intact plasma

membranes

• Directly proportional to the

membrane bound volume of

these viable cells.

28

ABER_VCD = ƒ(Norm_Cap , Inoculum density)

Baseline normalized (f-f0)

*In collaboration with MIT Engineering Practice School

Perfusion Control Scheme

29

FC XC FC

Nutrient Composition

Scale

Perfusion Apparatus

30

Harvest tank Feed tank

ATF

Scale

Harvest pump

Feed pump

Antifoam pump

Biomass Probe

Media and Sampling Plan

OptiCHO Medium – Commercial chemically defined medium

– NaHCO3 buffered

– 5 g/L glucose

– Supplemented with 10 mM glutamine

– 3% Antifoam C added periodically

Twice daily manual samples – Additional sample taken pre-perfusion

31

Culture Comparison: VCD

32

1.E+05

1.E+06

1.E+07

1.E+08

0 2 4 6 8 10 12 14

Via

ble

Ce

ll D

en

sit

y, c

ells

/mL

Culture time, d

ABER Perfusion

NOVA Perfusion

ABER Batch

NOVA Batch

ABER Fed-Batch

NOVA Fed-Batch

Culture Comparison: Glucose

33

0,0

1,0

2,0

3,0

4,0

5,0

6,0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Glu

co

se

(g

/L)

Time (day)

Perfusion

Batch

Fed-Batch

Fed Batch: Daily Bolus

Feeding Initiated

Perfusion Initiated

34

0

10

20

30

40

50

60

70

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

IgG

(u

g/m

L)

Time (day)

perfusion

batch

fed-batch

Culture Comparison: Product Yields

Total Product Yields

35

PAT Exercise Findings

• Nutrient Findings Feeding strategy was not sufficient for all

cultures

Automated feeding strategy developed from online data and feedback loops is needed

• DS Findings Probe was able to track trends and fluctuations

in cellular growth and viability

Requires development of an offset based on robust off-line data Adjustments for each cell line need to be made

36

Opportunities for PAT

• Real time decision making

Feeding: Timing, quantity, composition

Harvest: Go/No-Go

Gross contamination detection

Global culture health trajectory - Course

corrections

• Support Quality by Design (QbD)

• Support Multivariate Data Analysis (MVDA)

37

Acknowledgements

OBP:K. Brorson, S. Lute, B. Chavez, A. Williams, S.

Johnson, C.J. Hsu, J. Wang, M. Landreth, M. Brown,

D. Frucht

OPQ IO: A. Fisher

Academia: S. Yoon (UMass Lowell), K. Stein (MIT)

Funding Sources: CP, RSR, MCM

38