Process Validation in Manufacturing of Biopharmaceuticals (Biotechnology and Bioprocessing) 3rd Edition

Process Validation in Manufacturing of Biopharmaceuticals T H I R D E D I T I O N BIOTECHNOLOGY AND BIOPROCESSING SERIES Series Editor Anurag Rathore 1. Membrane Separations in Biotechnology, edited by W. Courtney McGregor 2. Commercial Production of Monoclonal Antibodies: A Guide for ScaleUp, edited by Sally S. Seaver 3. Handbook on Anaerobic Fermentations, edited by Larry E. Erickson and Daniel YeeChak Fung 4. Fermentation Process Development of Industrial Organisms, edited by Justin O. Neway 5. Yeast: Biotechnology and Biocatalysis, edited by Hubert Verachtert and René De Mot 6 . Sensors in Bioprocess Control, edited by John V. Twork and Alexander M. Yacynych 7. Fundamentals of Protein Biotechnology, edited by Stanley Stein 8. Yeast Strain Selection, edited by Chandra J. Panchal 9. Separation Processes in Biotechnology, edited by Juan A. Asenjo 10. LargeScale Mammalian Cell Culture Technology, edited by Anthony S. Lubiniecki 11. Extractive Bioconversions, edited by Bo Mattiasson and Olle Holst 12. Purification and Analysis of Recombinant Proteins, edited by Ramnath Seetharam and Satish K. Sharma 13. Drug Biotechnology Regulation: Scientific Basis and Practices, edited by Yuanyuan H. Chiu and John L. Gueriguian 14. Protein Immobilization: Fundamentals and Applications, edited by Richard F. Taylor 15. Biosensor Principles and Applications, edited by Löíefc J. Blum and Pierre R. Coulet 16. Industrial Application of Immobilized Biocatalysts, edited by Atsuo Tanaka, Tetsuya Tosa, and Takeshi Kobayashi 17. Insect Cell Culture Engineering, edited by Mattheus F. A. Goosen, Andrew J. Daugulis, and Peter Faulkner 18. Protein Purification Process Engineering, edited by Roger G. Harrison 19. Recombinant Microbes for Industrial and Agricultural Applications, edited by Yoshikatsu Murooka and Tadayuki Imanaka 20. Cell Adhesion: Fundamentals and Biotechnological Applications, edited by Martin A. Hjortso and Joseph W. Roos 21. Bioreactor System Design, edited by Juan A. Asenjo and José C. Merchuk 22. Gene Expression in Recombinant Microorganisms, edited by Alan Smith 23. Interfacial Phenomena and Bioproducts, edited by John L. Brash and Peter W. Wojciechowski 24. Metabolic Engineering, edited by Sang Yup Lee and Eleftherios T. Papoutsakis

Process Validation

Biopharmaceuticals

T H I R D E D I T I O N

1 Membrane Separations in Biotechnology, edited by W Courtney McGregor

2 Commercial Production of Monoclonal Antibodies: A Guide for Scale-Up,

edited by Sally S Seaver

3 Handbook on Anaerobic Fermentations, edited by Larry E Erickson and

Daniel Yee-Chak Fung

4 Fermentation Process Development of Industrial Organisms, edited by

7 Fundamentals of Protein Biotechnology, edited by Stanley Stein

8 Yeast Strain Selection, edited by Chandra J Panchal

9 Separation Processes in Biotechnology, edited by Juan A Asenjo

10 Large-Scale Mammalian Cell Culture Technology, edited by

Anthony S Lubiniecki

11 Extractive Bioconversions, edited by Bo Mattiasson and Olle Holst

12 Purification and Analysis of Recombinant Proteins, edited by Ramnath

Seetharam and Satish K Sharma

13 Drug Biotechnology Regulation: Scientific Basis and Practices, edited by

Yuan-yuan H Chiu and John L Gueriguian

14 Protein Immobilization: Fundamentals and Applications, edited by

Richard F Taylor

15 Biosensor Principles and Applications, edited by Löíefc J Blum and

Pierre R Coulet

16 Industrial Application of Immobilized Biocatalysts, edited by Atsuo Tanaka,

Tetsuya Tosa, and Takeshi Kobayashi

17 Insect Cell Culture Engineering, edited by Mattheus F A Goosen,

Andrew J Daugulis, and Peter Faulkner

18 Protein Purification Process Engineering, edited by Roger G Harrison

19 Recombinant Microbes for Industrial and Agricultural Applications,

edited by Yoshikatsu Murooka and Tadayuki Imanaka

20 Cell Adhesion: Fundamentals and Biotechnological Applications,

edited by Martin A Hjortso and Joseph W Roos

21 Bioreactor System Design, edited by Juan A Asenjo and José C Merchuk

22 Gene Expression in Recombinant Microorganisms, edited by Alan Smith

23 Interfacial Phenomena and Bioproducts, edited by John L Brash and

Peter W Wojciechowski

24 Metabolic Engineering, edited by Sang Yup Lee and Eleftherios T Papoutsakis

Series Editor

Anurag Rathore

Dane W Zabriskie

26 Membrane Separations in Biotechnology: Second Edition, Revised and

Expanded, edited by William K Wang

27 Isolation and Purification of Proteins, edited by Rajni Hatti-Kaul

and Bo Mattiasson

28 Biotransformation and Bioprocesses, Mukesh Doble, Anil Kumar Kruthiventi,

and Vilas Gajanan Gaikar

29 Process Validation in Manufacturing of Biopharmaceuticals: Guidelines,

Current Practices, and Industrial Case Studies, edited by Anurag Singh

Rathore and Gail Sofer

30 Cell Culture Technology for Pharmaceutical and Cell-Based Therapies,

edited by Sadettin S Ozturk and Wei-Shou Hu

31 Process Scale Bioseparations for the Biopharmaceutical Industry,

edited by Abhinav A Shukla, Mark R Etzel, and Shishir Gadam

32 Processs Synthesis for Fuel Ethanol Production, C A Cardona, Ó J Sánchez,

and L F Gutiérrez

33 PAT Applied in Biopharmaceutical Process Development And Manufacturing:

An Enabling Tool for Quality-by-Design, edited by Cenk Undey, Duncan Low,

Jose C Menezes, and Mel Koch

34 Stem Cells and Revascularization Therapies, edited by Hyunjoon Kong,

Andrew J Putnam, and Lawrence B Schook

35 Process Validation in Manufacturing of Biopharmaceuticals, Third Edition,

edited by Anurag Rathore and Gail Sofer

CRC Press is an imprint of the

Taylor & Francis Group, an informa business

Boca Raton London New York

Taylor & Francis Group

6000 Broken Sound Parkway NW, Suite 300

Boca Raton, FL 33487-2742

© 2012 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S Government works

Version Date: 20120404

International Standard Book Number-13: 978-1-4398-5094-7 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S Copyright Law, no part of this book may be reprinted, reproduced, ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.

transmit-For permission to photocopy or use material electronically from this work, please access www.copyright com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400 CCC is a not-for-profit organization that provides licenses and registration for a variety of users For organizations that have been granted a photocopy license by the CCC,

a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used

only for identification and explanation without intent to infringe.

Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com

Contents

Preface ix

Editors xiii

Contributors xv

Chapter 1 Guidelines.to.Process.Validation 1

Gail Sofer Chapter 2 Commentary.on.the.US.Food.and.Drug.Administration’s.2011. “Guidance.for.Industry,.Process.Validation.General.Principles and.Practices” 11

Hal Baseman Chapter 3 Applications.of Failure.Modes.and.Effects.Analysis. to Biotechnology.Manufacturing.Processes 51

Robert J Seely and John Haury Chapter 4 Process.Characterization 63

James E Seely Chapter 5 Scaled-Down.Models.for Purification.Processes:.Approaches. and Applications 89

Ranga Godavarti, Jon Petrone, Jeff Robinson, Richard Wright, Brian D Kelley, and Glen R Bolton Chapter 6 Adventitious.Agents:.Concerns.and.Testing.for. Biopharmaceuticals 141

Raymond W Nims, Esther Presente, Gail Sofer, Carolyn Phillips, and Audrey Chang Chapter 7 Lifespan.Studies.for Chromatography.and.Filtration.Media 159

Anurag S Rathore and Gail Sofer Chapter 8 Validation.of.a.Filtration.Step 185

Jennifer Campbell

Narahari S Pujar, Marshall G. Gayton, Wayne K. Herber,

Chitrananda Abeygunawardana, Michael L. Dekleva,

P. K. Yegneswaran, and Ann L Lee

Lynn Conley, John McPherson, and Jörg Thömmes

Chapter 18 Viral.Clearance.Validation:.A.Case.Study 491

Michael Rubino, Mark Bailey, Jeffrey C Baker, Jeri

Ann Boose, Lorraine Metzka, Valerie Moore, Michelle

Quertinmont, and William Wiler

Preface

This.updated.edition.of.Process Validation in Manufacturing of Biopharmaceuticals.

cussed.and.illustrated.with.industrial.case.studies

provides.insights.into.current.guidelines.and.expectations Current.practices.are.dis-opment, process qualification, and continuous monitoring—an approach that is.intended.to.provide.even.greater.assurance.that.a.process.will.perform.consistently

Chapter.1.presents.an.overview.of.the.new.validation.paradigm.of.process.devel-to produce a drug substance with its requisite critical quality attributes (CQAs) Some.of.the.current.process.validation.concerns.are.presented

Chapter 2 provides background and industry commentary on the final version.of.the.Food.and.Drug.Administration.(FDA).2011.“Guidance.for.Industry,.Process.Validation.General.Principles.and.Practices,”.commonly.referred.to.as.the.Process.Validation.Guidance.or.PVG,.which.was.issued.in.final.form.on.January.24,.2011 Understanding.the.background.and.intent.of.the.PVG.should.make.it.easier.to.develop.plans.that.are.aligned.with.the.recommendations.presented.in.the.PVG.and.result.in.more.effective.process.validation.approaches

sis [FMEA]) is presented as a means to prioritize process parameters for further.process.characterization.prior.to.validation FMEA.provides.a.logical.approach.that.can.aid.in.establishing.critical.parameters.and.ensure.process.robustness Specific.examples.on.the.use.of.FMEA.will.aid.readers.in.establishing.this.method.in.their.own.organization

In.Chapter.3,.the.use.of.a.risk.assessment.method.(failure.modes.and.effect.analy-Process.characterization.is.a.prerequisite.for.process.validation In.Chapter.4,.a.description.of.how.to.carry.out.thorough.and.consistent.process.characterization.is.presented Precharacterization.studies,.which.are.used.to.help.define.the.scope.of.the.actual.experimental.characterization.work,.are.also.discussed The.discussions.on.timing.of.process.characterization,.needed.resources,.and.a.stepwise.approach.provide.valuable.insights The.importance.of.scale.down.in.process.characterization.is.also.addressed

Accurately.scaling.down.to.mimic.manufacturing.processes.is.essential.in.several.aspects.of.process.validation Chapter.5.provides.further.guidance.and.strategies.for.scaling.down.unit.operations,.including.chromatography,.chemical.modification.reac-tions,.ultrafiltration,.and.microfiltration In.addition.to.general.scale-down.principles.and.parameters,.the.authors.address.specific.problems.and.present.some.examples.Prior.to.establishing.a.process.that.can.be.validated,.it.is.essential.to.consider.potential risks from adventitious agents, which include viruses, bacteria, fungi,.mycoplasma,.and.transmissible.spongiform.encephalopathies The.potential.sources.of.these.agents.and.testing.programs.for.them.are.described.in.an.updated.Chapter.6 Current.examples.of.contamination.events.in.biopharmaceutical.manufacturing.are.presented Bioburden.assessment.and.sterility.issues.are.also.addressed,.and.a.sum-mary.table.describes.adventitious.agents,.recommended.tests,.and.stages.at.which.to.perform.testing

In Chapter 7, the lifespan of both chromatography and filtration media is.addressed There.are.discussions.on.the.various.factors.that.influence.lifespan,.and.experimental.approaches.for.validation The.use.of.small-scale.models.for.validation.is.discussed The.application.of.concurrent.validation.to.provide.lifespan.data,.an.approach.that.is.gaining.more.acceptance,.is.discussed.in.this.chapter.

idation that can be performed in scaled-down studies as well as those aspects that.require.manufacturing.scale Next.is.a.section.on.the.validation.of.sterilizing.grade.filters Subsequent.sections.address.validation.of.filters.used.for.clarification.and.virus.removal.filters Details.of.tangential.flow.filter.validation.are.presented Also.included.are.descriptions.of.specific.validation.issues.in.clarification.of.bacterial.cell.harvest.and.lysate.clarification,.mammalian.cell.clarification,.and.protein.concentration.and.diafiltration Cleaning.validation.for.reusable.membranes.is.also.discussed

Chapter.8.begins.with.an.overview.of.filtration.validation.and.a.discussion.of.val-It.has.been.said.that.without.assays,.you.have.nothing In.an.updated.and.current.Chapter.9,.analytical.test.methods.are.discussed.with.a.special.focus.on.well-char-acterized.biological.and.biotechnological.products Appropriate.methods.for.testing.raw.materials.and.in-process.samples.during.the.various.manufacturing.steps.are.addressed The.authors.also.discuss.process.analytical.technology.(PAT),.which.is.being.driven.by.the.FDA.as.a.means.to.better.control.processes Another.section.of.this chapter presents methods used for product characterization, release, and sta-bility.testing Also.included.are.the.ever-problematic.potency.assay.and.strategies.for.choosing.a.quality.control.testing.scheme Other.topics.discussed.are.the.use.of.assays.for.demonstrating.comparability,.assay.validation,.dealing.with.out-of-speci-fication.(OOS).results,.and.assay.revalidation

In Chapter 10, the reader is provided with a regulatory perspective on facility.design.and.validation.issues Written.by.two.former.FDA.employees,.this.chapter.provides details on the regulatory requirements and the information that should.be.provided.in.a.license.application Also.presented.are.the.requirements.for.cell.inoculum.suites.and.areas.intended.for.fermentation/harvest,.purification,.and.bulk.filtration In addition, support areas, such as those used for preparation of media.and.buffers,.and.the.use.of.closed.systems.to.reduce.environmental.classifications.are.discussed There.are.extensive.sections.on.utilities,.cleaning,.and.environmen-tal.monitoring Multiproduct.facility.issues.are.addressed In.the.section.on.facility.inspections,.the.authors.provide.insight.into.the.current.focus.of.inspections.Chapter.11.has.been.updated.and.discusses.the.importance.of.taking.a.risk-based.approach toward computerized system compliance and how it adds value to the.product.and.process.that.is.commensurate.with.cost It.is.concluded.that.a.sound.computer.system.validation.(CSV).program.encourages.the.introduction.of.new.and.exciting.technologies.with.the.ultimate.promise.of.safer,.more.effective,.and.more.affordable.medicines

Today, many firms are dependent on contract manufacturing organizations.(CMOs).to.perform.process.validation

Chapter.12.presents.strategies.for.a.team.approach.to.ensure.process.validation.is.carried.out.according.to.current.expectations A.table.that.delineates.the.responsi-bilities.of.the.sponsor.and.CMO.provides.a.practical.tool

The.application.of.risk.management.in.validation.is.described.in.Chapter.13

ters.are.illustrated.with.case.studies,.including.two.that.are.new First,.we.learn.in.Chapter.14.about.process.validation.for.membrane.chromatography Chapter.15.pro-vides.a.case.study.on.leveraging.multivariate.analysis.tools.to.qualify.scaled-down.models In.Chapter.16.a.matrix.approach.for.process.validation.of.a.multivalent.bac-terial.vaccine.is.described Chapter.17.addresses.purification.validation;.in.this.case,.for a therapeutic monoclonal antibody that is expressed and secreted by Chinese.hamster.ovary.(CHO).cells Chapter.18.describes.viral.clearance.validation.studies.for.a.product.produced.in.a.human.cell.line.

In.Chapters.14.through.18,.many.of.the.concepts.described.in.the.previous.chap-We.hope.this.updated.book.will.provide.the.reader.with.valuable.insights.into.the.current.trends.in.process.validation By.sharing.their.knowledge,.the.authors.have.contributed.to.the.biopharmaceutical.industry’s.enhanced.application.of.science-.and.risk-based.approaches.to.process.validation

Editors

Anurag S Rathore is.a.consultant.for.Biotech.CMC.Issues He.is.also.a.faculty.

member.of.the.Department.of.Chemical.Engineering,.Indian.Institute.of.Technology,.Delhi,.India Prior.to.that.he.held.management.positions.at.Amgen.Inc.,.Thousand.Oaks, California and Pharmacia Corp., St Louis, Missouri His areas of interest.include process development, scale-up, technology transfer, process validation,.process analytical technology and quality by design He has authored more than

180 publications and presentations in these areas He is presently serving as the

editor-in-chief of Preparative Biochemistry and Biotechnology and serves on the editorial advisory boards for Biotechnology Progress, BioPharm International,

Pharmaceutical Technology Europe and Separation and Purification Reviews Dr Rathore has edited books titled Quality by Design for Biopharmaceuticals:

Perspectives and Case Studies (2009),.Elements of Biopharmaceutical Production (2007),.Process Validation.(2005),.Electrokinetic Phenomena.(2004).and.Scale-up

and Optimization in Preparative Chromatography.(2003) He.has.a.PhD.in.chemical.engineering.from.Yale.University

Gail Sofer has retired after serving as director of regulatory compliance for.

GE Healthcare Prior to that, she served as the director of regulatory services at.BioReliance Her publications include numerous articles and book chapters on.downstream.processing,.virus.inactivation,.and.validation,.and.she.has.coedited.and.authored.several.books She.served.on.the.Science.Advisory.Board.of.The.Parenteral

Drug Association (PDA),.the editorial.advisory.boards of BioPharm, BioQuality,.

and BioProcess International , and the Scale-Up Advisory Board of Genetic

Engineering News In.addition,.Sofer.chaired.a.PDA.task.force.on.virus.filters.and.the.Biotechnology.Advisory.Board,.as.well.as.serving.on.the.Board.of.Directors She.cochaired.the.American.Society.for.Testing.and.Materials.(ASTM).subcommittee.on.Adventitious.Agents.for.Tissue.Engineered.Medical.Products.and.holds.an.MS.degree.in.biochemistry.from.the.University.of.Miami

Lynn Conley

Biogen.IdecRaleigh,.North.Carolina

Susan Dana Jones

BioProcess.Technology.Consultants,.Inc

Woburn,.Massachusetts

Anurag S Rathore

Indian.Institute.of.TechnologyNew.Delhi,.India

Suma Ray

Sartorius.Stedim.India.Pvt.LtdBangalore,.Karnataka,.India

Nadine Ritter

Biologics.Consulting.GroupAlexandria,.Virginia

Process Validation

Gail Sofer

1.1  INTRODUCTION

In the first edition of this book, published in 2000, we posed the question: Why did

we decide to produce yet another book on validation? The same response is still applicable: Guidelines addressing validation are usually purposefully broad to allow for the variability in products, manufacturing methods, analysis, clinical indica-tions, patient populaindica-tions, and doses for biopharmaceuticals As a result, there is still much discussion related to validation approaches and specific issues that must be addressed to satisfy regulatory authorities and produce safe and efficacious biophar-maceuticals Furthermore, developing technologies, both analytical and manufactur-ing, can impact validation and it is expected that sponsors of biopharmaceuticals will remain current with new developments

Today, there is a new validation paradigm that reinforces a comment made in the previous edition that although validation is a regulatory requirement for licensed biopharmaceuticals, it also provides an economic value Understanding a process and controlling it within realistic ranges minimizes batch failures The US Food and Drug Administration (FDA) “Guidance for Industry, Process Validation: General Principles and Practices” defines three specific stages of validation: Process Design, Process Qualification, and Continued Process Verification.1 The guidance defines process validation as the collection and evaluation of data, from the process design

stage throughout production, which establishes scientific evidence that a process is

capable of consistently delivering quality products A few pertinent quotes from this draft guidance include

CONTENTS

1.1 Introduction 1

1.2 Current Validation Citations and Problems 2

1.3 Validation: Today and Tomorrow 6

1.3.1 Today 6

1.3.2 Tomorrow 7

References 9

• A successful validation program depends upon information and knowledge from product and process development.

• Focusing on qualification efforts without understanding the manufacturing process may not lead to adequate assurance of quality

• We recommend an integratedteam approach to process validation that includes expertise from a variety of disciplines, including process engi-neering, industrial pharmacy, analytical chemistry, microbiology, statistics, manufacturing, and quality assurance (p 2)

While not specifically mentioning the term Quality by Design (QbD), the ance references the International Conference on Harmonization (ICH) Q8 on development, and it is very clear that the document encourages the use of good devel-opment practices to support process control strategies derived using a systematic and science- and risk-based approach.2

guid-As noted in an ICH Q&A document, there is a significant benefit to using the QbD approach “For products developed following the minimal approach, the control strategy is usually derived empirically and typically relies more on dis-crete sampling and end product testing Under QbD, the control strategy is derived using a systematic science and risk-based approach Testing, monitoring or con-trolling is often shifted earlier into the process and conducted in-line, on-line or at-line testing.”3

A uniform approach to validation and avoidance of the pitfalls can provide ther economic advantage However, biopharmaceuticals encompass some vastly dif-ferent products—not only therapeutic monoclonal antibodies and proteins produced

fur-by recombinant DNA technology but also gene and cell therapies Is it possible to apply a consistent validation approach that is applicable to, for example, monoclo-

nal antibody products produced in sources as diverse as cows’ milk and E coli?

The answer in some respects is yes Certain practical steps can be applied for all therapeutic products A risk assessment is the starting point for determining how the manufacturing process should be designed so that it can, in fact, be validated And there must be sufficient resources, both human and financial, applied to validation Good science and common sense are essential, and basic regulatory requirements should be reviewed and followed

1.2  CURRENT VALIDATION CITATIONS AND PROBLEMS

The US Freedom of Information Act (FOI) benefits the biotechnology industry as it tries to anticipate validation issues that are of concern, at least to the US FDA FDA approval letters, Form 483s, and warning letters can be useful in trying to make sense of the latest validation issues This is not, however, the ideal way for industry

to determine what is appropriate Each product and its production method are unique

in at least some aspects and the risk assessment and good science should be the driving force in understanding validation requirements However, reviewing recent regulatory citations can be an interesting beginning if we keep in mind that we don’t have the full picture and that even reviewers writing 483s make mistakes

A review of the BioQuality database of Form 483s from 2008 through August

2010 revealed that problematic issues in process validation are very much the same

as they were five years ago.4 Validation failures still seem to fall into overlapping groups, some of which are presented here Some are due to lack of resources and upper management buy-in Those sponsors with multiple problems, many of which are related to a lack of thorough process validation, often appear to have an upper management that focuses on short-term profits and forgets that they or someone in their family might have to actually use the product Employees with experience and good intent often have to do a really good internal selling job to ensure validation is performed properly

Variability: During development, it is essential to study the potential sources of

variability Yet, today we still find citations such as control procedures are not lished that validate the performance of those manufacturing processes that may be responsible for causing variability in the characteristics of in-process material and drug product One sponsor received the following comment: “Repeated in-process failures and batch adjustments indicate that the manufacturing is not a validated process.” These comments sound like findings from organizations that did not have the structure

estab-to ensure validation was performed according estab-to regulaestab-tory requirements For small, start-up biopharmaceutical companies and academic institutions, lack of understand-ing of the regulations and, often more importantly, lack of resources lead to such cita-tions In addition, companies manufacturing older vaccines and blood products are often not current with the latest expectations This is not to imply that all companies must apply the new validation paradigm for older processes, but as changes are made to existing processes, new validation approaches and current methods should be applied

Bioburden: Bioburden control and related regulatory expectations in the

manu-facture of biopharmaceuticals have certainly raised concerns for a long time Some firms have claimed a high limit, e.g., 100 CFU/ml, is acceptable when they have consistently found only 10 CFU/ml If, in fact, one lot had 100 CFU/ml, it could overload process capabilities Even if the bacteria are inactivated, residual unantici-pated or unknown toxins might be co-purified with the product One has to wonder how a process could be validated for 100 CFU/ml, if this had never been seen Are spiking studies a realistic approach to solving this dilemma? Probably not, since it

is not likely feasible to measure all potential contaminants and their by-products Furthermore, the microorganisms introduced in manufacturing might be different from those used in the spiking study But a generic/family approach to validation

of sanitization agent capability has proved to be valuable.5 Such spiking studies, however, do not replace the need for validated in-process monitoring More recently,

one firm was told that their process validation protocol was deficient since it did not

include assessment of endotoxins or bioburden. Another was cited for a cleaning

validation report that was inadequate as it failed to include the testing of final rinse

water for bioburden.

Cleaning: Cleaning validation was an issue cited during several inspections and

described in the previous edition of this book It has remained one of the more nificant issues in the last five years Multiuse facilities and chromatography pro-cesses seem to draw the most concern One sponsor had no cleaning validation for cleaning critical manufacturing areas Yet another was observed to have no cleaning

sig-validation for laminar air flow hoods used for the pre-culture inoculation process In

another situation, it was observed that the manufacturer had not conducted cleaning validation to demonstrate that a cleaning detergent/antifoam agent could effectively remove an unidentified substance that accumulated on a column resin and interfered

with column packing A more recent observation pointed out that there was no

jus-tification for only executing one cleaning validation run for each piece of

manufac-turing equipment used for drug product and drug substance Another firm was cited

for not being able to demonstrate that worst-case cleaning validation was executed

Hold Times: Comments on hold time studies, or the lack thereof, seem to

indi-cate this is an issue often cited A reviewer noted that there were no hold time studies

for buffers and rinse solutions used in production Hold times are clearly an essential

element for ensuring consistent manufacturing, yet with inadequate validation plans and minimal resources, they are sometimes overlooked

Documentation and oversight: Documentation and oversight are critical areas In

one relatively recent case, a validation study was performed without the protocol being preapproved by quality assurance Another firm was cited for inadequate process valida-tion protocols and procedures Critical steps and critical parameters were not identified

in protocols, and protocols did not identify in-process tests for evaluating critical steps

Analysis: A process cannot be validated without validated analytical methods So

how did one sponsor think they could proceed with no acceptance criteria for ing analytical methods, deleted data, and missing sections that made it impossible to

validat-assess the results? A more recent citation referred to a computer program used to

elec-tronically calculate high-performance liquid chromatography (HPLC) data that was not validated and calculations performed by computer were not checked for accuracy

Unit Operations: The rush to be first to market is always a challenge and the

shortage of experienced personnel also causes oversights that lead to process tion mistakes Validation should be designed into the production process, but this requires time to understand the risks associated with each unit operation, cell sub-strates, and raw materials, as well as the expected results from the fermentation/cell culture and purification processes that will minimize those risks

valida-Citations related to fermentation processes that were not properly validated include the comment that production time limits had not been established for inocu-lum fermentation Another comment in the fermentation area related to hold times

In this case, there were no microbiological data supporting specified hold time for

autoclaved fermentation vessels In yet another, it was observed that the fermentation

process was validated to last for a certain amount of time, but batches were

termi-nated before the specified time due to contamination.

The validation of chromatography remains a source of reviewer comments Column lifetime, storage, and cleaning are all linked An assessment of carryover and its risks are important elements that should be included in the validation plan Validation of column storage times is a critical area, and it was observed at one manufacturer that there were no column storage time studies including bioburden and Limulus Amebocyte Lysate (LAL) determination In one postapproval inspec-tion, the FDA reviewer commented that the cleaning validation study was only con-ducted up to five uses of the column but the column could be used up to 46 lots based

only on a laboratory study In a review letter, a sponsor was asked to please provide

validation data to demonstrate there is no negative impact of extended use up to

150 production cycles on the efficacy of cleaning and regeneration of the column Column packing has raised some comments such as there were no studies on pack-

ing of purification columns In another situation, it was observed that a column was

consistently out of specification for a test, and there was no evidence that validation was reviewed to verify performance within the specification

Comments over the last three years are not much different:

• Validation has not been completed for sanitization and storage of tion columns

purifica-• No cleaning validation for entire lifespan of purification columns

• No studies to generate data supporting storage of columns between uses

• No bateriostasis or fungistasis studies have been conducted on buffers

• Cleaning validation of chromatography skid is inadequate as it did not include the assessment of endotoxin or bioburden

Validation of filter reuse has also drawn some attention In one case, standard

operating procedures (SOPs) were generated without validation data for the cation the operator was expected to meet It was observed that the SOP required fil-ters to be replaced after a specified time or after a defined number of production runs

specifi-No data were generated to support the requirement for 200 runs It is not uncommon

to find that filtration needs to be repeated during manufacturing Reprocessing of tration requires validation, but one citation read reprocessing (e.g., re-filtration) was

fil-performed without validated reprocessing procedures More recently, it was noted

that a microfiltration skid membrane was not validated for reuse, and validation of cleaning of ultrafiltration system failed since samples of both permeate and retentate

failed to meet predetermined acceptance criteria for TOC and conductivity.

Scaled-down validation studies: When scaled-down units are used in process

validation, the models themselves must be qualified (or validated) to ensure they reflect manufacturing results One firm was cited for using model vessels for clean-ing validation that did not adequately demonstrate the effectiveness of cleaning of the actual production vessels For clearance studies, purity and impurity profiles should be the same at both scales Viral clearance and microorganism sanitization studies must be done outside of the actual facility, often resulting in differences in operators, buffer preparation, and unit operations All too often there is a disconnect between those who perform validation and clearance studies and personnel in manu-facturing, leading to the inconsistencies cited during regulatory review

Small-scale studies can be invaluable for predicting resin lifetime FDA persons have noted that small-scale studies are useful for determining resin lifetime, but that failure to monitor during manufacturing is unacceptable, as evidenced by the observation that there was no concurrent validation of column performance at full-scale production For other validation concerns, such as viral clearance, small-scale studies remain the only viable option at this time

spokes-Viral clearance validation/evaluation studies have been problematic since the first biotherapeutics were produced The reasons are manyfold Among those reasons are safety issues, sensitivity and inhibition of infectivity assays, scale-down accuracy,

effect of spike on process, cost of studies, and data interpretation In the past, most sponsors have waited until they were almost ready to begin clinical trials to perform viral clearance studies Polymerase chain reaction (PCR) now provides the process development scientist with a more rapid, more sensitive, and less costly alternative that allows for the assessment of viral clearance capabilities during development Clearance studies should also address sanitization studies One sponsor was asked

to provide data that show complete removal of viral contamination prior to reuse of

the system

In the FDA’s “Points to Consider in the Manufacture and Testing of Monoclonal Antibody Products for Human Use,” it is noted that an appropriately conducted clearance study may be an acceptable substitute for lot-to-lot testing for potential contaminants and additives.6 Validated clearance studies combined with final prod-uct testing on at least three production lots can significantly reduce quality con-trol costs and expedite product release However, it is essential that these studies

be repeated when process changes with potential for changing clearance are made, since these clearance studies are often a critical element in ensuring patient safety The capability to perform the validated assays used for the clearance studies must be maintained This can be problematic during clinical studies in the event that the one analyst who can perform the method leaves the company

Changes and validation: Firms are sometimes observed to make changes

subse-quent to or during validation, as in the case where changes were made to equipment cleaning procedures subsequent to validation, but equipment cleaning procedures were not revalidated to establish the impact of the changes on the process In another case, process parameters were changed during validation, but the master batch record was not changed to reflect these modifications, and the precise nature and

time of these changes was not included in the validation report Changes are desired

to reduce failures, improve product quality, and improve productivity and economy The application of QbD and design space can facilitate the ability to improve pro-cesses after licensure without extensive revalidation and lengthy preapproval times With a more traditional approach, it is more difficult to make changes For example, one traditional vaccine manufacturer was observed to lack a quality assurance sys-tem that requires the timely revalidation of processes whenever there are changes in formulation and processes, which could have impact on the effectiveness or product characteristics, and whenever there are changes in product characteristics

1.3  VALIDATION: TODAY AND TOMORROW

Validation begins with good process development It requires that process developers understand the necessity to design a process that will be capable of ultimately meet-ing predetermined specifications without being subject to deviations within a defined range of preset operating parameters Development reports are invaluable when pro-cess changes are to be implemented and validated But it has also been observed that companies usually put constraints on the time allotted for development, and the development reports are often not very effective This is in spite of the fact that other

companies find the development reports a means to expedite reviewer understanding

of critical process parameters, which can lead to a reduced regulatory burden In fact, the Common Technical Document requires a development summary.7

Technology transfer from process development to pilot and/or manufacturing

is a challenge The frustration level is high when a manufacturing process change

is made that invalidates previously validated studies It’s a two-way process, ever The process developers must understand manufacturing capabilities One frus-trated manufacturing head noted that every process developer should spend a year

how-in manufacturhow-ing

Validation documentation is extensive and includes master validation plans, validation protocols, and validation reports A master validation plan is a useful essential, and is now specified in EU Annex 15 as a requirement.8 This Annex to the EU “Guide to Good Manufacturing Practice” provides an overview of several validation-related documents You are less likely to overlook validation items that are specified in a plan As noted by one Center for Drug Evaluation and Research (CDER) compliance officer, validation master plans and documentation will still be critical components of GMP compliance in FDA’s new GMP initiative He went on

to state that firms ought to start viewing process validation not just as a step in the manufacturing process, but an ongoing activity from design to testing and continu-ous improvements.9

Validation protocols are also an essential basic The protocol must state what will

be done, how it will be done, and what the outcome must be for the validation to be a success Validation cannot be just going back to a process step repeated three times and stating that it is validated

As noted above, validation is an ongoing process It is not a one-time effort that can then be ignored For biotherapeutics, most validation is performed prospectively, that is, prior to market approval However, today there is more acceptance of also using concurrent validation for some aspects Certainly, data collected at the manu-facturing scale can be more relevant provided that in-process analysis is sufficiently sensitive

New technologies and a risk-based approach applied to biopharmaceutical facturing are enabling more in-process monitoring The FDA is driving Process Analytical Technology (PAT) and, although more commonly used for synthetic drugs, some firms producing biopharmaceuticals are already using it Newer, highly sensitive, at-line, on-line, and in-line measurements allow more control of processes Although unlikely to replace the need for prospective validation, it has the potential

manu-to reduce that effort PAT has been used in fermentation control Cell viability has been measured by nicotinamide adenine dinucleotide (NAD)/NADH fluorescence; total cell counts by turbidity and optical density–based sensors; product and nutrient concentration by HPLC, ion chromatography (IC), near-infrared (NIR), and infrared (IR); and respiratory quotient by off gas analysis with mass spectrometry, pH, DO2, DCO2 In purification columns, PAT has been used to provide feedback of gradient control by NIR, ultraviolet (UV), and conductivity HPLC and a UV sensor have also

been used to determine when to collect product.10 As noted by Dr Kathyrn Zoon, elements of PAT could be applicable even to traditional biologics, like plasma deriva-tives Dr Zoon also commented that “PAT could be used for online monitoring of adventitious agents found in biotech therapeutics.”11

Another interesting validation approach is the use of generic or modular clearance studies Several publications have provided significant data that may lead to regula-tory acceptance of these studies for viral clearance A generic retrovirus low-pH inactivation study was performed and it was shown that bracketed generic conditions were sufficient to inactivate X-MLV in cell-free intermediates produced in either NS0 or Chinese hamster ovary (CHO) cell substrates Both monoclonal antibody and recombinant protein processes were evaluated In all cases, when the bracketed con-ditions were adhered to, a log reduction value of ³ 4.6 log10 was obtained.12 In another study with monoclonals, a generic/matrix chromatography virus removal step was evaluated on Q-Sepharose Fast Flow The column was operated in a flow-through mode so that the virus, not the product, would bind The clearance of SV-40 was shown to be ³ 4.7 log10 in three model antibodies with pIs ³ 8.8.13 Furthermore, the data were consistent in resins reused more than 50 times Caution should be taken, however, since this approach is not universally applicable With other, more complex separation modes, the generic/matrix approach may not work, and at this time the approach is not accepted by regulatory agencies

As more data is accumulated and disseminated, the biotechnology industry is finding it easier to apply risk assessments, QbD, experimental design (DOE), and design space, and maintain validated manufacturing processes Technical Reports from the Parenteral Drug Association (PDA) and other organizations provide insight into current practices in areas such as cleaning validation.14 Today, viral clearance

is better understood for certain types of products, and target values can be designed into a process A journal article from the Office of Biotechnology Products, CDER/FDA, describes a viral clearance database that contains product information, unit operation process parameters, and viral clearance data from monoclonal antibody and antibody-related regulatory submissions to the FDA The FDA authors pre-sented averages and ranges of viral clearance results by Protein A and ion exchange chromatography steps, low pH chemical inactivation, and virus filtration, focusing

on retro- and parvoviruses.15

Why then hasn’t the new validation paradigm been adopted by all? There are many potential advantages, such as fewer failed batches—a big cost saver—and potentially enhanced patient safety due to a better understanding of critical qual-ity attributes (CQA) and more consistent production within defined ranges of criti-cal process parameters (CPP) But these advantages come with a cost many firms can’t afford or are not willing to pay, such as potentially longer and more expensive development times, and restructuring of company protocols and even departments For new products and new companies, this is a real opportunity For older firms, processes, and products, change is usually more difficult However, as more regula-tory relief is granted to make more rapid and more frequent process improvements without costly delays; as more information is shared in the industry; and as techno-logical improvements arise, especially in analysis, change in validation will likely

be implemented

4 BioQuality http://www.bioquality.biz/subscribers.htm.

5 Hiraoka-Sutow, M., and C Broughton “Validating the Sanitization of Chromatographic

Resins: A Sample Case Study.” BioPharm (April 2001): 26–30.

6 US Food and Drug Administration (FDA) “Points to Consider in the Manufacture and Testing of Monoclonal Antibody Products for Human Use,” 1997.

7 The Gold Sheet 37 (2003): 3–4.

8 “Final Version of Annex 15 to the EU Guide to Good Manufacturing Practices, Qualification and Validation.” 2001 http://www.gmp-compliance.org/eca_guidelines_ search2_annex%2015.html.

and Recombinant Proteins.” Biotechnol Bioeng 82 (2003): 321–329.

13 Curtis, S., K Lee, G S Blank, K Brorson, and Y Xu “Generic/Matrix Evaluation

of SV40 Clearance by Anion Exchange Chromatography in Flow-Through Mode.”

Biotechnol Bioeng 84 (2003): 179–186.

14 Parenteral Drug Association (PDA) Technical Report No 49: Points to Consider for

Biotechnology Cleaning Validation Bethesda, MD: Parenteral Drug Association, 2010, www.pda.org.

15 Miesegaes, G., S Lute, and K Brorson “Analysis of Viral Clearance Unit Operations

for Monoclonal Antibodies.” Biotechnol Bioeng 106 (2010): 238–246.

Impact 182.2.7 General Commentary on PVG Scope and Purpose 192.3 Part 3: 2011 Final Version of the PVG 202.3.1 Legacy Processes 262.3.2 Re: ASTM E2500-07 352.4 Part 4: Final Thoughts 47References 48

iments and tests needed to validate processes Some have expressed the opinion that at times regulatory requirements are not aligned with scientific principles The thought is that the industry has the higher level of understanding of their manu-facturing processes and therefore more expertise at qualifying and validating those processes Performing equipment qualification and process validation without sound scientific principles and risk-based objectives have resulted in ineffective process control and inefficient validation programs.

For years, the industry had been asking for more flexibility in designing the exper-The 2011 final version of the FDA 2011 “Guidance for Industry, Process Validation: General Principles and Practices,” referred to in this chapter as the Process Validation Guidance (PVG) was issued in final form on January 24, 2011 This guidance appears to address these concerns The guidance presents a relatively non-prescriptive, science- and risk-based, life cycle approach to process validation

It is important for companies to understand the principles presented in the guidance and the intent of the agency authors Trying to follow the guidance without this understanding will likely result in recurrence of the same issues that the industry has experienced in the past

This chapter does not describe how to validate processes Other publications and chapters address that subject Nor does the chapter present a plan for complying with the PVG Doing so would misrepresent the intent of the PVG Rather, the chapter provides the reader with some insight into the intent and content of the 2011 final version PVG To do so, the chapter will present background and commentary on the document, based on responses to the 2008 draft PVG, Parenteral Drug Association (PDA)-sponsored workshops, and related discussions Understanding the back-ground and intent of the PVG should make it easier to develop plans that are aligned with the recommendations presented in the document, and that should result in more effective process validation approaches

The first part of the chapter will include background into the subject of process validation and guidance The second part will present the response to the 2008 draft PVG The third part will present a more detailed analysis and commentary on each

of the sections of the 2011 final PVG

2.1  PART 1: INDUSTRY BACKGROUND

stand the events and industry conditions leading to the release of the 2011 PVG The practical concept of the drug manufacturing process validation began in the 1970s The need for process validation may have been a response to the 1970–1971 out-

To understand the perspective presented in this chapter, it may be helpful to under-breaks of E cloacae and Erwinia contamination in large-volume parenteral bottles

at several hospitals.1

The outbreaks were apparently caused by moisture seeping into a space under the screw cap of the bottles during cooling after sterilization The contamination was released when the bottles were opened Routine sterility testing of the solution would not necessarily have uncovered the contamination, as the contamination would not have had time to spread to the point of being apparent The testing of product quality was not enough to assure product quality or patient safety A systematic approach to

confirm the effectiveness of all process steps and conditions that affect product qual-Companies or individuals wishing to distribute drug products that affect the health and well-being of people have a business and moral obligation to assure that those products are safe and effective In the United States and in many foreign mar-kets, that obligation includes a regulatory requirement to do so as well In the United States that requirement is presented in section 501(a)(2)(B) of the U.S Food, Drug, and Cosmetic Act (21 U.S.C 351(a)(2)(B)), which states that “A drug shall be deemed to be adulterated if the methods used in, or the facilities or controls used for, its manufacture, processing, packing, or holding do not conform to or are not operated or administered in conformity with current good manufacturing prac-tice to assure that such drug meets the requirements of this Act as to safety and has the identity and strength, and meets the quality and purity characteristics, which it

purports or is represented to possess.”2

facturing practice (CGMP) for finished pharmaceuticals as provided in 21 CFR part 211.100 of the Current Good Manufacturing Practice regulations state that “There

Food and Drug Administration (FDA) regulations describing current good manu-shall be written procedures for product and process control designed to assure that

sented to possess.” Other CGMP regulations define the various aspects of validation For example, § 211.110(a), Sampling and testing of in-process materials and drug products, requires that control procedures “be established to monitor the output and

drug products have identity, strength, quality, and purity they purport or are repre-sible for causing variability in the characteristics of in-process material and the drug product” (emphasis added).3,4

to validate the performance of those manufacturing processes that may be respon-The requirement for assurance of process control can be met by one of two means

1 Observation or Monitoring Where the outcome of a process, a condition,

or a product quality attribute can be fully observed, inspected, or tested For this to be effective, the attribute or outcome must be apparent for all product units, e.g., dimensions on a medical device or printing on a label For observation to be effective, all units of product must be sampled and inspected and the inspection must be fully reliable

ity attribute cannot be fully observed, inspected, or tested, assurance can

2 Prediction Where the outcome of a process, a condition, or a product qual-be provided by predicting the outcome based on information that can be observed or calculated, e.g., environmental monitoring, statistical sam-pling, personnel training, confirmation of equipment design, or represen-tative temperature mapping For prediction to be effective there must be correlation between what is observed and the predicted outcome

Validating a process is providing assurance that the process is scientifically sound and capable of resulting in the desired outcome on a consistent and reliable basis It is the combination of observation and prediction It is accomplished by predicting the outcome of the process by making a correlation between the performance that can be observed and the desired product quality outcome

rility assurance and sterilization The validation studies included moist and dry heat temperature mapping and biological challenges of autoclaves, dry heat ovens and tunnels, and media fills This emphasis on contamination control and sterilization made sense on a relative risk to patient safety basis.

Validation efforts in the 1970s focused on contamination control, specifically ste-In the 1980s, validation programs included equipment and facility qualification, cleaning validation, mixing studies, sanitization/disinfection effectiveness valida-tion, analytical test method validation, computer system valuation, environmental monitoring qualification, hold studies, clean-in-place and sterilize-in-place studies, container formation, ethylene oxide, gamma radiation, and electron-beam steriliza-tion validation

In the 1990s, more detailed design qualifications, installation and operational qualification studies, validation planning, packaging qualification, and container integrity studies were performed to a greater degree Over the past 30 plus years, the focus has included the validation of manufacturing processes for all dosage forms and manufacturing technologies

Throughout this period, it appears that much of process validation was based on evaluating samples of product taken from a series of consecutive commercial-scale or near-commercial-scale “demonstration” batches of product If these batches did not meet all of the prescribed quality specifications, then an investigation and evaluation was performed The evaluation of results became a combination of deviation investi-gation and justification, rather than a presentation of process control supporting evi-dence In other words, more time was spent explaining or defending what was wrong, than presenting why it provided assurance of process control and product quality It was apparent that for processes to work properly, the equipment and systems that supported those processes needed to be suitable for use and function in a reliable manner Equipment, facility, and system qualification provided that assurance

In 1987 the FDA issued its Process Validation Guidance The 1987 PVG is a presentation of industry best practices, as well as regulatory expectations It codified practices and regulatory expectations, including equipment qualification and process validation Europeans soon follow with different emphasis, but basically the same format for validation

Qualification and validation became an important topic and business endeavor If

mentation The company would spend tens of millions of dollars or more designing and constructing a facility (not to mention the drug product development costs) The facility could not manufacture commercial product until validation was complete If that took six months or a year or longer, then the investment would show no return

a company wanted to market a drug product, it needed to provide validation docu-in that time There was a validation department, which planned and completed the validation The validation process was laborious and time consuming Companies hired additional staff and contractors/consultants to expedite the process With the potential operating margins on products high, the cost was of less consideration than schedule Products represented billion-dollar revenue streams Hundreds of millions of dollars of investment were at stake, not to mention market share and time to market The validation effort would cost a fraction of the benefit it would bring Therefore, doing anything and everything to assure regulatory compliance

and product approval made business sense This led to a lack of focus on the objec-Validation consultants would transfer procedures from company to company

“influencing” each with the other’s program, regardless of appropriateness to their process Trade associations published best practices and held conferences to pres-ent case studies Everyone wanted to know what the others were doing If another company did something and got approval from the FDA, then it must be right Cost was not an issue One dared not put a price on quality More to the point, it was not prudent to take a risk of product approval delay by attempting a novel or unique approach to validation It was more prudent to be conservative This resulted in an additive or ratcheting effect, where each new project added another level of valida-tion test or procedure with little consideration of appropriateness or effectiveness for the company’s specific process

cussion of the sources of that variability or of the impact that variability might have

There was only limited consideration of process variability There was little dis-on process performance and product quality There was limited interaction with the process development people to determine process variability or weakness There was little feedback after the completion of the process validation studies on pro-cess performance once the process was released for commercial manufacturing use Further, since the equipment and systems were qualified early, often before criti-cal process parameters were identified, the parameters tested might not be the ones needed to verify that the equipment was capable of reliable process performance.The regulators were looking to the industry for best practices This meant that they observed what companies were currently doing and reported on it Therefore, internal inspection and regulatory submissions guidance reflected and reinforced those approaches However, some of these practices were not optimal at best and not effective at worst This resulted in increasingly high validation costs, project delays, and ineffective process validation Companies needed significant quality resources

to police validation efforts, identify deviations and errors, and investigate failures Quality units often found themselves in conflict with engineering and operations Operations received only limited benefit and knowledge from the validation effort There was an ineffective transfer of useful information from validation to mainte-nance or operations

In the 1990s several companies were found to have deficiencies with various quality-related systems, resulting in warning letters and consent decrees Over 25 consent decrees were entered into from 1990 to 2004.5 In many cases the response

grams to appease regulators In some cases these programs proved to be difficult to sustain in operations and added to the ratcheting effect In the 2000s many compa-nies faced financial pressures There was increased scrutiny on cost of drug manu-facturing placed on the industry The result was closer scrutiny of validation-related costs As the decade closed, the FDA appeared to increase its enforcement policies Regulators encouraged companies to utilize risk management techniques to make product quality-related decisions and prioritize efforts

to the consent decree was to create very robust validation and process control pro-The result was two decades of less than optimal validation ratcheting, increased validation and manufacturing cost, and decreased effectiveness The increased emphasis on cost and additional regulatory compliance pressures meant that there needed to be change in the approach to process validation in the industry The factors that led to this need for change can be summarized in the following issues, which the industry faced from 1990–2010:

tices that were not aligned with good science One-size-fits-all validation approaches did not always address newer technologies Therefore, the use of these practices hindered innovation Some in the industry suggested that regulators should set the ground rules that companies provide assurance for safe and effective products, but leave the method for providing that assurance to the company manufacturing pro-cess experts

This resulted in criticism from the industry that regulators were promoting prac-2.2  PART 2: ANALYSIS OF THE 2008 DRAFT PVG

In 2006, the industry began to hear that the FDA was going to revise its 1987

“Guidance on the Principles of Process Validation.” Throughout 2007 and 2008, the industry waited for the release of a PVG draft In anticipation of the release of the draft PVG, the Parenteral Drug Association (PDA) organized a task force of industry experts to be ready when the draft guidance was released Trade associations and companies requested information from the FDA on the content of the revised PVG

On November 17, 2008, the FDA released the draft revision of the PVG This was meant to be a shift from the documentation focus of the 1987 PVG to a less prescrip-tive, more risk- and science-based approach The draft revision was released the week before the winter holiday season, initially allowing for 60 days to submit com-ments (an extension was later granted) Excessive comments were not anticipated, because the revision was a reflection of what was believed to be current industry practice and the document was nonprescriptive and general in nature Therefore, there should have been little on which to comment

as to the intent of the PVG and therefore there was a need for further clarification The comments received by the PDA were reviewed and organized into the six major categories discussed in the following sections.6,8

The use and definition of specific terms and language had the most questions and comments Collectively, these comments promoted the value of including a document glossary, as well as the desire to use terminology that is consistent with International Conference on Harmonization (ICH) and other FDA regulatory guidance defini-tions in order to reduce potential misinterpretation This was important, because the PVG would be used internationally and the interpretation of terminology should be consistent

These comments noted the need for clarification of expectations for determining the level of assurance required to initiate commercial product manufacture and release batches for commercial distribution Related to this issue was the concern that a limited number of developmental batches would not be sufficient to develop a statis-tically sound rationale for commercial product distribution

The guidance indicated that extensive testing on early commercial batches to achieve statistically sound process controls might be required, yet offered no indica-tion of expectation for what constitutes the acceptable level of assurance in order to reduce this level of testing Further, the guidance was interpreted as not adequately promoting risk assessment as a means to reduce the number of samples and level of monitoring on relatively low-risk processes and steps

Objections were raised relative to the expectation that viral and impurity clearance studies performed at small scale should be performed under full CGMP conditions The objections noted the additional burden this expectation would place on companies

rent release could be used Several concerns were raised on the recommendation for stability testing of all concurrently released batches being overly prescriptive and that the release of batches was not in the scope of the PVG

The comments indicated that there was some confusion over when and how concur-2.2.5  s cope and  l egacy  s ysTems  (p rocesses )

Scope:

Questions were raised on whether the guidance covered clinical product sup-plies, investigational medical products, blood products, in vitro diagnostic products,

and vaccine products These also included questions related to whether processes such as cleaning, sterilization, sanitization, holding, and distribution of commercial products were included in the scope of the guidance

Legacy processes: Clarification was sought regarding how the PVG should be

applied to the validation of existing processes or processes previously validated using the 1987 PVG

and  r egulaTory  i mpacT

Qualification: Significant concerns were expressed regarding the expectation to

pated production times, especially where extended processing times are encountered

demonstrate the capability of equipment to maintain operating ranges over antici-Documentation and organization: Clarification was sought on the difference

between qualification plans and protocols There were recommendations to remove language that prescribes organizational dynamics and personnel activities such as having a variety of disciplines and project plans, as well as trending production line operator’s errors

Regulatory impact: There were several comments requesting clarification of

expectations related to information and the format of information to be included

in a regulatory submission There were concerns raised over apparent or perceived conflicts between the PVG and some regulatory submission requirements

The PDA comments were submitted to the FDA in a letter and report on March 3,

2009.6 The FDA received these comments along with others from companies, other trade associations, individuals, and other interested parties Throughout 2009, the PDA conducted a series of workshops designed to present the draft PVG and obtain addi-tional comments from the industry The workshops included participation by one or more of the draft PVG authors and were held in San Francisco, Chicago, Las Vegas, Bethesda, Munich, and Puerto Rico.7 The workshop discussions indicated that questions remained on the intent of the PVG authors and raised concerns over how to comply with its recommendations As one consultant put it, the problem is that we just spent the past

30 years being told that process validation was a matter of good documentation practice Now you are saying it is based on science That is quite an adjustment in philosophy to

overcome 30 years of thinking and practice in a matter a months.

It may be interesting to note that in workshops the FDA has consistently described the 2008 and 2011 PVGs as nothing really new In fact, the FDA representatives at the workshops emphasized that the requirements specified in the 2008 draft PVG are all presented in the 1978 GMPs The FDA asserts that the principles and recommenda-tions in the 2008 draft PVG were just a reflection of what the industry was already doing More than one FDA representative expressed surprise when industry attendees responded that this may not be the case It may not be the case with all companies

The industry perception was that the regulators expected three conformance batches to demonstrate process control The revised 2008 draft PVG was silent on the number of batches required to validate the process The PVG spoke more to the need for data, which would come from batches, than number of batches as accep-tance criteria This perceived change in the three-batch expectation raised questions

As of April 2011 (and my guess is for some time after), questions related to how many batches should be run for process validation continue to be raised

In the 1987 PVG, the FDA noted that the document outlines the acceptable elements

of process validation for human and veterinary drug products and medical devices

In the 2008 draft PVG and 2011 final version PVG, the FDA adds biological and active pharmaceutical ingredients, but omits medical devices The 2008 and 2011

PVG versions change acceptable elements to appropriate elements and keep the

scope the same as the 2008 PVG The 2011 PVG notes that guidance on medical device process validation is provided in a separate document, “Quality Management Systems—Process Validation,” edition 2.8

ceutical ingredients (APIs), not just finished pharmaceutical dosage forms During the early 1990s, some firms in the industry debated the applicability of process validation principles and approaches to the manufacture of bulk drug products or bulk drug chemicals They argued that long-standing processes were art forms and not well understood, but were achieving reliable and predictable results Today, few would use such arguments as the basis for trying to convince regulators that process validation was not needed, so this is an evolution in attitude and understanding.The 1987 PVG has an “Introduction” section The section first presents a general regulatory basis for process validation, as a requirement in the CGMP regulations for finished pharmaceuticals (21 CFR Parts 210 and 211), and for medical devices (CFR Part 820) The section further notes that industry firms have requested guidance

The 2011 PVG scope is applicable to active drug substances and active pharma-on what the FDA expects them to do to assure compliance with this requirement for process validation This is an important statement because it (1) establishes that process validation is a requirement not just a recommendation and (2) it makes it a request of the industry to provide an approach, rather than the regulators unilaterally dictating an approach.9

The 1987 PVG goes on to state that the recommendations are not intended to be all inclusive, and that a variety of methods and techniques (are and) may be used to assure compliance with the requirement of process validation There is an acknowl-edgment that many methods are and may be employed This leads one to imagine that the agency reviewed many of these methods and is presenting what it feels is a compilation of current industry best practices, rather than a carefully devised treatise

on process control

The 2008 draft PVG mentions the 1987 version and goes on to point out that since that version, the FDA has obtained additional experience through regulatory

of process validation This is an interesting passage The authors do not say that the industry has gained experience or that new and more effective methods have been developed for process validation, or even that new technologies and challenges have emerged that warrant changes in the guidance The authors focus on the agency’s regulatory enforcement experience as the impetus for the new guidance This should not be surprising Over the years, the preferred method of process validation for many companies has been to run three consecutive full-scale commercial batches, monitoring process parameters and product output

At times these runs did result in some out-of-specification results or deviations

cate an invalid or failed process In some cases, processes continued to exhibit fail-ures during commercial production These resulted in modification to the process or incorporation of additional control measures, that is, enhanced product inspection

In these cases companies often spent time explaining why the deviation did not indi-If process validation provides a high degree of assurance that the process works reliably and predictably, then how does one explain process failure? These are often explained as anomalies or isolated failures of procedure or systems This suggests a lack of full process understanding Do we know all process variables, and if so, have

we taken effective steps to adequately control those variables? There were several reasons for these issues, including:

• Poor process understanding when the process validation protocol was developed

• Insufficient knowledge of critical process parameters (CPPs) when user requirement specifications (URS) and process control strategies were developed and process equipment was designed and qualified

• Undocumented or inadequately addressed changes to process as a result of additional information gained after URS and control strategy development

• Focus on speed to market resulting in back-loaded process development and validation approaches

• Poor communications among process design, validation, and operation functions

• ations and back to process design functions

Inefficient transfer of knowledge from process design to validation to oper-2.3  PART 3: 2011 FINAL VERSION OF THE PVG

This part of the chapter analyzes the 2011 final version PVG It does so by presenting the text of the PVG, with commentary indented below that section Most of the com-mentary is based on comments received during the 2008–2009 review period, along with discussions that occurred at PDA PVG workshops, and related discussions and meetings.4,7,10,11,12,13

I Introduction

This guidance outlines the general principles and approaches that FDA considers appropriate elements of process validation for the manufacture of human and animal drug and biological products, including active pharmaceutical ingredients (APIs or

drug substances), collectively referred to in this guidance as drugs or products This guidance incorporates principles and approaches that all manufacturers can use to validate manufacturing processes.

The word appropriate replaces the word acceptable in the 1987 PVG Appropriate

is a broader term that indicates the approach is sound and effective, rather than just compliant Active pharmaceutical ingredients and biologicals have been added to clarify that API manufacture does require validation Medical devices have been removed, but a reference does point the reader to an acceptable guidance source.This guidance aligns process validation activities with a product lifecycle concept and with existing FDA guidance, including the FDA/International Conference on Harmonization (ICH) guidance’s for industry, Q8(R2) Pharmaceutical Development, Q9 Quality Risk Management, and Q10 Pharmaceutical Quality System 2 Although this guidance does not repeat the concepts and principles explained in those guidances, FDA encourages the use of modern pharmaceutical development concepts, quality risk management, and quality systems at all stages of the manufacturing process lifecycle Harmonization to international guidance is mentioned in this section This is an addition to the 2008 PVG One of the industry comments relayed to FDA by the PDA in 2009 brought up concerns over the lack of alignment with ICH and other international guidance

Quality risk management is mentioned This is an addition to the 1987 PVG Risk management and risk assessment appear twice as frequently in the 2011 final PVG as

in the 2008 draft PVG The limited mention of risk management and assessment in the 2008 PVG was brought up by attendees at PDA workshops One of the answers given by FDA at those workshops was that the authors felt risk principles were so engrained in our current practice that it did not have to be mentioned repeatedly However, the 2011 PVG does reinforce the use of risk approaches to a higher degree than did the 2008 PVG

mercial manufacturing process, and maintenance of the process in a state of control during routine commercial production This guidance supports process improvement and innovation through sound science.

The lifecycle concept links product and process development, qualification of the com-The 2011 PVG provides a method for defining and developing a lifecycle approach

to process validation A lifecycle approach is not new However, truly effective life cycle approaches were not always well implemented The 2011 PVG recommends that companies develop an approach that links, relies on, and transfers information from process development to process qualification to commercial manufacturing.The 2011 PVG mentions process improvement and innovation as acceptable out-comes, even key objectives of process validation In the PDA workshops, partici-pants raised questions and concerns over the reaction of regulators to changes or improvements made after process qualification They expressed concern that regula-tors would see this as an admission of allowing an imperfect process to be released into commercial production and therefore they would be penalized FDA participants

of additional information becoming available due to commercial manufacture and

if the initial process was sound and effective, but perhaps not yet optimal However, companies should consider taking steps to better identify areas of improvement prior

to completion of process design stages

tory submission Interested persons can refer to the appropriate guidance or contact the appropriate Center in determining the type of information to include in a submission.Industry comments made through the PDA in 2009 and at PDA sponsored work-shops raised concerns that the 2008 PVG did not address regulatory submissions and would potentially be in conflict with regulatory guidance on the inclusion of process validation–related information in submission The FDA participants in the work-shops recommended that companies consult individual center rules for guidance on submissions, but noted that conflicts should be minimal A prudent approach would

This guidance does not specify what information should be included as part of a regula-be to consider a plan that presents the most stringent regulatory expectation.This guidance also does not specifically discuss the validation of automated process control systems (i.e., computer hardware and software interfaces), which are commonly integrated into modern drug manufacturing equipment This guidance is relevant, how- ever, to the validation of processes that include automated equipment in processing The 2011 PVG does not address computer software validation, but since most items of equipment and instruments used to manufacture and test covered products rely on some level of automation, assuring the functionality and reliability of those systems would be part of this guidance

II Background

ing the availability of a guidance entitled Guideline on General Principles of Process Validation (the 1987 guidance) 7 Since then, we have obtained additional experience through our regulatory oversight that allows us to update our recommendations to industry on this topic This revised guidance conveys FDA’s current thinking on pro- cess validation and is consistent with basic principles first introduced in the 1987 guid- ance The revised guidance also provides recommendations that reflect some of the goals of FDA’s initiative entitled “Pharmaceutical CGMPs for the 21st Century—A Risk-Based Approach,” particularly with regard to the use of technological advances

In the Federal Register of May 11, 1987 (52 FR 17638), FDA issued a notice announc- ment and quality system tools and concepts 8 This revised guidance replaces the 1987 guidance.

in pharmaceutical manufacturing, as well as implementation of modern risk manage-Three important points are made in this paragraph:

1 The FDA is revising the 1987 PVG in part because of knowledge they gained from their experience with industry approaches over the previous 20 years This includes good experience, as in identifying effective approaches, and bad experience, as in failures of process validation and ineffective process