©BioPhorum Operations Group Ltd OVERVIEW

BIOMANUFACTURING TECHNOLOGY ROADMAP

OVERVIEW

BPOG Technology Roadmap 1 ©BioPhorum Operations Group Ltd OVERVIEW

Acknowledgments

The following member company participants are acknowledged for their efforts and contributions in the production of this roadmap document. (Overview section authors highlighted by *.)

Abbvie Janssen PM Group Derek Sawyer Patrick Sheehy Andy Rayner Li Malmberg (formerly) Ranjit Thakur* Roche Natarajan Ramasubramanyan Stefan Merkle* Oliver Stauch* Timo Simmen Asahi Kasei Bioprocess Paul Bezy Kimo Sanderson Kaiser Optical Systems (formerly) Tina Larson* David Strachan AstraZeneca Sanofi Rick Lu* Lonza Beate Mueller-Tiemann Rajesh Beri* Thomas Sauer Bayer Edgar Sur M+W Group Sartorius Stedim Ingmar Dorn David Estapé Lars Boettcher Janet Wendorf Merck KGaA, Shire Joerg Heidrich Darmstadt, Germany (formerly) Alan Glazer Thomas Daszkowski Jonathan Souquet* (formerly) Bert Frohlich* Biogen David Beattie Chun Zhang Eliana Clark* (formerly) Joerg Ahlgrimm Merck & Co., Inc., Phil McDuff Kenilworth, NJ, USA. Takeda CRB David Pollard Palani Palaniappan Kim Nelson Wayne Froland Thermo Fisher Scientific (formerly) Beth Junker* Emerson Craig Smith Robert Lenich NNE UCB Morten Munk Fujifilm Diosynth Biotechnologies John O’Hara Stewart McNaull* Novasep BioPhorum Jean-Luc Beulay G-CON Operations Group Maik Jornitz Pall Steve Jones* Hélène Pora Paul Ilott* GE Healthcare Life Sciences Linda Wilson* Jonas Astrom Pfizer Bob Brooks Joris Smets* GSK Clare Simpson* Paul McCormac* Charles Heffernan* Jonathan Dakin Will Waterfield David Paolella Justin John (formerly) Louise Duffy Publication Team ImmunoGen, Inc. Les Pickford Sandra Poole Karen Kilbride Thomas Ryll* Steve Pitt

The steering committee would like to thank the following non-member participants for their contribution to the roadmap meetings:

Alexion AMBIC, John Hopkins University MIT Al Boyle Michael Betenbaugh Charles Cooney Chris Love Biopharm Services BioProNet Alan Calleja* Alan Dickson NIST Andrew Sinclair Michael Tarlov BPTC Eli Lilly Tom Ransohoff UCL Todd Winge Gary Gilleskie Dan Bracewell Novavax Centre for Process University of Delaware, Tim Hahn Innovation (CPI) NIIMBL Richard Alldread Kelvin Lee A*STAR (Agency for Science, Technology & Research) ETH Zurich Andre Choo Massimo Morbidelli May Win Naing

BPOG Technology Roadmap 2 ©BioPhorum Operations Group Ltd OVERVIEW

Contents

1 Introduction...... 5 1.1 Current state of the industry: the case for change...... 5 1.2 Objectives...... 5 1.3 Scope...... 5 1.4 Roadmap participants...... 6 1.5 The biopharmaceutical roadmap methodology...... 6 2 Market trends ...... 8 2.1 Market growth...... 8 2.2 New product classes...... 9 2.3 Uncertainty of product success and sales...... 11 2.4 Cost pressures ...... 12 3 Business drivers and metrics...... 13 3.1 Facility flexibility ...... 14 3.2 Speed...... 14 3.3 Quality...... 15 3.4 Cost...... 16 3.5 Metrics...... 17 4 Biomanufacturing scenarios...... 19 4.1 Drug substance scenarios...... 19 4.1.1 Scenario 1: Large-scale stainless steel fed batch ...... 22 4.1.2 Scenario 2: Intermediate-scale single-use perfusion ...... 23 4.1.3 Scenario 3: Intermediate-scale multiproduct single-use fed batch...... 24 4.1.4 Scenario 4: Small-scale <500L portable facility...... 24 4.1.5 Scenario 5: Small-scale <50L for personalized medicines...... 25 4.2 Drug product scenarios...... 26 4.2.1 Introduction...... 26 4.2.2 Low-volume drug product manufacturing scenario...... 26 4.2.3 High-volume drug product manufacturing scenario...... 27 4.2.4 Initial considerations to deliver 10-year targets...... 28 5 Regulatory strategy ...... 29 5.1 Introduction...... 29 5.2 Scope...... 30 5.3 Regulatory needs, challenges and potential solutions...... 31 5.3.1 Needs and challenges...... 31 5.3.2 Potential solutions...... 32 5.4 Regulatory interaction recommendations...... 32 6 Conclusions and recommendations...... 33

BPOG Technology Roadmap 3 ©BioPhorum Operations Group Ltd OVERVIEW

7 References...... 34 8 Acronyms/abbreviations...... 35 9 Glossary...... 37 10 Appendices...... 38 Appendix A – Product classes...... 38 Appendix B – Detailed modeling results ...... 39 Appendix C – Antitrust statement...... 47

List of figures

Figure 1: Roadmap development and structure...... 7

Figure 2: The four major biopharmaceutical market trends...... 8

Figure 3: Worldwide sales of monoclonal antibodies and their derivatives by year ...... 8

Figure 4: Total number of recombinant protein products on the market and in clinical development by phase ...... 9

Figure 5: Number of biopharmaceuticals approved by the FDA since 1952 by product class...... 10

Figure 6: Number of biopharmaceuticals in the development pipeline by product class...... 11

Figure 7: Relationship between market trends and industry business drivers...... 13

Figure 8: Product lifecycle activities and cost components...... 16

Figure 9: Comparison of business driver profiles for selected scenarios...... 21

Figure 10: Range of throughputs targeted by facility type...... 22

Figure 11: DP metrics for low-volume scenario...... 27

Figure 12: DP metrics for high-volume scenario...... 27

Figure 13: Scenario 1 process cost breakdown...... 44

List of tables

Table 1: Five and 10-year metrics for flexibility, speed, quality and cost...... 18

Table 2: Main drug substance facility-type scenarios with typically associated product types and business drivers...... 20

Table 3: Regulatory agency policy and communications...... 29

Table 4: Illustrative examples of regulatory stakeholders...... 30

Table 5: Illustrative summary of regulatory needs and challenges...... 31

Table 6: Assumptions used for process cost modeling...... 40

Table 7: Process flow used to conduct initial modeling exercise...... 41

Table 8: Scenarios used for initial round of modeling...... 41

Table 9: High-level modeling results (cost per gram of antibody and capital expenditure)...... 43

Table 10: High-level modeling results for each scenario...... 43

Table 11: Input parameter ranges for sensitivity analysis...... 45

BPOG Technology Roadmap 4 ©BioPhorum Operations Group Ltd OVERVIEW

1.0 Introduction Considering this assessment and other public commentary, 1.0 the biopharmaceutical industry has recognized the 1.1 Current state of the industry: the case advantages of collaborating on a common technology for change and capability roadmap to drive the needed transition in biomanufacturing. It is now appreciated that a collective The future for biopharmaceuticals remains bright as effort in the pre-competitive space will have greater demand for drugs and therapies continues to rise. The impact and mutual benefit than what companies can do on ability to treat an increasingly wider variety of diseases their own with limited resources. has never been greater. Many diseases that previously were untreatable are now within reach of modern 1.2 Objectives pharmaceutical science. However, the biopharmaceutical The primary objective of this technology roadmap is industry is facing a number of global challenges. While to establish a dynamic and collaborative technology demand is growing, the cost of new product development management process for the industry to accelerate continues to rise while the pressure to reduce drug change. A roadmap will focus the efforts of the prices and increase patient access has intensified. biomanufacturing community and provide direction to Although the bulk of new product development costs industry stakeholders by: are still associated with clinical studies, there is also pressure on biopharmaceutical process development • determining pre-competitive critical needs and drivers and manufacturing to increase speed, improve flexibility • identifying technology and/or manufacturing targets and reduce the overall cost to supply a drug or therapy • prioritizing potential solutions through – while simultaneously sustaining or raising quality. As a assessment and modeling, where appropriate. heavily regulated industry, the extensive time and effort Initially focusing on a 10-year horizon, the roadmap is required to obtain global product licensure and approval meant to be a starting point for future discussions and of significant post-approval changes from regulatory activities. While generated by the member companies of authorities, often with disparate regional requirements, the BioPhorum Operations Group (BPOG) Technology only adds to the challenge. Roadmap collaboration, publication of this first edition is Newer technologies and greater innovation are required meant to reach a far wider community. The roadmap will to meet these challenges. Compared to other industries, be made freely available to all in the industry with the aim it is generally recognized that the pharmaceutical sector of stimulating a response and aligning efforts. The intent is has been slow to innovate and implement new approaches also to recruit additional stakeholders and contributors for and technologies. The perceived regulatory barriers are implementation planning and the creation of subsequent a major factor in this aversion to risk, as are the lack of editions. Like any strategy document, the roadmap will harmonization and coordination within the industry. Most be refreshed and expanded as situations change and new biomanufacturers, if they do develop a new technology, knowledge comes to light. choose to do so in isolation resulting in highly customized 1.3 Scope solutions. With little industry-wide standards or guidance, The scope of this first edition of the roadmap is focused the risk of incurring program delays, or even failing to on commercial biomanufacturing of drug substance. obtain a license with the introduction of a new technology, The overview includes a high-level view of the scenarios is often considered too great. This environment also makes impacting on drug product, but this is a topic for more it difficult for suppliers to know where to innovate and detailed consideration in future roadmap editions. the risk/reward balance for launching new approaches to biomanufacturing is poor. The technology areas selected are based on their perceived readiness for implementation within the As a result, biomanufacturing production platforms and next decade. These readiness assessments were also manufacturing infrastructure has not changed much over influenced by discussions on pre-competitive deployment the last 30 years. Conventional facilities are extremely opportunities, team knowledge and expertise, and their capital intensive to construct and have long build and perceived urgency and short-term impact on the industry. start-up lead times. Initial scale selection must rely on Where disruptive technologies could be identified or forecasts that rarely match real demand and often results predicted to appear in this initial timeframe and potentially in high underutilization costs. Typically built in stainless alter the roadmap, they are also pointed out. steel for a fixed product type, these facilities are inflexible, yet market trends point towards a more diverse mix of products types and therapies.

BPOG Technology Roadmap 5 ©BioPhorum Operations Group Ltd OVERVIEW

Specific longer-term technologies requiring significant future vision. During its assembly, more than 160 development were deprioritized for this edition. For this individuals participated, including representatives from reason, the creation of new and improved cell lines or 18 biopharmaceutical manufacturers, 12 suppliers to the other protein expression systems are not included in the biopharmaceutical industry, eight academic/government scope, despite their profound impact on biomanufacturing institutions involved in biotechnology research and/ efficiencies to date and their future potential. In addition, or training activities, and four engineering consulting drug devices, anti-counterfeiting (such as serialization), firms. Steering Committee members are mentioned in packaging and labeling, and supply chain were omitted the Acknowledgements with their affiliations. Roadmap from this initial effort. concepts and interim outcomes were also reviewed with Since recombinant proteins currently represent, by far, the global regulators, and comments and suggestions were largest class of biological drugs for both therapeutics and solicited from a wider industry cross-section during vaccines, the ‘backbone’ of the roadmap will emphasize conference presentations and communications. opportunities for improvements in the manufacturing of 1.5 The biopharmaceutical these important products. More specifically, due to their roadmap methodology current and short-term future dominance, monoclonal antibodies (mAbs) produced by mammalian cell cultivation Development of the roadmap was based on the process 2 are the key focus of this first edition roadmap. pioneered at the University of Cambridge . Additionally, best practices from other industry roadmaps, such as the Despite this initial emphasis on the mAb product class, International Technology Roadmap for Semiconductors significant portions of the proposed technology strategies and the National Aeronautics and Space Administration apply to other classes, such as therapeutic proteins (non- Technology Roadmaps, were identified to create a method mAbs), antibody-drug conjugates and vaccines, as well for the biopharmaceutical roadmap effort. The BPOG as microbially derived products. Less focus was placed facilitated a series of workshops and team meetings that on cell therapy products due to the recent publication invited BPOG member companies to (1) identify market of a roadmap for that modality1. Gene therapies and trends followed by the development of quantified business combination products, while certainly on the horizon, drivers (the ‘why’), (2) identify the critical commercial are also considered to be further out and are currently value streams that later evolved into six technology a small portion of the biopharmaceutical market. In roadmap teams (the ‘what’), and (3) catalog the required future roadmap editions, additional technologies will be enabling technologies and enablers (the ‘how’). highlighted or identified as the other product classes begin to emerge in larger numbers. In the interim, and where possible, the roadmap will point out potential synergies To assess the sufficiency of the expected impact among the product classes and treatment modalities in of proposed technology strategies, and to guide terms of the development and manufacturing technologies prioritization, a set of manufacturing scenarios was that can be brought to bear. devised to represent the 10-year vision for the facilities Technologies that solely support the development of of the future. Teams then projected the likely evolution manufacturing processes and associated analytical of these five scenarios over 10 years to further guide the methods, such as high-throughput laboratory systems roadmaps of key enabling technologies and capabilities and laboratory automation and robotics, are out of scope – Process Technologies, In-line Monitoring and Real- of this first edition of the roadmap. While it is recognized time Release, Modular and Mobile, Automated Facility, that development costs are at least as influential as facility Knowledge Management and Supply Partnership design, construction and manufacturing on overall costs Management. These six teams engaged industry subject and the ultimate price of a drug, the focus of this roadmap matter experts to roadmap details of needs, challenges is on commercial manufacturing. and potential solutions. The roadmap teams linked the enabling technologies and capabilities to market trends 1.4 Roadmap participants and business drivers and highlighted overlaps and This roadmap is the first major global collaborative effort dependencies. The structure of the biopharmaceutical of the biopharmaceutical industry to establish a collective technology roadmap is illustrated in Figure 1.

1 National Cell Manufacturing Consortium, A Technology Roadmap to 2025, Feb 2016. http://cellmanufacturingusa.org/sites/default/files/NCMC_Roadmap_021816_high_res-2.pdf

2 http://www.ifm.eng.cam.ac.uk BPOG Technology Roadmap 6 ©BioPhorum Operations Group Ltd OVERVIEW

Figure 1: Roadmap development and structure

Market trends and business drivers

Cost pressure ncertainty Market growth New product classes Paer pressure Produt suess merin marets onms s Biosimiars emand ariaiit Goa reah Gene therap eeopment ompetition nreion manuature e therap

Cost Flexibility Speed uality manuaturin ost haneoer uid time roustness P emand response ead time ost o poor uait

iomanufacturing scenarios (Facility types) 5. Small-scale <50L for personalized medicine

4. Small-scale <500L portable facility

3. Intermediate-scale multi-product single-use fed batch rug product Lo oume 2. Intermediate- se perfusion scale single-u rug product ih oume 1. Large-scale stainless steel fed batch

ae istriuted

Enabling technologies and capabilities

Process technologies Automated facility Modular and mobile In-line monitoring nowledge Supply partnership and real-time release management management

OG aiit uid uid time Produt reease ost o proess ae innoatie supp proess speed P da deeopment hains

inestment OP osts Quality, efficiency ime to introdue a ost o uait rom urrent and supp hane to an eistin ime proess to month ost o nonuait to o operatin osts

· Process intensification ie hihuait · Quick to configure nhaned inine · Efficient technology Partnerships ith and omination o unit and roust and sae monitorin transer uait uit in operations iomanuaturin tandard desins ndiret and nterated noede tandard orin ontinuous proessin Pu and pa mutiariate sensors uait throuhout interation and treamined aidation tehnooies ouped iee reatime eetroni ith adaned proess Open data standards utiariate anasis data ehane and preditie modein ontro nteroperaiit hared pannin

ADCs – antibody drug conjugates, mAb – monoclonal antibodies, COGS – cost of goods, CAPEX – capital expenditure, OPEX – operational expenditure

BPOG Technology Roadmap 7 ©BioPhorum Operations Group Ltd OVERVIEW

2.0 Market trends The considerations and contributing factors for each of 2.0 these major trends are summarized below: How the market is changing 2.1 Market growth Early in the roadmapping effort, senior executives and The number of biological drugs has risen steadily since subject matter experts from 18 biopharmaceutical the advent of recombinant DNA technology in the manufacturers exchanged perspectives and identified 1980s. More recently, this trajectory has accelerated four important market trends that currently underlie with the explosive growth of the biopharmaceutical the industry (see Figure 2). These four major trends market in the past five years. In 2011, biopharmaceuticals were subsequently used to form the basis of technology accounted for only 5% of the total biopharmaceutical and selection for this roadmap. These major trends are: (1) pharmaceutical market revenue. By 2020, it is estimated the continued growth of the market for that biopharmaceutical sales will contribute at least 50% biopharmaceuticals, (2) the introduction of new product of total revenues. This trend is evident from the increasing classes, (3) the pressure to reduce costs, and (4) the percentage of biopharmaceutical product candidates in uncertainty in the clinical success, approvals and sales of the development pipeline, which is predicted to comprise any one new product. 80% of the pipeline, resulting in a concomitant decline of small molecule product candidates from their current Figure 2: The four major biopharmaceutical market trends 3 iure he our maor iopharmaeutia maret trends majority of 57% . Recombinant proteins alone will continue to occupy New product classes Market growth e treatment modaities trenth o saes the largest portion of demand. The largest class of Diversification of product groups oust deeopment pipeines recombinant proteins is mAbs, whose structural and danes in sstems ioo Goa reah and emerin marets Personaied mediine functional diversification are further fueling their growth. Market Figure 3 represents the global therapeutic mAb market trends and indicates the contribution by sub-class: antibody Cost pressure ncertainty of fragments, conjugates or fusion proteins. It has been said Paer pressure on prie approvalsiure and ordide sales saes o monoonathat “At antiodies this growth and their rate, deriaties sales of currentlyear approved mAbs Biosimiars and ompetition inia eficacy; Dose requirements ost o inia aiures ompe oa reuations plus sales of from new products approved in the coming saatin deeopment osts emand ompetition and maret share Regional/political requirements years will drive the worldwide sales of mAb products to over $94 billion by 2017 and nearly $125 billion by 2020.” 4

Figure 3: Worldwide sales of monoclonal antibodies and their derivatives by year 5

aes Biions

eominant proteins produed uenth monoona antiodies onoona antiod onuates in mammaian e uture produed in mammaian e uture raments and usion proteins produed in mammaian e uture eominant proteins produed onoona antiod produts produed iopharmaeutia produts in miroia ermentation in miroia ermentation produed in pant e uture

3 Datamonitor: Pharmaceutical Key Trends 2011 – Pharmaceutical Industry Infrastructure Overview 4 Ecker, Dawn M, Susan Dana Jones, and Howard L Levine. “The Therapeutic Monoclonal Antibody Market.” mAbs 7.1 (2015): 9–14. PMC. Web. 21 Mar. 2017. BPOG Technology Roadmap 8 ©BioPhorum Operations Group Ltd OVERVIEW

This anticipated continued growth trajectory is timelines typically required to conduct pivotal clinical confirmed by the number of new biological drugs in the trials, file for approval and to gain market authorization to various phases of clinical development (see Figure 4). launch product candidates that provide unique and urgent Approximately three times as many new products are patient advantages. This acceleration in the number of in the development pipeline compared to the number of drug candidates moving through Phase I, II and III clinical current products and many of these will come to market trials will contribute to the rapid market growth. It is within the next decade. Additionally, regulators such as the apparent that the ‘boom’ of recombinant protein products European Medicines Agency (EMA) and the Food and Drug is likely to continue for a while, even though many other Administration (FDA) in the US have recently developed exciting biologic therapies are also in development. The iure ota numer o reominant protein produts on the maret and in inia deeopment phase breakthrough, accelerated and other fast-track licensure emphasis on this class of products in this first edition of the routes. These resulted in substantial reductions in the roadmap appears well justified.

Figure 4: Total number of recombinant protein products on the market and in clinical development by phase 6

umer o produts

aret BL Phase Phase Phase

eominant proteins tpe andor stae o deeopment

BLA – biologics license application, MAA – marketing authorization application, NDA – new drug application

2.2 New product classes However, alongside the growth of this product class and the more traditional biologically derived products, such In addition to the growth of currently available as blood factors and vaccines, other product classes are biopharmaceuticals, the number of different types of emerging, such as cell and tissue therapies. These other products continues to expand. Figure 5 shows the number therapies, also derived from cell culture, will continue to of products approved by the FDA by type and by year since grow and generate an even greater variety of product the early 1950s. Recombinant proteins have expanded classes as new treatment modalities demonstrate most dramatically, starting from the 1980s after the effectiveness. introduction of recombinant DNA technology.

5 BPTC. Bioprocess Technology Consultants, Inc. bioTRAK® database; 2016. http://www.bptc.com 6 Ibid. BPOG Technology Roadmap 9 ©BioPhorum Operations Group Ltd OVERVIEW

iure umer o iopharmaeutias approed the sine produt ass

Figure 5: Number of biopharmaceuticals approved by the FDA since 1952 by product class 7

New product classes New product Market growth classes

MarketMarket The number of cell culture-based products is expanding trendsTrends Cost pressure ncertainty of approvals Multiple applications

aines ira prodution

eominant proteins monoona antiodies reeptor anaos enme repaement therapies et issue uture tissue rats

utooous and other e therapies

tem es and reeneratie mediine

nth

issue e tem eom transen eom

umuatie numer o approed iopharmaeutias irus iro Bioo

Currently, biopharmaceutical products loosely fall into healthcare industry. Figure 6 shows a predicted growth the following classes and subclasses. Definitions are trajectory for these newer products over the next few included in Section 9: Glossary and a complete description decades based on the types of products currently in is provided in Appendix A, along with an overview of the clinical development. These newer therapies will begin to production methods typically employed. shift treatment paradigms as they replace or supplement • traditional biologics more traditional therapies. For example, there are more than 1,000 clinical trials in progress worldwide based • mAbs and derivatives on stem cell technology8. In a survey of 250 global • other therapeutic proteins (non-mAb) biopharmaceutical executives, 48% indicated that their • antigens and viruses company is developing or will develop a novel therapy, • gene and mRNA/oligonucleotide therapies such as gene and cell therapies9. Thus this roadmap • cell and tissue therapies. will need to account for this increasing diversification New therapeutic technologies and treatment modalities and find flexible solutions to accommodate the added will be transformative and even disruptive for the manufacturing complexity.

7 Sourced from Bert Frohlich, Shire, UML Biopharmaceutical Summit, March 2012 8  clinicaltrials.gov 9 Market Insight: The changing biopharma risk equation, DDNews, August 2016, (http://www.ddn-news.com/index.php?newsarticle=10840) BPOG Technology Roadmap 10 ©BioPhorum Operations Group Ltd OVERVIEW

iure he numer o iopharmaeutia in the deeopment pipeine produt ass

Figure 6: Number of biopharmaceuticals in the development pipeline by product class10

Perent o produts

aret BL Phase Phase Phase

ntiod Bood protein toine nme usion protein ormone Peptide Protein nsuin

2.3 Uncertainty of product success and sales With the global expansion of the biopharmaceutical marketplace, product candidates will need approval The biopharmaceutical industry experiences a high degree in many different countries and therefore different of uncertainty surrounding the successful development regulatory jurisdictions. Since many regulatory agencies of a product that will eventually launch globally and do not have closely aligned approval requirements, ultimately treat patients and generate revenue. This achieving global approvals can be technically challenging, uncertainty is likely to increase given the complexity of time-consuming and sometimes uncertain. According to mechanisms of action and the rise of novel and complex a recent survey, regulatory uncertainty tops the list of biological drugs and treatment modalities. Clinical trials risks that biopharmaceutical industry experts think might remain expensive endeavors with variable timelines disrupt their company’s strategy in the next five years11 . and the results are subject to regulatory scrutiny. Their outcomes are complicated by heterogeneous patient The varying requirements for biomanufacturers come populations and the ability to appropriately select and not only from health authorities, but also from other recruit patients. Even when a product candidate drug government agencies charged with economic development shows proof of concept in humans in early trials, there and taxation. In developed and developing countries is still the uncertainty of finding the right dose that alike, there is political pressure to maintain a domestic demonstrates efficacy with acceptable patient risk, while manufacturing base with its associated jobs. Therefore, forecasting market demand and potential revenue. there is considerable pressure to create in-region manufacturing rather than building centralized production facilities that are potentially more efficient.

10 BPTC. Bioprocess Technology Consultants, Inc. bioTRAK® database; 2016. http://www.bptc.com 11 Ibid. BPOG Technology Roadmap 11 ©BioPhorum Operations Group Ltd OVERVIEW

2.4 Cost pressures very different. Spending distributions are substantially different and there is a greater emphasis on the prevention A number of recent trends in the pharmaceutical industry of infectious diseases and in treatments that are low are impacting on both sides of the financial ledger. Gross cost. Also, the barrier to entry for competing drugs and margins are increasingly challenged due to price controls, therapies can vary considerably depending on the country competition and other uncertainties while operating or region. Thus, innovator companies and generic drug costs continues to rise. Also, the paradigm of how health makers alike will feel pressure on their revenues, which are services are delivered to the public is beginning to shift, used to fund new drug commercialization programs and requiring providers to demonstrate product value and maintaining the supply of the drugs and therapies in their fully understand the patient experience. There is increased current portfolios. attention on pharmacoeconomics, specifically for new treatments to demonstrate a lower overall social cost On the other side of the ledger is the cost involved in through cost-effective analysis (e.g. National Institute for developing, commercializing and manufacturing drugs. Health and Care Excellence type model (NICE)) and health The cost of drug development, in particular, has increased technology assessments. substantially. Over the decade of 2002–2011, the average cost of bringing a drug to market went from $2.4bn to Seeking efficiencies, health insurance companies and approximately $4.4bn. A substantial component of this national formularies are leveraging their purchasing power increased development cost is due to the additional time through tendering per molecule, price unification and it takes to successfully discover and develop a drug, which referencing. For example, payer groups are purchasing has increased rather than decreased over the past decade. set volumes of a specific biopharmaceutical for all of Over the past 15 years, development timelines to bring their patients with a specific ailment. Individual countries a new drug to market have increased from 10 years to are also negotiating unified pricing by region, leveraging approximately 14 years12. In addition to the high cost their higher aggregated market share. In addition, there of human clinical trials, averaged into these overall cost is the matter of public perception, which will drive figures are those drugs that fail in the clinic, are never political policy. New business models and paradigms may approved or that fail in the market due to reimbursement emerge that are closer to the patient and payer, such issues or competition. Thus, clinical success is a major as collaborative ecosystems of multiple stakeholders, factor in the financial well-being of an innovator company crowdsourcing and open source platforms. since the cost of failed developments will have to be Another consideration is the emerging markets where carried by the drugs that eventually do make it to and the healthcare paradigm and public health priorities are succeed in the market.

12 KMR, Bernstein Analysis, Bernstein-Global-Pharma_22Sept2015 https://kmrgroup.com/wp-content/ uploads/2015/09/Bernstein-Global-Pharma_22Sept2015_RnD-Productivity-Trends.pdf BPOG Technology Roadmap 12 ©BioPhorum Operations Group Ltd OVERVIEW

3.0 Business drivers and metrics will also help determine the relative value of certain 3.0 technology options and solutions. Understanding the connection between market demands, Figure 7 illustrates how the major market trends map which are largely outside of a company’s direct control, to the predominant business drivers. Each of the major and its business reality will be key to its success by drivers (i.e. facility flexibility, speed, quality and cost providing direction in decision-making and the ability to reduction) is further explained below. It is worth noting respond to changes. Specifically, in regard to bioprocess that there is not a one-to-one correspondence between development and biomanufacturing, a quantitative market factors and specific business drivers or metrics, but understanding of these drivers and their related metrics a more complex relationship.

iure eationship eteen maret trends and industr usiness driers Figure 7: Relationship between market trends and industry business drivers

Market Trends Business drivers

Market growth Facility flexibility ih demand aiit desin and sae umer o drus suppied utiprodut apaiit Goa reah and emerin marets eiona manuature

New product classes e treatment modaities Speed Personaied mediine peed to ini peed to uid peed to maret peed to supp

Uncertainty Quality Clinical efficacy; Dose requirements Produt attriutes and harateriation Product approvals; Complex regulations Comparability requirements Demand, competition and market share uaitris manaement Regional/political requirements Cost of non-quality

Cost pressure Cost reduction Paer pressure on prie Development costs Biosimiars and ompetition Facility investments; Timing ost o inia aiure onstrution and aidation osts saatin deeopment osts anuaturin osts

BPOG Technology Roadmap 13 ©BioPhorum Operations Group Ltd OVERVIEW

3.1 Facility flexibility 3.2 Speed The need for manufacturing flexibility is driven by With the rapid expansion of the biopharmaceutical market, several key trends: market growth along with increasing the industry will need to respond to the growing demand diversification of product modalities (which includes the by adding capacity quickly. However, the need for speed emergence of personalized medicine); fast-to-market is also driven by uncertainty and the high costs associated approaches for new drugs; the uncertainty of demand with biopharmaceutical development and manufacturing. forecasts and obtaining regulatory approval in a rapidly Similar to the drive towards flexible facility designs, changing healthcare environment. faster construction could allow capacity to be expanded in smaller increments. If large investment decisions can Flexibility in throughput be delayed because capacity can be added at a later point As new products are developed, new capacity will need to in the lifecycle of a drug, the financial risk and the level of be built. However, deciding on the scale of a new facility uncertainty will be reduced. Similarly, if a company can or how much existing capacity to reserve is a difficult reach a clinical development milestone more quickly, less decision. Whether this capacity is reserved in-house or research and development funds are put at risk. at a contract manufacturing organization, the decision carries with it financial consequences. Unfortunately, Over the lifecycle of a drug, there are three major market forecasts are almost always wrong. The eventual components to speed that are related to the major phases sales of a new biopharmaceutical can, ultimately, only be of development and commercialization. There are: determined by the products’ market success. Network • speed to clinic (first in human) or proof of concept capacity strategy is heavily influenced by a company’s • speed to market launch, which may include the speed pipeline. Overbuilding will tie-up precious capital and the to build if new capacity is required product may be burdened with higher depreciation costs. • speed to supply, i.e. end to end (E2E) manufacturing In particular, for a smaller company with more limited cycle time. resources, overbuilding can represent a very high business risk due to cash flow. Underbuilding can result Speed to clinic in having insufficient capacity to capture market share, Demonstrating absence of toxicity of a new product is which is critical to recovering the investment in drug a critical milestone in the development of a new drug. development and commercialization and in funding future Showing proof of concept in humans is a key indicator development efforts. of the ultimate success of a clinical development program. Achieving these outcomes quickly will greatly Thus, facility designs are needed that can flex with reduce the financial risk of continued development and changing demand, market conditions and throughput commercialization activities. Also, clinical-scale production requirements for new products. This scale dilemma is often on the critical path for initial toxicology studies in becomes more acute when market-specific regulatory animal and sometimes human trials. Thus, speed to clinic requirements limit what product can be sold in each includes the manufacture of finished drug product to market. In-region manufacturing further drives the need supply both animal pre-clinical and human clinical trials. for flexibility since manufacturing is distributed across more manufacturing sites. Further pressure for speed to achieve manufacturing readiness arises from the creation of faster regulatory Flexibility in process and production platforms pathways for approvals of biopharmaceuticals for unmet Multiple production platforms are expected as new medical needs. These accelerated/breakthrough pathways product modalities emerge. Building fixed infrastructure now exist in several developed and even emerging for different product types is likely to be prohibitive. markets, and more countries are expected to adopt them in Thus, production lines that can be reconfigured with the near future. Timelines will be shortened for approvals, relatively little effort, time and cost to accommodate requiring even faster development. And since accelerated/ new manufacturing methods and product types will breakthrough designation can be made based on early prove to be highly valuable as product diversification clinical data (e.g. Phase I), speed to the toxicology studies continues in the industry. and the clinic can also be a distinct competitive advantage.

BPOG Technology Roadmap 14 ©BioPhorum Operations Group Ltd OVERVIEW

Speed to market 3.3 Quality Speed to market includes the time required to conduct Quality management and control have always been late-stage clinical trials and to build appropriate launch paramount as patient safety is at stake. However, good capacity and/or retrofit existing facilities, if necessary. This quality management also requires maintenance of a phase includes the time taken to commission and validate consistent and reliable supply of a drug to the patient. the new or modified facility and equipment, and the process Attention to product quality during development is a key itself. It also includes the time required to file a regulatory enabler since the ability to accurately measure critical application and wait for the health agency’s approval. quality attributes of the candidate molecule is a real Achievement of this milestone permits a company to advantage and reduces risk. Given the uncertainty in finally capture revenue associated with the new product the successful development and commercialization of and to begin to recoup its substantial investment. If a biopharmaceuticals, a highly defined and characterized company can launch a product sooner, it may be able to product can also ensure quicker approvals. The complexity capture a greater market share, increasing and accelerating of the global regulatory environment and competition return on investment while accelerating patient access to will only add to the need for measurable quality and to be new medicines. able to differentiate similar molecules in the marketplace The need for speed to market is further challenged with the and the approvals process. Also, with the considerable accelerated paths to approval by the regulatory agencies advances in analytical sciences and methods, greater described above and, in some cases, with streamlined regulatory scrutiny can be expected, especially concerning trial strategies. While acceleration offers benefits to both product comparability. the patient and the manufacturer, the opportunity places As regulatory agencies across the globe are working to additional pressure on the development organization and on harmonize their requirements and expectations, there is a its facility construction. new common theme emerging in guidance documents that Speed to manufacture surrounds the expected use of modern quality systems Speed to manufacture captures the E2E cycle time to to manage risk. This is consistent with the goal of many plan, procure materials, set up, manufacture release and regulators to shift the burden of quality management distribute a product. Strategies and technologies that onto the pharmaceutical sector and towards a more self- can streamline processes and analytical testing will allow regulated industry. Thus, it can be assumed that a more more batches to be produced and ensure a faster release proactive quality management culture will be encouraged to distribution. Shorter cycle times also enable a company and eventually required in the pursuit of this ‘Desired to more reliably meet market demand, respond to changes State’, as defined by the FDA. and reduce inventories. If less safety and pipeline stock of Having achieved licensure of their biopharmaceuticals and finished product are needed, the cost of inventory can be successfully launched into the global marketplace, many significantly reduced. companies are now tasked with maintaining a consistent supply to their patients. Manufacturing processes that are not sufficiently robust and/or are impacted on by the variability of raw materials can lead to lots that fail to meet their final specifications and therefore must be rejected and discarded. Such issues can easily lead to stock- out situations and even to patients not receiving their medicines, which in some cases can be life threatening. The number of drug shortages rose steadily over the first decade of this century and remains unacceptably high. Shortages negatively impact on a company’s reputation and the trust between the company and its customers and the regulators. These are costly both in terms of the direct expenses associated with mitigation as well as lost opportunities and possible market share. Furthermore, deviations (or non-conformances) in the manufacturing process, investigations and quarantined or rejected lots put a large strain on the manufacturing, technical and quality organizations. Events are often unpredictable and result in sudden shifts of resources,

BPOG Technology Roadmap 15 ©BioPhorum Operations Group Ltd OVERVIEW

which are highly disruptive to an organization. 3.4 Cost The concepts of ‘quality by design’ and ‘quality risk Crucially important in selecting and evaluating a management’ were once viewed as burdensome and technology is the understanding of how it will impact unnecessary but are gaining acceptance in many the overall value chain in a product’s lifecycle. Cash companies. expenditure increases significantly as a product candidate A paradigm shift in quality management is likely to be moves through the initial stages of its lifecycle, from required if the industry seeks to achieve its goals. If quality early investments in research into development, towards is to be improved in biopharmaceutical manufacturing, commercialization and the provision of manufacturing then increased product and process understanding will be infrastructure. Decisions made earlier in the lifecycle needed. Characterization data collected before licensure will impact on subsequent activities and costs. can be used to design an improved control strategy, thus The major phases of a pharmaceutical’s lifecycle are minimizing process upsets and failures. Process and assay depicted in Figure 8, along with the activities and cost variability will also need to be monitored and controlled components associated with each phase. As described through a lifecycle approach to process validation. This previously, the most significant cost to an innovator understanding needs to applied to the full E2E supply company lies in the clinical development of a new chain, including process raw materials through to finished drug or therapy. However, significant costs also lie in good shipment, and ultimately through to the patient developing the manufacturing process and analytical and payer. A comprehensive control strategy built on assays, providing manufacturing capacity and then this understanding increases comfort levels such that producing and distributing the product. There is also the inventory cushions are no longer needed to mitigate overall cost of ensuring quality that can be significant supply risks, leading to improved product availability and and may provide opportunities for improvement, as potentially to a smaller quality organization. described in the previous section. Thus, technologies that increase manufacturing success rates, reduce lead times and inventory costs, and ensure high and reproducible quality, will undoubtedly also have great value.

Figure 8 Product lifecycle activities and cost components

Figure 8: Product lifecycle activities and cost components

Research evelopment Implementation Manufacturing

Product research Product development Construction Cost of goods isoer Preinia anima ite seetion a materias nima modes inia trias ite preparation Laor inia researh ru deier and deie onstrution tiities aret researh ormuation ommissionin uait ontro oioo uait assurane Proo o onept ehnia support Other site oerhead arehousin hippin Process development Start up istriution ru sustane proess ehnoo transer ru produt proess · Qualification natia methods aidation hippin studies rainin taiit studies iin

BPOG Technology Roadmap 16 ©BioPhorum Operations Group Ltd OVERVIEW

Risk: another form of cost 3.5 Metrics Over the long term, business risk is intimately tied The four major business drivers – Flexibility, Speed, to cost. Major investment decisions often have to be Quality and Cost – are the means to establish a successful made under the high uncertainty characteristic of this biopharmaceutical enterprise. The relative importance industry. Generally, the level of risk is higher the larger of controlling these drivers will depend on the business the investment and the earlier in a given development model and circumstances surrounding a given company program this investment has to be committed, since there or organization. Each of the major drivers is comprised of is less knowledge or understanding of all of the factors various component drivers. It is helpful to quantify these influencing success. Similar to speed as a driver, the cost component drivers to measure the extent to which a given to develop a product candidate, including the resources technology or approach will influence the major drivers. consumed to propel it through the significant hurdles prior to and including successful regulatory approval, are at risk Table 1 summarizes the metrics chosen by the until the product is successfully launched. collaborating companies to represent the component drivers felt to be most relevant for this roadmap. Technologies or business approaches that allow a The metrics are grouped by the major drivers. more rapid response to changing conditions, or allow a Based on the market trends and business needs decision to be delayed until a greater level of certainty or described earlier, targets were set for each metric confidence can be obtained, will dramatically reduce risk for five- and 10-year horizons, including a summary of and therefore long-term costs. Given the large expense the current state for each metric. These metrics align associated with biopharmaceutical manufacturing with the demands of the identified industry trends facilities, faster construction of additional capacity and the business drivers requiring significant innovation enables a later investment and possibly a more staged to meet future patient needs. or incremental expansion, rather than building for an expected peak capacity. Maintaining idle capacity represents a large opportunity cost as well as a significant financial drain. A new strategy, approach or technology may impact on one or more of the various cost components described above. How many and to what extent will determine its value to the industry.

BPOG Technology Roadmap 17 ©BioPhorum Operations Group Ltd OVERVIEW

Table 1: Five and 10-year metrics for flexibility, speed, quality and cost

Driver Metric Current state Five-year target 10-year target

Facility utilization percentage <70% >85% >95%

Titer range in upstream that is directly accommodated by Fed batch: 0.1–2g/L Fed batch: 1–10g/L Fed batch: 2–40g/L downstream facility fit Perfusion: 0.05–1g/L/day Perfusion: 0.5–5g/L/day Perfusion 0.5–10g/L/day

Product changeover time for one production train 3 days <18 hours <8 hours Flexibility Time to reconfigure suite for new process >2 weeks <1 week <2 days

Number of platforms per suite (e.g. ability to change between CHO, E. coli, yeast and gene therapy within a 1 3 >5 suite)

Time to produce first GMP material for the clinic 18–24 months 12 months 8 months

Facility build speed 3 years 2 years 1 year

Speed to market 7–10 years 5 years 3 years

4–6 months 2 months 1 month Time to make product (E2E speed) 100% 50% reduction 75% reduction Speed Time to release product (E2E speed) 4–12 weeks 2 weeks 1 day

6–12 months 2 months 1 month US/EU US/EU US/EU Time to introduce a change into an existing process 18–24 months 18 months 6 months ROW ROW ROW

Cost of non-quality >10% of operating costs 10% of operating costs 2% of operating costs

Process variability (ppk) <1.2 >1.5 >1.8

Quality Assay quality (ppk) <1.2 >1.5 >1.8

Inventory quantity 50% reduction 90% reduction Inventory cover 3–6 months 2 months 2 weeks

$100/g (mAbs) $50/g (mAbs) $10/g (mAbs) Total cost to supply 100% 50% reduction 90% reduction

$500m+ DS facility $100m DS facility $50m DS facility Cost Cost of upfront investment in manufacturing (depends on capacity) $50m DP facility $25m DP facility $100m+ DP facility

Cost from Phase III process analytical and formulation 100% 25% reduction 75% reduction development to launch

DS – drug substance, GMP – good manufacturing practice, ROW – rest of the world, CHO – Chinese hamster ovarian, DP – drug product

These metrics are used by the roadmapping teams to estimate the relative impact of proposed technologies and solutions and will help prioritize activities going forward.

BPOG Technology Roadmap 18 ©BioPhorum Operations Group Ltd OVERVIEW

4.0 Biomanufacturing scenarios Table 2 summarizes the drug substance scenarios and 4.0 includes a list of characteristic products and facility 4.1 Drug substance scenarios features. More detail on each facility type/scenario Given the complexity of the biopharmaceutical industry follows in the subsections below to provide a vision from and the increasing diversity of products and companies, which enabling technologies were identified. Since the it is clear that there will be no ‘one size fits all’ solution relative value or impact of a technology will likely depend to biomanufacturing. Instead, we see a range of on the set of business circumstances and facility type, biomanufacturing scenarios playing out over the next 10 the scenarios are designed to help with visualization and years. Five high-level scenarios were selected for drug estimating the value of technologies and alternatives. A substance manufacturing and two for drug product to technology that is determined to have a high impact across cover the full spectrum of process and facility types. Each multiple scenarios would reinforce its importance to the facility type is associated with a representative set of industry and likely elevate its priority as a technology to be business conditions reflecting a typical mix of key drivers developed further. and business dimensions.

BPOG Technology Roadmap 19 ©BioPhorum Operations Group Ltd OVERVIEW

Table 2: Main drug substance facility-type scenarios with typically associated product types and business drivers

Case/Scenario Facility type and base case process description Typical products and product classes Typical business scenario and key drivers

Large-scale stainless steel facility • low-complexity proteins, such as mAbs Very-high-volume product(s) • fed-batch upstream process • mAb fusions • demand range = 100s–1,000s kg/yr • stainless steel bioreactors >10,000L • other recombinant proteins with low • relatively high certainty of approval and • high (relative) cell culture expression rate degrees of glycosylation market acceptance (approx. 3g/L) • no or little requirement for regional 1 • batch or continuous DSP manufacturing • high (relative) purification recoveries • desire to maximize use of existing assets or to build new (greenfield) • single or few products • low to moderate need for fast launch • low COGS with high utilization

Intermediate-scale facility with single-use • mAbs and/or High- to medium-volume product perfusion • more complex proteins (non-mAb), • demand range = 100 kg/yr • perfusion USP with 2,000L such as therapeutic enzymes and other (range 50–500/kg/yr) single-use bioreactors unstable proteins, such as growth factors • moderate certainty market acceptance • semi-continuous or continuous DSP and/or peak market demand 2 • relatively small product mix (one to several products) with limited need for campaign changeovers • possible requirement for regional manufacturing

Intermediate-scale multiproduct facility mAbs and/or higher complexity proteins: Medium- to low-volume product single-use fed batch • bispecifics • low volume product since targets a specific • batch USP • or multicomponent biologics patient population • batch or semi-continuous DSP • conjugates • demand range = 100 kg/yr • gene therapy for large patient (range 50–500 kg/yr) populations (>1,000 patients per year) • multiple products to be manufactured at same site due to changing product 3 mix and/or possibility of regional manufacturing required. On a per-product basis but several products possible • good for low-volume products that target a specific patient population • clinical-scale manufacturing and/or lower certainty of approval and market acceptance

Small-scale <500L portable facility • potentially highly complex biologic, such Low volume product(s) • single-use perfusion cell culture as vaccine or allogeneic cell therapy • low volume product 1 kg/yr • continuous DSP (range 0.5–5 kg/yr) • portable process desired for regional • relatively low certainty of approval or 4 manufacturing or design for rapid facility market acceptance change out • requirement for regional manufacturing highly likely • high need for rapid deployment and/or into regional markets

Small-scale <50L for personalized medicines • potentially highly complex biologic, such Very low volume or personalized product • isolated, patient-specific preparation of as autologous vaccine or gene or cell • since customized to patient. Base case therapeutic protein or biological therapy therapy products for rare diseases and 0.1 kg/yr (range 0.0001–0.5 kg/yr) • typically manual cell culture or cell low patient populations (<100 patients 5 preparation per year) • batch or continuous DSP • highly portable but limited set of unit operations for possible bedside deployment or centralized parallel processing

DSP – downstream process, USP – upstream process, COGS – cost of goods

BPOG Technology Roadmap 20 ©BioPhorum Operations Group Ltd OVERVIEW

Figure 9 shows the various business profiles for each of will support robust sales as long as significant market the selected scenarios. These business dimensions include share can be established. Possibly because treatment capacity requirements and the business drivers that were is long-term and the dosage is relatively high, there is a described in the previous sections: facility flexibility, speed desire to minimize the cost of production. There may also to clinic and to market, quality and cost of manufacturing. be significant competition with other players entering For example, Scenario 1 represents a high-volume product the therapeutic space and therefore being the low-cost where there is a high degree of confidence that the market alternative may be critical to maximizing market share. Figure 9: Comparison of business driver profiles for selected scenarios

Figure 9: Comparison of business driver profiles for selected scenarios

ih

s enario

enario

Medium enario

Relative contribution of business driver enario

Lo enario

Capacity aiit peed peed to Manufacturing requirements flexibility to ini build/market cost (per unit)

As depicted in Figure 9, there is a considerable range in These business driver profiles are relative to each other all these potential requirements as drivers. Production and represent the trade-offs that are typically expected outputs can vary from metric tons of a recombinant and progress from one type of profile to another in terms protein (as has been the case for some blockbuster of overall capacity versus cost, flexibility and speed of mAbs) to very small quantities of a drug or preparation construction. In general, a given facility type will be most intended for a single patient. A facility may be assessed appropriate for a certain range of throughput. However, to manufacture proteins, virus or polysaccharides for there are certainly other factors influencing the chosen vaccines, vectors for gene therapy, a specific phenotype design of a new facility. Figure 10 shows an approximate for cell therapy or any combination of these. Depending on range of production rates that would typically be targeted the market and competitive situation faced by a company, by a given scenario and facility type. speed to clinic or to market may be paramount. The cost of goods (COGS) may be critical in gaining or maintaining market share and/or in demonstrating superior value.

BPOG Technology Roadmap 21 ©BioPhorum Operations Group Ltd OVERVIEW

Figure 10: Range of throughputs targeted by facility type

Figure 10: Range of throughputs targeted by facility type

Scenario Product output (kg/yr) 0.1 1 10 100 1000

Scenario 1 Laresae stainess stee ed ath

Scenario 2 ntermediatesae sineuse perusion Multiproduct Scenario 3 ntermediatesae mutiprodut sineuse ed ath

Scenario 4 masae L portae aiit Multiproduct Scenario 5 masae L or personaied mediine

While the scenarios collectively span several orders of 4.1.1 magnitude in throughput, they also overlap. Particularly Scenario 1: Large-scale stainless steel fed batch in the designs aimed at multiproduct manufacturing, The facility type for this first scenario is a traditional large- scale-out options can extend the range of overall scale, stainless steel biomanufacturing facility that has production in a given year. been characteristic of the biopharmaceutical industry for Quantitative computational models provide valuable the past 30 years. This scenario is still highly relevant as insight in estimating the value of a technology or many companies currently own and operate such facilities process innovation that would otherwise be difficult and, understandably, wish to continue to leverage their 13 to predict. The BioSolve analysis tool (from Biopharm existing assets in the most efficient manner possible . Services Ltd) was used to model the biopharmaceutical The facilities were primarily built for the production manufacturing scenarios. of recombinant proteins and mAbs with large annual For the first edition of the roadmap, the collaboration volume requirements. Traditionally, large batch processes team agreed on a set of assumptions for the scenario have been run both upstream and downstream in these types defined above that are representative of the typical facilities. Furthermore, many were designed based on manufacturing processes in commercial operation today. lower cell culture titers than what is currently being Sensitivity parameters were defined for key operations achieved in development and thus the production within these scenarios to reflect how the parameters could bioreactor volumes are in the order of 10,000–20,000L vary over time with the process improvements that the in scale. technology roadmap envisages. The base assumptions The set of business conditions and drivers that would and the sensitivity parameters can be found in Appendix typically accompany a facility of this type would be B. Process cost models were built to identify high-cost representative of a company that is in the fortunate components and the features of both the process and position of having a high degree of certainty around the the typical facility design, and to perform sensitivity full utilization of a fixed asset of this type. A robust and/or analyses to prioritize technology needs in the roadmap. growing product portfolio with a high aggregate certainty The application of findings from the models to the process of capacity needs, and/or a product with a high certainty technology requirements can be seen in the Process of market penetration (such as a follow-on biologic), can Technologies report. justify the investment in a large-capacity facility and the Manufacturing process modeling is a topic for further risks associated with a relatively inflexible design. consideration in future editions of the roadmap. It offers Thus, these large-scale facilities may remain well benefits to all stakeholders in understanding the value suited for business circumstances requiring a very high and impact of innovation opportunities. For example, a capacity of protein per year (in the order of a metric ton technology that is determined to have a high impact across or more) and seeking a low COGS. Presumably, the multiple biomanufacturing scenarios would reinforce its company is not encumbered with needing to provide a importance to the industry and likely elevate its priority high degree of regional manufacturing and can therefore for development and adoption. centralize manufacturing.

13 Tom Ransohoff. “Challenges in Forecasting Biopharmaceutical Demand and the Value of Flexibility.” BioProcess International West Conference (Feb 27-Mar 2, 2017, San Francisco) BPOG Technology Roadmap 22 ©BioPhorum Operations Group Ltd OVERVIEW

An example of this situation is a contract manufacturing and continuous downstream processes will be explored as organization that has ready clients and with contracts part of this study since single-use equipment for protein already in place for significant production quantities. purification is beginning to emerge on the market. Because of this confidence, the need to supply this The set of business drivers that would indicate a facility additional capacity quickly is not as great due to proactive of this type would be representative of a company that supply planning and forecasting and the use of existing has a need for a moderate to high volume of manufacture capacity that will suffice until the new facility or capacity with a relatively small product mix. Since a limitation of can be built. continuous processing is the relatively longer start-up and For modeling purposes, a conventional mAb process was shut-down times required, too many product changeovers chosen as a base case and is further described in Appendix would not be productive. Instead, to handle additional B. Process assumptions were made that represent a typical products or respond to increasing demand, 2,000L process that is currently in commercial operation as a bioreactor trains would need to be added. Scaling out, basis for comparing other configurations and improved rather than scaling up, allows a high degree of assurance or alternative technologies. An important question is how in technology transfer and product comparability if scale should these facilities evolve to accommodate the new changes can be avoided. Scaling out can also accommodate business drivers? Another is how does a company minimize a lower degree of certainty in product success and demand or optimize additional investments to adjust to new forecasts. This scenario can also be an advantage from throughput requirements, improved process conditions the market launch perspective as a company moves and/or yields and recoveries and/or to different product out of late-stage development towards commercial types? For a company considering the construction manufacturing, provided that product demand is fairly of a new facility of this type, there may well be design consistent throughout the year. considerations or features that could be incorporated that By reducing new construction times and costs with the will allow additional flexibility in the future and/or the single-use technology, increased flexibility allows a more adoption of new technologies. rapid response to increases in demand that may have been 4.1.2 difficult to predict. This approach will likely resound with a Scenario 2: Intermediate-scale single-use perfusion smaller company, or one with a smaller product portfolio, In this scenario, a facility is designed to take advantage of that cannot take the same risk on a single product as a single-use technologies; i.e. to reduce the capital outlay larger company with a greater number of assets. Building needed to establish the required capacity and to reduce a traditional facility that then does not achieve full the requirements for ‘steam and clean’ during operation. utilization and/or cannot be easily adapted to other needs, However, significant throughput is still required, may place a financial burden on a smaller organization that potentially approaching that of Scenario 1. As a new could be difficult to sustain. construction, this facility type may also be enabled with A conventional mAb process was chosen as the base the higher cell-line productivities recently demonstrated case for this scenario and for modeling purposes similar on development-phase products. Thus, the high outputs to Scenario 1. This allows direct comparison to a large- typical of a more conventional stainless steel facility may scale stainless steel bioreactor facility at high throughput be possible in smaller reactors operated in perfusion mode. (although, to date, not many mAbs are produced using A perfusion-based platform could be a highly productive perfusion cell culture). The base case process and one and would be particularly well suited to unstable assumptions for Scenario 2 are described in detail in molecules and/or cell lines that may have high media feed Appendix B as the first point of comparison. However, requirements. this same perfusion platform can also be employed for While single-use bioreactors are now available up to poorly expressing proteins, enabling acceptable bioreactor 3,500L, 2,000L is currently the largest size in a readily- volumetric productivities that would otherwise require available single-use configuration. Thus, the upstream extremely large batch-mode reactors. Thus, Scenario 2 production bioreactors in this scenario are limited to could also be used to represent a low productivity case, 2,000L. To achieve the significant mass output of drug such as the production of a highly complex glycoprotein of substance, a continuous cell culture is employed. As the low demand (e.g. rare diseases). base case for this scenario, the downstream portion of the facility is still operated in a batch mode, which is typically the case regardless of whether the upstream process is continuous or not. However, semi-continuous

BPOG Technology Roadmap 23 ©BioPhorum Operations Group Ltd OVERVIEW

4.1.3 from multiple trains and pooling intermediates accordingly. Scenario 3: Intermediate-scale multiproduct Also, with multiple products, each with a limited demand, single-use fed batch the advantage of a highly optimized and productive process The facility type in this scenario, similar to Scenario 2, may not justify the additional development work needed. leverages single-use technology. The same 2,000L scale Thus, Scenario 3 represents a set of market influences bioreactors are used since they are the largest that and business drivers requiring an even higher degree are readily available at present. However, rather than of flexibility than Scenarios 1 and 2. A company may be operating them in perfusion (continuous) mode, they facing a high degree of uncertainty in its product mix and are run here as fed-batch reactors, similar to Scenario 1, a business calling for the production of several products, albeit at a smaller volume. In this scenario, batch mode is potentially with varying processing requirements. The employed in the upstream process to enable more products scenario also places a premium on speed by avoiding to be accommodated in the same equipment and with the relatively longer start-up and shut-down times shorter cycle times than if they were run in a continuous associated with continuous processing. Similar to mode. This design is intended for small- to medium-volume Scenario 2, increasing output can be accomplished by products that individually have demands on the order scaling out, rather than scaling up, which allows a high of kilograms per year but in aggregate could amount to degree of assurance in technology transfer and product hundreds of kilograms of annual output. comparability since scale changes are avoided. There is The advantage of a batch process is that variations in the added advantage that dramatically different product process requirements and unit operation sequence may demands can be accommodated. be easier to accommodate for a portfolio of products that The base case process and assumptions for Scenario 3 are do not all use the same process. Longer processes with a described in detail in Appendix B along with all major and larger number of operations, some of which are potentially minor parameters required to construct such a model. complex (such as conjugation reactions), would be Again, for simplicity, the base case is derived from a typical considerably harder to link together and automate if they mAb process similar to Scenarios 1 and 2. were to be run as a continuous process. 4.1.4 A facility of this type offers considerable flexibility both Scenario 4: Small-scale <500L portable facility in terms of adapting to different products in a campaign This scenario represents a very different set of mode but also by allowing scale-out in a similar manner requirements than the previous three. Here, a highly to Scenario 2. Initially, with perhaps just a single train, modular construction would be required so that the production campaigns of various durations could be whole process and any associated required infrastructure scheduled and sequenced in a way to respond to demand could be made portable. The process equipment may for individual products. If one train can no longer support even be built into a vehicle to enable rapid transport overall demand, more production trains could be added and deployment to disparate locations. An example of the same type and scale. Campaign durations would be of this may be a highly complex biologic, such as a relatively short since the bioreactor is sized for overall vaccine or allogeneic cell therapy, which requires capacity and not the annualized demand of any one given regional manufacturing. The flexibility of decentralized product. Shorter campaign lengths may be advantageous manufacturing may be a key driver because of regional in terms of the speed of response, lower quality control governmental policies and/or the need to be close to the costs (due to fewer batches for release) and for critical- intended patient population because of product shelf- path clinical production. For this reason, this scenario is life and stability, shipping requirements and/or regional becoming increasingly popular for manufacturing products emergency responses, such as pandemic flu. in clinical development by contract manufacturing To be portable, by definition, such a facility would be organizations and innovator companies due to the large for relatively small production quantities. For products number of clinical candidates, low facility fixed costs and with large overall global demand or increasing demand, reduced changeover times. It is also being considered by more modules could be distributed in strategic locations. larger companies as a more cost-effective way to launch To be responsive to growing demand and construction, new products, particularly those that are targeting smaller installation and qualification would have to be fairly rapid. patient populations. Alternatively, modules could be built ahead in anticipation This facility could also more easily accommodate varying of deployment and held in reserve, but this presents an cell-line productivities by mixing and matching equipment investment risk and additional inventory costs.

BPOG Technology Roadmap 24 ©BioPhorum Operations Group Ltd OVERVIEW

To make these portable facilities as compact as possible, are removed from the patient, brought to a centralized a highly intensified process would be advantageous to processing facility, expanded and/or treated and then minimize space requirements. Process intensification using returned to the same patient. continuous process technologies upstream, downstream Since a patient has to wait for treatment delivery once or both, could be employed. Single-use technology their sample has been collected and genotype and/ would likely be the design of choice, but not necessarily, or phenotype has been identified, production cycle if construction can be rapid and the equipment can be times would need to be short. As such, batch processing dedicated. Stainless steel process equipment would probably is the most efficient approach due to the small not be as dependent on a steady supply of disposables/ volumes and the simpler set-up and execution compared consumables, which could be problematic depending on to continuous parallel processing. The volumes and the the facility’s location. bioreactors, if required, would likely be less than one liter. In addition to the business drivers already mentioned, this These procedures would likely begin as manual processes scenario may be associated with a relatively low certainty but automation may be applicable as the patient count of approval or market acceptance. Unknown or uncertain increases. Due to the critical need to segregate patient requirements for regional manufacturing may also be a samples, single-use technology is the only alternative. factor. This reality would be accommodated by a gradual While automation equipment for vessel and fluid handling build out, or in this case a scale out, with the addition of could be shared, any surface wetted with a patient- more portable facilities. Thus, a high up-front investment specific cell or gene therapy preparation would have to be could be avoided by starting with just a few prototypes. completely dedicated. More production trains could be added of the same type Cycle time includes shipping the sensitive preparations to and scale. and from the processing facility and thus quality release If this scale-out could be accomplished quickly, this and shipping also need to be done as expediently as application could also be used to satisfy the needs of possible. Thus, quality control assays testing for safety and multiple products by quickly assembling modules with potency would have to be designed with this critical-path varied process sequences. However, in a continuous production schedule and limited product shelf-life in mind. mode, this approach would most likely be unsuitable for a Clearly, speed of manufacturing is paramount in this large number of products running in parallel. Production scenario. However, speed to build or the ability to expand cycle times would not have to be short if the equipment is capacity rapidly may also be important as patients are dedicated and can run over extended periods of time and identified for treatment. Through the use of modular demand is predicable for a given region. construction, a company could avoid a high up-front Also, depending on how remote or isolated the location of investment by starting with a small facility and then operation, advanced infrastructure may not be available. expand its footprint as demand increases or add additional In this case, generation of clean water and other utilities facilities that are strategically located. Perhaps modules may have to be built into the modules. could be designed as a single unit to be placed in a facility The base case process and assumptions for Scenario 4 are designed for parallel processing. The product mix would described in detail in Appendix B along with all major and probably be relatively small but some processing flexibility minor parameters required to construct a model. may be required to adapt to patient needs. 4.1.5 Specifically for cell therapies, the reader is referred to Scenario 5: Small-scale <50L for the recently published National Cell Manufacturing personalized medicines Consortium (NCMC) Cell Therapy Roadmap. In the In contrast with Scenario 1, with a facility designed to future however, synergies regarding this facility type produce one or a small number of products in very high and associated new technologies with other modes of quantities, Scenario 5 represents a facility where many biomanufacturing should be explored. Also, while cost may individual preparations are made in parallel. These be less of an issue in these early therapies, at some point preparations would typically be made in very small the COGS will have to be addressed. This would include volumes as would be the case for treating individual understanding what overall design aspects most impact patients. Some such facilities have already been built the COGS, including how the final product is shipped to the and are in operation but they stand at the dawn of patient and administered. personalized medicine. These products may be as complex At some point, a whole new paradigm may be considered as those mentioned in Scenario 4 and perhaps even more for delivering a therapy to a patient. Currently, the specialized, such as an autologous cell therapy. Here cells therapeutic preparations are made in a centralized or

BPOG Technology Roadmap 25 ©BioPhorum Operations Group Ltd OVERVIEW

regional facility. However, manufacture at bedside could 4.2.2 be envisaged using a modular and mobile approach. Low-volume drug product manufacturing scenario However, a number of questions would have to be Low-volume products are expected to become an essential addressed, such as how do E2E costs compare with a more part of the industry’s portfolio in the next 5–10 years. centralized facility? And could this be made highly portable While small in overall batch size, many of these products and brought to a patient’s bedside for deployment or can be considered high value hence any loss of product would it have to be set up in a hospital permanently? during manufacturing operation has a considerable business impact. The ‘right first time’ quality is of utmost 4.2 Drug product scenarios importance. Today’s relatively high reject rates, often due The primary scope of this first roadmap edition is drug to the insufficient quality of raw materials (particularly substance. This section provides a high-level view of the primary packaging) or processing steps, are deemed not scenarios impacting drug product, with metrics and initial acceptable. Continuous improvement efforts in the coming considerations to deliver the anticipated 10-year targets. years will drive the yield up to >99%. Future portfolio This is a topic for more detailed consideration in future needs will require faster product changeover, format roadmap editions. changes and product release times. It is expected that drug 4.2.1 product manufacturing processes will be further simplified Introduction e.g. by consistently using pre-sterilized ready-to-fill From a drug product perspective, there are two base primary packaging and, hence, avoiding in-house washing scenarios: ‘low volume’ and ‘high volume’, which are and sterilization steps. determined by the overall production volumes of the A high degree of automation further reduces operator particular product in terms of annual throughput. Devices interference and testing requirements. As an example, and combination products are considered out of scope for consider the time and resources currently spent for glove this first edition of the technology roadmap. In general, integrity testing for isolators, or the full-time equivalent technologies for drug product are primarily determined needs for processing in classified areas, which still by the product presentation, i.e. the primary container needs operators and exhibits very complex operating and container closure system rather than the therapeutic procedures. Further, the broader use of single-use/ modality. As a standard, fill/finish processes are operated disposable components instead of multi-use stainless steel with minimal human interface, with isolator technology will reduce cleaning and validation needs. Highly flexible and restricted access barrier systems both considered multiproduct/multipresentation lines (‘modular filling equally qualified technical solutions for ensuring sterility. lines’) will become a new industry standard. As is the current conventional approach, it is assumed the bulk drug substance is received at the drug product manufacturing site as pre-formulated, or formulated, and ready to fill. It is also assumed that the product has extended room temperature stability, such as formulations of mammalian cell culture (i.e. CHO) and E. coli-derived active pharmaceutical ingredients, as well as vaccines. This will significantly reduce the complexity of logistics of drug substance, drug product and all drug product manufacturing operations. Although not specifically addressed, personalized medicine can be considered to be the extreme version of the ‘low volume’ scenario, although there may be some additional challenges that are not specifically addressed in this version of the technology roadmap.

BPOG Technology Roadmap 26 ©BioPhorum Operations Group Ltd OVERVIEW

The metrics and 5–10-year targets for the low-volume DP scenario are shown in Figure 11.

Figure 11: DP metrics for low-volume scenario

Driver Metric Current state Five-year 10-year

Total cost to supply reduction 25% 50%

Cost Cost of upfront investment in manufacturing (includes $80–100m $70–80m $50m (due to cloning and utilities, equipment for formulation and filling) saving in validation)

Time to release product (E2E) 6–8 weeks 3 weeks 1–2 days

Speed Speed to market 5 years 3 years 1 year

Facility build speed 3–5 years 2 years <1 year

Flexibility Technology transfer from development to commercial 6–12 months 2–3 months (cloned filling) 2–4 weeks

Success rate 96–97% 98–99% >99%

Quality Deviation-free fill lot 80% 90% >95%

Reject fill rate 1–2% 0.5–1.0% <0.1%

4.2.3 High-volume drug product manufacturing scenario Besides the trend for more personalized medicines that require less overall manufacturing volume, high-volume products are still considered part of industry’s future portfolio, especially when dealing with mass-vaccination products. These high- throughput production lines still require significant upfront investment but prices will fall due to industry standards because of less customized technical user requirements. This will also benefit the speed of building a new facility. The metrics and 5–10-year targets for the high-volume DP scenario are shown in Figure 12.

Figure 12: DP metrics for high-volume scenario

Driver Metric Current state Five-year 10-year

Total cost to supply reduction 5–10 % 15–20%

Cost Cost of upfront investment in manufacturing (includes $80–100m $80–90m $75–80m (e.g. standardization utilities, equipment for formulation and filling) among companies)

Time to release product (E2E) 6–8 weeks 3 weeks 1–2 days Speed Facility build speed 3–5 years 3 years 1–2 years

Flexibility Technology transfer from commercial to commercial 12 months 6 months 3 months

Success rate 96–97% 98–99% >99%

Quality Deviation-free fill lot 80% 90% >95%

Reject fill rate 1–2% 0.5–1.0% <0.1%

BPOG Technology Roadmap 27 ©BioPhorum Operations Group Ltd OVERVIEW

4.2.4 Initial considerations to deliver 10-year targets The vision is that both scenarios will be in commercial production in 10 years and the technology will be readily available to support these priorities. The following changes are anticipated to achieve the five- and 10-year targets. In five years: 1. a reduction in the current level of defects by 50% for primary packaging materials 2. availability of formulated ready-to-fill bulk drug substance, moving the formulation activities closer to the manufacture of the drug substance 3. Multipresentation fill lines so that vials, cartridges and prefilled syringes can be handled on the same line 4. product changeover in less than two hours 5. for validation, a facility is available for commercial manufacture in less than two years 6. product release within 15 days.

In the 10-year timeframe: 1. minimize the process losses to less than 1% and aim for ‘zero’ defects 2. particle-free operations 3. product release time of less than one day 4. have the option of a fully continuous process from drug substance through drug product, or drug substance and drug product in one facility 5. unique identifier for each primary container 6. new facilities to go from scoping to commercial production in less than one year and to push the validation towards the equipment manufacturers so when arriving at the facility the equipment is fully validated 7. product changeover in less than 15 minutes from the last good vial of one batch to the first good vial of another batch and product lines capable of doing many different types of products including highly potent compounds 8. reduced costs, including new facilities with less than $15m in capital expenditure and less than $1m per year in operating costs. More specifics on drug product-related needs can be found in the In-line Monitoring and Real-time Release, and Modular and Mobile reports.

BPOG Technology Roadmap 28 ©BioPhorum Operations Group Ltd OVERVIEW

5.0 Regulatory strategy 5.1 Introduction

The vision Current industry initiatives The Technology Roadmap Regulatory Strategy will plan The stakeholders include global health authorities that interactions between regulators and industry to enable authorize initial and revised licenses throughout the successful new technology implementations. This will biopharmaceutical product lifecycle and other agencies require support from the regulatory agencies of multiple involved in the generation of standards and guidance. global health authorities. Updated or new regulations and The scope and initial outcomes have been shared through standards may be needed to enable implementation of the the FDA’s Emerging Technology Team (ETT) and its roadmap for legacy, current and future manufacturing. Office of Biotechnology Products, and the Medicines The regulatory vision will be achieved by assuring early and Healthcare products Regulatory Agency’s (MHRA) and continuous involvement of regulatory stakeholders Strategy and Innovation Group. There is recognition of as the industry develops and adopts the new technologies the importance of partnering support if the industry is to and manufacturing approaches envisioned by the successively innovate and be aligned with the expectations technology roadmap. set by agencies (see Table 3).

Table 3: Regulatory agency policy and communications

Policy or communication Issuing agency Issued Related links Key points relevant to the BPOG Technology Roadmap

Advancement of emerging technology FDA Draft guidance, FDA ETT • commitment to supporting and enabling the applications to modernize the December Guidance modernization of pharmaceutical manufacturing pharmaceutical manufacturing base 2015 FDA blog, • one long-term solution to avoid drug shortages M Kopcha: • implementation of emerging technology is critical Modernizing to modernizing pharmaceutical manufacturing and FDA blog, L Yu: improving quality Continuous • leveraging existing resources to facilitate regulatory Manufacturing review of submissions involving manufacturing technologies likely to improve product safety and quality

21st century cures US Gov. and December FDA blog, R • builds on the FDA’s ongoing efforts to advance medical FDA 2016 Califf: 21st product innovation and ensure that patients get access Century Cures to treatments as quickly as possible, with continued assurance from high-quality evidence that they are safe and effective

Innovation and technology MHRA 2012 MHRA • single point of access to expert regulatory information, Innovation advice and guidance Office • help organizations to develop novel manufacturing processes

Innovation and technology EMA August 2014 EMA • discussion platform for early dialogue with applicants, Innovation particularly small- and medium-sized entities, to Task Force proactively identify scientific, legal and regulatory issues of emerging therapies and technologies

There are shared expectations for technology innovation suppliers and government hybrid organizations. BPOG between regulators and industry (e.g. product quality will provide opportunities to share progress along with and security of supply). Initial interactions with the future plans with a more extensive regulatory and FDA and MHRA recognize that the successful global biopharmaceutical audience. implementation of technology innovation requires There are multiple regulatory stakeholders and the interaction with country-specific agencies and with global Regulatory Strategy will be to continue to engage with regulatory groups. BPOG will also seek to collaborate with them to provide updates and seek feedback. Examples of other industry groups, government agencies, academia, key regulatory stakeholders are in Table 4.

BPOG Technology Roadmap 29 ©BioPhorum Operations Group Ltd OVERVIEW

Table 4: Illustrative examples of regulatory stakeholders

Regulatory stakeholders Name*

Country Agency ANVISA, MHRA, EMA, FDA, PMDA, SFDA

Global Agency WHO, PIC/S, ICDRA, ICMRA

Government BARDA, NIST, NIIMBL, AMBIC

Industry PhRMA, EFPIA, CASSS, ISPE, PDA, IFPAC

Standards ICH, USP, ASTM, ASME

*Acronyms are defined in Section 8

Context of how regulatory challenges 5.2 Scope can be overcome This section reflects the needs, challenges and solutions The biopharmaceutical industry has extensive related to regulatory requirements from the individual experience of where the rapid implementation of roadmap reports. The scope for the Regulatory Strategy is: improved or new technology has been necessary to continue its mission to reliably provide therapeutics 1. anticipating regulatory needs, challenges and to support public health. Examples include: potential solutions for the implementation of the BPOG Technology Roadmap for new and legacy • unmet medical needs – the development of penicillin (approved) products based on the modalities and streptomycin introduced the ‘new’ technology and product classes defined in Section 3 and the of large-scale stirred submerged fermentations. biomanufacturing scenarios described in Section 5 The seasonal influenza vaccine introduced non- 2. assessing cross-company experiences and egg-based manufacturing substrates. More observations from new product development and recent viral pandemics (e.g. Ebola and Zika) have lifecycle activities to determine regulatory challenges accelerated the need to explore technologies to incremental and step-change technology changes such as small-scale single-use systems 3. understanding and interpreting the scope of • drug shortages – there is an industry focus to existing regulations and standards (associated minimize the frequency and length of occurrence with product development, product maintenance, • data integrity – there was an international response equipment, facilities and supplier services) to to reduce the risk to computer systems of potential identify those regulatory solutions that enable year 2000 failures. A global effort ensued to assess an effective roadmap implementation and remediate impacted computer systems. 4. strategic planning to engage with regulators and The Regulatory Strategy will leverage applicable other agencies, including identification of specific learnings from other industries. In addition, other points of contact within agencies and industry, regulated industries (e.g. aviation, nuclear, automotive to ensure an ongoing and consistent dialog. and food) have worked successfully on implementing The scope does not directly include other regulatory cross-industry technologies with their respective areas (i.e. product labeling, pharmacovigilance, clinical regulatory agencies. There has been a consistent safety and efficacy), although there could be indirect recognition of benefits, including a joined-up (global) impacts that will be assessed and mitigated. inspectorate, cross-industry standards, consideration of human factors, training/education programs and IT capabilities. The Regulatory Strategy will plan interactions between regulators and industry to enable successful new technology implementations.

BPOG Technology Roadmap 30 ©BioPhorum Operations Group Ltd OVERVIEW

5.3 Regulatory needs, challenges 2. patient needs – industry alignment on risk and potential solutions mitigations for enhanced security of supply and support for accelerated submissions to address 5.3.1 unmet patient needs Needs and challenges 3. process improvement – increased process robustness In this report, regulatory considerations are described that and data integrity; implementation of new relate to the various technology needs and challenges. technologies, such as continuous manufacturing and These are summarized in Table 5 and illustrate the need real-time release for revised or new standards and guidance. In summary, these are expected to impact on: 4. standards and harmonization – driving global harmonization while recognizing current diverse 1. resourcing – reduced resources to ensure requirements, including supplier/manufacturer regulatory oversight interfaces.

Table 5: Illustrative summary of regulatory needs and challenges

Technology Roadmap team Regulatory needs and challenges considerations

Cross-team Global acceptance of Real-time Release

Validation strategy for continuous manufacturing

Modular and Mobile Room assessed/inspected like equipment

Change management process adapts to include Modular and Mobile approaches

Qualification/validation adapts to include Modular and Mobile approaches

Harmonize regulatory and building code requirements

Process Technologies Single-use standards

Continuous processing standards and guidance

Reduce/eliminate changeover between products

‘Ball room’ design and declassified facility

Global regulatory harmonization

Viral validation strategy

Parallel processing of multiple products

Cross-use of consumables among products

In-line Monitoring and Real-time Development of rapid microbiological methods and testing Release Diverse global regulatory requirements

Process robustness

Security of supply chain

Knowledge Management Expectations on compliance activities, including training workforce, facility and production characteristics

Expectations on implementation of International Council on Harmonization Q10

Data integrity expectations for product and facility lifecycle

Automated Facility Use of open system reference standard architecture

Use of open system reference software modules

Establishing data integrity in an open system environment

Real-time upgrade can be done without product impact

Integration: interface protocols and vendor specifications

Standards for integration: vendor specifications

Common lexicon

Data integrity

Increased use of automation/robotics

Supply Partnership Management Transparency and trusted supplier data

Standardized data packages

BPOG Technology Roadmap 31 ©BioPhorum Operations Group Ltd OVERVIEW

5.3.2 5.4 Regulatory interaction recommendations Potential solutions 1. the Technology Roadmap team will continue to share Regulatory policy, guidance and standards progress and seek global regulatory stakeholder agency An analysis of technology gaps is being performed to identify engagement related policies, guidance and standards, and assess them as 2. effective collaborations will be established with being: related industry groups and standards bodies to assure • existing and sufficient: those that remain valid alignment, minimize overlap and enable effective • supplement/change: those that may require updating implementation • non-existent/insufficient: new or major revisions 3. solutions will be developed to address the needs and required. challenges identified as described in 5.3.1. The Technology Roadmap team intends to further engage 4. regulatory updates on the Technology Roadmap content with regulators to co-develop solutions, depending on the and its implementation will be made at key industry topic/challenge, with focused discussions between subject events and through publications matter experts and cross-industry groups. 5. training and education events will be designed and made available for global industry and regulatory Initiatives and global approaches supporting new stakeholders. technology policy, guidance and standards • contact has been made with representatives from current regulatory initiatives. The Regulatory Strategy proposes additional engagements to establish the right connections, seek alignment and avoid duplication • key industry events are being leveraged to share Technology Roadmap progress, develop industry positions and widen engagement with stakeholders • a holistic approach is being taken to develop regulatory solutions across suppliers, manufacturers and regulators; further strengthened with opportunities for independent academic peer review. Education and training • a shared understanding of challenges and proposed solutions is critical to successfully implementing the Technology Roadmap. The team provided an education session to the FDA’s Office of Biotechnology Products in September 2016 and further sessions will be offered as the Technology Roadmap develops. Other approaches are being explored to engage with relevant regulatory agencies.

BPOG Technology Roadmap 32 ©BioPhorum Operations Group Ltd OVERVIEW

6.0 6.0 Conclusions and recommendations Participation levels in this first edition of the BPOG Technology Roadmap demonstrate that the industry will openly share technology strategy and that organizations are broadly moving in the same direction with shared challenges. The development of the Technology Roadmap has highlighted the importance of collaboration to overcome challenges and develop solutions that will benefit all stakeholders and, ultimately, our patients. The first edition kick-starts the roadmap initiative and has imbued a sense of momentum to evolve to the next level of maturity and operationalize the roadmap. The steering committee recommends several steps that the industry can take to move the industry forward. ‘Industry’ is defined as inclusive of all stakeholders (end-users, suppliers, regulators, etc.). These steps are: 1. build awareness of the Technology Roadmap and encourage engagement through proactive communication activities with industry public events, networks and within your organization 2. engage with key industry organizations and gather feedback from the industry to form a response to the roadmap, align efforts and consider funding routes 3. identify collaboration opportunities in response to roadmap needs to accelerate innovation initiatives and roadmap ‘quick wins’ 4. develop and track industry analytics to understand the ever changing market trends and progress of innovation 5. widen the participation to engage key stakeholders, including regulators and academics, to effect the implementation of the roadmap vision 6. broaden the scope of the roadmap effort with new areas of focus and continued future editions 7. nominate and recruit subject matter expertise for future roadmap activities (e.g. industry benchmarking and tracking, trend analytics, collaborative projects, communications, regulatory interactions and/or input) for the roadmap’s second edition process. This technology roadmapping effort is an evolving, dynamic and open process. We welcome comments from all industry stakeholders and look forward to continued growth in membership, further accelerating and broadening our industry impact. Please go to the BPOG website to learn how to become part of this worldwide effort for the biopharmaceutical industry. http://www.biophorum.com/category/resources/technology-roadmapping-resources/introduction/

BPOG Technology Roadmap 33 ©BioPhorum Operations Group Ltd OVERVIEW

7.0 7.0 References [1] National Cell Manufacturing Consortium, A Technology Roadmap to 2025, Feb 2016. http://cellmanufacturingusa.org/sites/default/files/NCMC_Roadmap_021816_high_res-2.pdf [2] http://www.ifm.eng.cam.ac.uk [3] Datamonitor: Pharmaceutical Key Trends 2011 – Pharmaceutical Industry Infrastructure Overview [4] Ecker, Dawn M, Susan Dana Jones, and Howard L Levine. “The Therapeutic Monoclonal Antibody Market.” mAbs 7.1 (2015): 9–14. PMC. Web. 21 Mar. 2017. [5] BPTC. Bioprocess Technology Consultants, Inc. bioTRAK® database; 2016. http://www.bptc.com [6] Ibid. [7] Sourced from Bert Frohlich, Shire, UML Biopharmaceutical Summit, March 2012 [8] clinicaltrials.gov [9] Market Insight: The changing biopharma risk equation, DDNews, August 2016, (http://www.ddn-news.com/index.php?newsarticle=10840) [10] BPTC. Bioprocess Technology Consultants, Inc. bioTRAK® database; 2016. http://www.bptc.com [11] Ibid. [12] KMR, Bernstein Analysis, Bernstein-Global-Pharma_22Sept2015_RnD-Productivity-Trends.pdf [13] Tom Ransohoff. “Challenges in Forecasting Biopharmaceutical Demand and the Value of Flexibility.” BioProcess International West Conference (Feb 27-Mar 2, 2017, San Francisco) [14] National Cell Manufacturing Consortium, A Technology Roadmap to 2025, Feb 2016. http://cellmanufacturingusa.org/sites/default/files/NCMC_Roadmap_021816_high_res-2.pdf

BPOG Technology Roadmap 34 ©BioPhorum Operations Group Ltd OVERVIEW

8.0 Acronyms/abbreviations

Acronym/abbreviation Definition ADCs Antibody drug conjugates AMBIC Advanced Mammalian Biomanufacturing Innovation Centre ANVISA Agência Nacional de Vigilância (Brazil) ASAP Accelerated seamless antibody purification ASME American Society of Mechanical Engineers ASTM American Society for Testing and Materials AEX Anion exchange BARDA Biomedical Advanced Research and Development Authority BCC BCC Research BPOG BioPhorum Operations Group CASSS California Separation Science Society CMO Contract manufacturing organization CEX Cation exchange COGS Cost of goods CPP Critical process parameter CQA Critical quality attributes DF Diafiltration DP Drug product DS Drug substance DSP Downstream process or downstream processing E2E End to end EFPIA European Federation of Pharmaceutical Industries and Associations EMA European Medicines Agency EPO Erythropoietin FDA Food and Drug Administration (USA) FDA ETT Food and Drug Administration Emerging Technology Team FDA OBP Food and Drug Administration Office of Biotechnology Products FTE Full-time equivalent GMP Good manufacturing practice ICDRA International Conference of Drug Regulatory Authorities ICH International Conference on Harmonisation ICMRA International Coalition of Medicines Regulatory Authorities

BPOG Technology Roadmap 35 ©BioPhorum Operations Group Ltd OVERVIEW

Acronym/abbreviation Definition IFPAC International Forum for Process Analysis and Control IP Intellectual property iPSC Induced pluripotent stem cells ISPE International Society for Pharmaceutical Engineering ITRS International Technology Roadmap for Semiconductors KMR KMR Group, Bernstein Research mAb Monoclonal antibody MHRA Medicines and Healthcare Products Regulatory Agency (UK) MVDA Multivariate data analysis mRNA Messenger RNA NASA National Aeronautics and Space Administration NICE National Institute for Health and Care Excellence NIIMBL National Institute for Innovation in Manufacturing Biopharmaceuticals NIST National Institute of Standards and Technology OPEX Operational expenditure PDA Parenteral Drug Association PhRMA The Pharmaceutical Research and Manufacturers of America PIC/S Pharmaceutical Inspection Co-operation Scheme PMDA Pharmaceuticals and Medical Devices Agency (Japan) QA Quality assurance QBD Quality by design QC Quality control QRM Quality risk management RABS Rapid access barrier system RTR Real-time release SFDA State Food and Drugs Administration (China) SUS Single-use system TOC Total organic carbon UF Ultrafiltration USP Upstream processing USP US Pharmacopeial Convention WHO World Health Organisation

BPOG Technology Roadmap 36 ©BioPhorum Operations Group Ltd OVERVIEW

9.0 Glossary

Term Definition Batch USP Both batch and fed-batch production bioreactor processes, seed bioreactor perfusion coupled to batch or fed-batch production bioreactor Continuous USP Perfusion process in production bioreactor where either a) only cells are retained or b) both cells and product are retained; low molecular weight components are removed Batch DSP Traditional purification process where multiple chromatography cycles are pooled prior to next unit operation Semi-continuous DSP Includes many variations on the same theme, e.g. connected process, straight through process, ASAP, pool-less process; pooling occurs at final ultrafiltration/diafiltration step, typically only one column of the same resin for each chromatography step Continuous DSP Continuously moving product, typically multiple columns of the same resin for each chromatography step Traditional biologics An aggregate of prophylactic and therapeutic products derived from natural sources. This class would include traditional whole-cell vaccine preparations and immunoglobulins derived from donor plasma, such as blood factors and anti-venoms Monoclonal antibodies All immuno-proteins that have a mAb backbone, typically genetically engineered and derived (mAb) and derivatives from recombinant organisms. The class contains naked mAbs, bispecific, multi-specific mAbs, ADCs, fragmented antibodies and mAb-fusion proteins. The hallmark of these therapeutic proteins is their specificity for their target, such as a receptor or antigen. In the latter case, they can be used to impart passive immunity Other therapeutic All other therapeutic proteins except those that may be used as a vaccine antigen. proteins (non-mAb) Replacement enzymes, scaffolds, growth factors, erythropoietin, insulins and related fusion proteins are all part of this class. Many of these proteins are highly glycosylated, which imparts additional specificity and/or activity, whereas most mAbs have a less complex carbohydrate structure and, in some cases, none Antigens and viruses A wide range of molecules used to provide active immunity as a vaccine. The antigen molecule(s) can be proteins alone or in combination with other moieties that can elicit an immune response in the host. Examples include chemically detoxified proteins (toxins) from pathogenic bacteria, recombinant surface proteins, viral capsid proteins and oligosaccharides or polysaccharides conjugated to a protein expressed in recombinant organisms. Viruses propagated in a host cell line can be used as either attenuated live or an inactivated vaccine Gene and mRNA/ Gene therapy is an emerging field with the very recent approval of first products. These oligonucleotide treatments are attractive as they may be curative by replacing or repairing a defective gene. therapies The coding sequence is typically delivered via a viral vector but other delivery mechanisms are also used, such as lipids. Short-chain oligonucleotides are also being used as therapeutic agents as a means of controlling or altering the regulation or expression of a native gene Cell and tissue therapies In this product class, whole cells or a tissue are delivered as the therapeutic agent, such as a graft. A cell therapy can be autologous where the cells are derived from the patient, modified or activated and returned, or allogenic where the cells are all derived from a common source. Thus, in the latter case, all patients receive the same cell or tissue genotype. Induced pluripotent stem cells (iPSC) are derived from skin or blood cells that have been reprogrammed back into an embryonic-like pluripotent state that enables the development of an unlimited source of any type of human cell needed for therapeutic purposes

BPOG Technology Roadmap 37 ©BioPhorum Operations Group Ltd OVERVIEW

10.0 Appendices Appendix A – Product classes Below are some further details on the five major product classes described in Section 2.2. • Recombinant proteins – mAbs and their derivatives: CHO is the predominant expression system while new mammalian and non-mammalian expression systems emerge mostly dependent on glycosylation/glycol engineering needs. While fed-batch operations are dominant, repeated fed-batch or continuous upstream operation are also established technologies. Initial product recovery is widely generic since Protein A can be used to capture most antibodies. Purification technologies for mAb backbone are to a large extent generic and performed with aqueous buffer systems. Post-translational modifications like pegylation or drug conjugation are typically performed on purified intermediates. • Other therapeutic proteins: (non-mAb): Several different expression systems have been employed for these proteins due to the specific product characteristics required. Different production bioreactor modes are used to accommodate specific growth needs or product stability attributes. Many of these proteins are highly glycosylated and may be subject to degradation or proteolytic cleavage where the mean residence time of the product species may be critical to its quality. Perfusion cell culture is typically applied in the latter case. Purification technologies for proteins are, to a large extent, generic unit operations and performed with aqueous buffer systems. Post-translational modifications, such as enzymatic modifications, are typically performed prior to purification while pegylation or acylations are typically performed on purified intermediates. • Antigens: Vaccine antigens are produced with a variety of processes. Toxins are made from pathogenic bacteria and are then chemically detoxified. Recombinant surface proteins and oligosaccharides are typically expressed in recombinant organisms or attenuated-live or inactivated (possibly recombinant) viruses. • Gene therapy: Oligonucleotides as a therapeutic agent can be manufactured with virus-infected cell cultures and thus require very similar technical equipment as traditional CHO systems. However, the use of such facilities for making other products is restricted due to regulatory constraints and concerns of cross-contamination of virus particles. Solid-phase chemistry can, in some cases, be an alternative for the synthesis of oligonucleotides or peptides rather than being biologically produced. Downstream processing of gene therapy products uses similar unit operations as some therapeutic proteins, sometimes involving solvents but with adapted ligands for chromatography steps. • Cell therapy: In this case, the cells themselves are the product and delivered as the therapeutic agent. Here, technologies have to be developed and validated to transfer these living cells aseptically into appropriate devices and quickly delivered to the patient. For autologous therapies, cells are taken from the patient for expansion and/or modulation and thus also require careful extraction, shipping and tracking to the production facility. For allogenic cell therapies, characterization and release of the seeding material and the final product requires new ways of thinking.

BPOG Technology Roadmap 38 ©BioPhorum Operations Group Ltd OVERVIEW

Appendix B – Detailed modeling results The impact of technologies that may confer a significant benefit to the biopharmaceutical manufacturing industry can be assessed using quantitative modeling. Process cost models can be useful in elucidating the complex relationships between cost, operating efficiency, speed, time value of money and the labor and material requirements for the various unit operations in a given manufacturing process. In building the Technology Roadmap, models were built to compare the five manufacturing scenarios considered in the roadmap. The modeling was intended to identify the high-cost components and features of both the process and the typical facility design, and to perform sensitivity analyses as a means of prioritizing aspects of the roadmap. The approach taken, and options to expand on this work in the future, is discussed in this section.

Scope of the model The process models were based on the biomanufacturing scenarios defined within the Technology Roadmap: 1. Large-scale stainless steel fed batch – low cost at high utilizations, high capital and long build times 2. Intermediate-scale single-use perfusion – medium throughput production of a broad variety of proteins, more easily reconfigured or ‘scaled across’ 3. Intermediate-scale multiproduct single-use fed batch – medium to low throughput production of a very broad variety of proteins, more easily reconfigured or ‘scaled across’ 4. Small-scale <500L portable facility – low throughput production units, also can be rapidly ‘scaled across’ and deployed into multiple regional markets 5. Small-scale <50L for personalized medicine – very low throughput, patient-specific preparation. Many production units, globally distributed. Assumptions have been mapped for Scenarios 1–4, with Scenario 5 remaining undefined at this time, to focus on the most common production modes in current operation. The data work focused on Scenarios 1 and 2. The remaining scenarios are intended to be worked on for future editions. The data produced is intended to support the roadmap conclusions and inform subsequent roadmapping efforts within the biopharmaceutical industry.

Model assumptions and input parameters The BioSolve software application from Biopharm Services Ltd has been used to construct process models for the different scenarios described above. The Biosolve model is available on the Biopharm Services Limited website. The BioSolve modeling tool is widely used (by more than 40 companies), 50% of which are suppliers modeling the value of their technologies to end-users. The application area encompasses a wide variety of operational configurations, including batch and continuous upstream operation and batch, and semi and fully continuous operation downstream. The assumptions defined for each of the scenarios considered by the Roadmap can be seen in Table 6. These were supported by a wide range of parameters defining both the upstream and downstream operations for each scenario type.

BPOG Technology Roadmap 39 ©BioPhorum Operations Group Ltd OVERVIEW

Table 6: Assumptions used for process cost modeling

Minimum Maximum Notes

Indicative volume L 12,500 Working volume

Duration Months 2 12

Capacity kg/yr 500 4,000 Low titer, low kgs

Operation USP 24/7 Scenario 1 Operation DSP 24/7

Changeover Days 2 Redundant inoculum, frozen

Indicative volume L 200 2,000

Duration Months 12

Capacity 100 500

Operation USP 24/7 Scenario 2 Operation DSP 24/7

Changeover Days 0.5 Redundant inoculum, frozen

Indicative volume L 500 2,000

Duration Months 2 12

Capacity 100 1,000

Operation USP 24/7 Scenario 3 Operation DSP 24/7

Changeover Days 0.5 Redundant inoculum, frozen

Indicative volume L 10 200

Duration Months 12

Capacity

Operation USP 24/7 Scenario 4 Operation DSP 12/7

Changeover Days 0.5 Redundant inoculum, frozen

Application to the roadmapping exercise BioSolve process models and modeling results were used to map the main areas for improvement opportunities of today’s biomanufacturing processes. To achieve this objective, process scenarios for a typical mAb production process were designed using target production amounts of 1,000 or 100kg per year.

BPOG Technology Roadmap 40 ©BioPhorum Operations Group Ltd OVERVIEW

Table 7: Process flow used to conduct initial modeling exercise

Centrifugation/ 2° depth Protein A Virus CEX AEX F/T Viral filtration UF/DF Final 1° depth filtration inactivation filtration filtration

87% 96% 97% 98% 97% 97% 98% 98% 98% Yield Flow

Removed when 200 LMH Batch 35g/L Batch 45g/L 200 g/L 250 LMH 40 LMH 300 LMH using perfusion 400L/m2 PCC - 55 g/L PCC - 100 g/L 600L/m2 250L/m2 USP Comments

Modeling work focused on Scenarios 1 and 2 as a case study to investigate the value that process cost modeling could bring to a roadmapping process. Each of Scenario 1 and 2 was further split into several model instances; the process parameters assigned to these can be seen in Table 8.

Table 8: Scenarios used for initial round of modeling

Scenario Target production Batch (B)/ Reactor scale Campaign duration Number of reactors Reactor scale number per year Continuous (C)

1a 1,000kg B/B 12,500L stainless steel Floating Fixed Fixed

1b B/C

1c B/B 2,000L single-use bioreactor Fixed Floating

1d B/C

1e C/B

1f C/C

2a 100kg B/B 2,000L single-use bioreactor Fixed Floating Floating

2b B/C

2c C/B

2d C/C

BPOG Technology Roadmap 41 ©BioPhorum Operations Group Ltd OVERVIEW

To enable a better comparison of continuous and batch mode operations, cellular productivity was kept constant for batch and perfusion operations. The other parameters were selected based on our understanding of today’s best processes exercised in industrial settings. The model has the following assumptions built into it: 1. all cases refer to new build (greenfield) facilities 2. all cases refer to commercial manufacture (not clinical trials) 3. process changeover takes seven days for stainless steel processes and two days for single-use technology processes 4. the model keeps a stable 80% utilization assumption and allows downstream throughput to increase as titer increases 5. downstream processing becomes a bottleneck once the model hits eight cycles, after which bioreactor volumes start to reduce as far as 30% volume 6. labor costs are factored according to the following: • labor salary costs are input and the labor effort is estimated • direct operational hours are calculated based on direct operations, solution management and cleaning • Other costs are factored as: Supervisors 10% Quality assurance 22% Quality control 30% Indirect 38% Indirect breakdown Logistics 8% Engineering 15% Other 15%

BPOG Technology Roadmap 42 ©BioPhorum Operations Group Ltd OVERVIEW

Modeling results Table 9 summarizes the high-level results from the process cost modeling exercise. The reader should bear in mind that these are only initial results and will be further investigated in subsequent editions of the Technology Roadmap. The table shows the selected scenarios with their high-level costs of production per gram of antibody and associated capital expenditure. There are clear differences between the scenario types in terms of both COGS and capital cost of each facility. Table 10 maps out the COGS against batch size and campaign length in more detail.

Tableae 9: High-level ihee modeling modein results resuts (cost ost per per gram ram of o antibody antiod and and capi apitatal expenditure) ependiture

enarios

r a d e

ost per 49.5 ram

nstaed 65.7 apita

Leend r a d ed ath ioreator

Perusion ioreator

ost per Pooin point ram Bath donstream nstaed ontinuous donstream apita

© opriht Biopharm eries Ltd rihts resered sed ith permission

Key: gray circles – stainless steel, blue circles – single-use systems

Table 10: High-level modeling results for each scenario

Total Titer Throughput Bioreactor Cost Batch size campaign TIC MM Process Upstream Downstream Batches (g/L) (kg/yr) vol (L) ($/g) (kg) length USD (months)

Scenario 1a Fed-Batch SS Batch 5 1,010.7 12,500 36.4 23 43.9 9.0 112.1

Scenario 1b Fed-Batch SS Continuous 5 1,010.7 12,500 32.1 23 43.9 9.0 106.9

Scenario 1c Fed-Batch SU Batch 5 1,012.4 2,000 49.5 144 7.0 12.0 65.7

Scenario 1d Fed-Batch SU Continuous 5 1,012.4 2,000 42.4 144 7.0 12.0 60.2

Scenario 1e Perfusion SU Batch 1.2 1,000.9 2,000 54 193 5.2 12.0 46.9

Scenario 1f Perfusion SU Continuous 1.2 1,000.4 1,999 44.6 193 5.2 12.0 47.6

Scenario 2a Fed-Batch SU Batch 5 100 1,778 108.8 16 6.3 12.0 27.3

Scenario 2b Fed-Batch SU Continuous 5 100 1,778 95.8 16 6.3 12.0 25.4

Scenario 2c Perfusion SU Batch 1.2 100.1 603 168.8 96 1.0 12.0 23.0

Scenario 2d Perfusion SU Continuous 1.2 100.1 603 112.8 48 2.1 12.0 21.5

TIC MM USD – total install cost (TIC) in millions (MM) USD

BPOG Technology Roadmap 43 ©BioPhorum Operations Group Ltd OVERVIEW

Table 10 shows that the COGS decreases in every instance where continuous processing is deployed downstream. In most cases, the capital investment required to install a continuous process versus a batch process is also reduced. It is also clear that the capital investment required to install a factory based on single-use technologies is significantly less than that required to install stainless steel. This is balanced by a higher consumables cost required to run a single-use technology factory. The benefits of a modular approach, which enables a decrease of these costs, is discussed further in the Modular and Mobile roadmap report, while the Process Technologies roadmap report describes opportunities to reduce the consumables and process cost of each unit operation. Further cost benefits of a continuous facility are described in the Automated Facility roadmap chapter. Figure 13 depicts the contribution of different cost drivers on the cost per gram of antibody for Scenario 1 (1,000 kg/year production). Using stainless steel reactor systems at 12,500L scale (1a, 1b) results in capital expenditure being the biggest relative contribution to the costs of goods. This shifts to consumables when producing a ton of antibody per year using iure enario proess ost readon 2,000L disposable reactor systems. When using perfusion mode at 2,000L scale, the main cost drivers are consumables and media.

Figure 13: Scenario 1 process cost breakdown

enario Outputs

enarios

a

ost per ram

d enario a enario enario enario d enario e enario Other Laor e onsumaes aterias apita

Other nsurane aste et ontinuous proessin redues apita impat Laor Operators et Perusion sstems hae hih media ost impat onsumaes iters resins as Proesses d hiher onsumae omponent due to sineuse sstems aterias outions hemias apita Annual finance charge

© opriht Biopharm eries Ltd rihts resered sed ith permission OL

BPOG Technology Roadmap 44 ©BioPhorum Operations Group Ltd OVERVIEW

Parameter sensitivity analysis To better understand the sensitivity of different expenditure elements (such as capital, consumables and media costs on input variables such as cellular productivity, perfusion rate and downstream yields), a sensitivity analysis was completed using a continuum of input variables. The selected variables and ranges are listed in Table 11.

Table 11: Input parameter ranges for sensitivity analysis

Sensitivity range

Base - Base Base + Five years 10 years

Cell line productivity (pg/cell/day) 15 20 25 30 60

Fed-batch titer (g/L) 1.5 3 5 10 20

Fed-batch media cost ($/L) 30 15 10 7.5 5

Fed-batch culture duration (days) 12 14 16 10 7

Perfusion steady state cell mass (million/mL) 30 60 80 100 120

Perfusion media cost ($/L) 15 10 7.5 5 2.5

Concentrated fed-batch perfusion rate over last 10 days (reactor volumes per day) 2 1

Concentrated fed-batch peak cell mass (million/mL) 80 120

Concentrated fed-batch specific productivity 20 60

Concentrated fed-batch titer (g/L normalized for cell mass) 12 36

Protein A resin capacity 30 35 45 60 80

CEX resin capacity 60 90 120

AEX resin capacity (FT) 200 500 800

Chromatography cycles number Protein A 2 4 6 8 16

Chromatography cycles number polishing steps 1 2 3 4 8

Viral filter loading (kg/m2) 5 20 50

UF/DF loading (L/m2) 400 1,000 5,000

Chromatography cycle time, Protein A (hours) 12 24 36 32 48

Chromatography cycle time, polishing (hours) 3 6 9 12 18

UF – ultrafiltration, DF – diafiltration, ProA – protein A, CEX – cation exchange, AEX – anion exchange

The results of the sensitivity analysis were plotted to visualize the effect of technology improvements in each parameter. Overall, the factors of cell line productivity and cost of media showed a direct impact on the COGS for every case. Downstream costs of resins were surprising low due to the assumption of high resin cycling numbers (>100) that are typically used for commercial manufacture. Figure 14 demonstrates the trends shown in this analysis, using Scenario 1a as an example. The data shows percentage changes in the COGS for the modeled base case and each of the sensitivity models given in Table 11.

BPOG Technology Roadmap 45 ©BioPhorum Operations Group Ltd OVERVIEW

iure enario a sensitiit resuts isuaied on a time pot Figure 14: Scenario 1a sensitivity results visualized on a time plot

Base Base Base ears ears

e ine produtiit edath media ost L Pro resin apait resin apait resin apait Viral filter loading (Kg/m UF/DF flux (L/mhr

UF – ultrafiltration, DF – diafiltration, ProA – protein A, CEX – cation exchange, AEX – anion exchange

The low impact of downstream operations is due to the model considering each resin step independently of the rest. When looking at clinical operations, the cost of chromatography resins can be very high so gains in resin capacity are assumed to be of benefit in those cases. When the downstream data is considered in isolation (i.e. without the upstream processes), it is clear that the Protein A and viral clearance steps are the key operations where there is an opportunity for improvement. There may be further value to be elucidated in considering the downstream processes together in a more interlinked manner.

Process cost modeling recommendations Perhaps the greater value of the modeling for the roadmap is to test the impact of technology improvements within each scenario. Identifying technology areas that reduce the COGS within each scenario is very valuable in identifying which technologies will help us to meet our five- and 10-year metrics goals. Metric targets for facility flexibility, speed, cost and quality are summarized in Table 1. The modeling results contained in this Appendix can be reviewed against the technology recommendations in the Process Technologies roadmap chapter, to see which technologies may provide the biggest benefits and help achieve process goals. For individual organizations, the choice of manufacturing design and benefits of each technology must be considered on a case-by-case basis depending on their product demands and state of technology development or adaptation. Future value may be found by further developing the model to describe the process gains envisaged by each roadmap chapter. For example, modeling the effect of the in-line monitoring and real-time release processes described against the associated reductions in the cost of quality management could help guide teams in which parts of those proposals for focus on adopting for their particular manufacturing model. Likewise, modeling the benefits of automating a production facility, or of adopting a modular manufacturing approach can guide teams in making decisions in those areas.

BPOG Technology Roadmap 46 ©BioPhorum Operations Group Ltd OVERVIEW

For future roadmap editions, the following recommendations can be considered: 1. develop the model further to assess the impact of new technologies in monitoring, automation and facility management 2. develop the model to assess the impact of introducing new technologies into existing production facilities 3. further assess the benefits of continuous processing and its modes of introduction into a plant 4. assess the inter-relationships between unit operations, especially in the downstream operations 5. assess the impact of changes described to capital expenditure, as well as the COGS currently considered 6. consider additional scenarios (as defined in the roadmap) and consider new scenario modalities, such as cell therapy production models.

Appendix C – Antitrust statement It is the clear policy of BioPhorum that BioPhorum and its members will comply with all relevant antitrust laws in all relevant jurisdictions: • All BioPhorum meetings and activities shall be conducted to strictly abide by all applicable antitrust laws. Meetings attended by BioPhorum members are not to be used to discuss prices, promotions, refusals to deal, boycotts, terms and conditions of sale, market assignments, confidential business plans or other subjects that could restrain competition. • Antitrust violations may be alleged on the basis of the mere appearance of unlawful activity. For example, discussion of a sensitive topic, such as price, followed by parallel action by those involved or present at the discussion, may be sufficient to infer price-fixing activity and thus lead to investigations by the relevant authorities. • Criminal prosecution by federal or state authorities is a very real possibility for violations of the antitrust laws. Imprisonment, fines or treble damages may ensue. BioPhorum, its members and guests must conduct themselves in a manner that avoids even the perception or slightest suspicion that antitrust laws are being violated. Whenever uncertainty exists as to the legality of conduct, obtain legal advice. If, during any meeting, you are uncomfortable with or questions arise regarding the direction of a discussion, stop the discussion, excuse yourself and then promptly consult with counsel. • The antitrust laws do not prohibit all meetings and discussions between competitors, especially when the purpose is to strengthen competition and improve the working and efficiency of the marketplace. It is in this spirit that the BioPhorum conducts its meetings and conferences.

BPOG Technology Roadmap 47 ©BioPhorum Operations Group Ltd OVERVIEW

Roadmap intended use statement This roadmap report has been created, and is intended to be used, in good faith as an industry assessment and guideline only, without regard to any particular commercial applications, individual products, equipment, and/or materials.

Our hope is that it presents areas of opportunity for potential solutions facing the industry and encourages innovation and research and development for the biopharmaceutical industry community to continue to evolve successfully to serve our future patient populations.

Permission to use The contents of this report may be used unaltered as long as the copyright is acknowledged appropriately with correct source citation, as follows “Entity, Author(s), Editor, Title, Location, Year”

Disclaimer Roadmap team members were lead contributors to the content of this document, writing sections, editing and liaising with colleagues to ensure that the messages it contains are representative of current thinking across the biopharmaceutical industry. This document represents a consensus view, and as such it does not represent fully the internal policies of the contributing companies.

Neither BPOG nor any of the contributing companies accept any liability to any person arising from their use of this document.

BPOG Technology Roadmap 48