Many industries, such as food producers or carmanufacturers, have embraced the concepts of highvolume/low profit operations for many decades (1,2);but why have such methodologies recently become soimportant to the biopharmaceutical industry?

Over the last five years, the industry has witnessed anumber of new challenges. While companies face the usualrequirement to grow revenues and profits, we seedeveloping pressure on drug prices as the large healthcareproviders (both public and private) re-evaluate many of thetherapies they use. Biopharm manufacturers are also facingrising operational costs (labour, energy and materials), aswell as having to manage substantial financial risks – forexample, deciding whether to build new facilities or useexisting plants. Finally, there remains the competitive threatof biogenerics, particularly from the emerging economiesof Asia. We now have a situation where manufacturersmust (in line with the process of industrial maturation)implement operational improvements across the entirevalue chain (see Figure 1).

In response to these challenges, the industry isconsolidating and embracing the concepts of ‘Lean’ and‘Operational Excellence’. Approaches include the use ofprocess analytical technology, platform-based processesand disposable production components, all of which canreduce time-to-market and the costs of development andmanufacture, through increasing product and processquality, and reducing waste – keys to achieving so-called‘Operational Excellence’.

WHERE ARE WE HEADING?There is no doubt that the product portfolio of thebiopharmaceutical industry is being driven bymonoclonal antibodies (MAbs) (see Figure 2). At the

same time, however, the industry is debating the exactscale of MAb manufacturing required – is multi-tonmanufacturing on the horizon? In a case study of verylarge-scale monoclonal antibody production (3), Kelleylooked at a single-branded product need of up to10,000kg/year; he concluded that such production ispossible with current technology and sees little or no needto turn to unproven alternatives simply for cost reasons.In our own study, looking at the production scale ofcurrently marketed biopharmaceuticals, we see at leastfour recognisable trends that will reduce averageproduction scales for novel protein drugs (4), concludingthat the future of antibody manufacturing will need tooffer solutions for a small group of therapeutics at theton-scale, and much more economical production ofproteins at 50-500kg volumes – something that can beachieved through Operational Excellence.

57Innovations in Pharmaceutical Technology

Bioprocessing

Operational Excellence – the Future of Biopharmaceutical ManufacturingIn response to mounting challenges, biopharmaceutical manufacturers are looking into ways of achieving Operational Excellence; Lean-enablingtechnologies play an important role in lowering costs and reducing waste,helping to make a process as efficient as it can be.

By Günter Jagschies at GE Healthcare Life Sciences

Günter Jagschies is Senior Director, Strategic Customer Relations (R&D), at GE Healthcare Life Sciences (Uppsala,Sweden). Now in his 24th year with GE Healthcare Life Sciences (formerly Amersham), he has held senior managementpositions in sales, marketing and R&D within the BioProcess division of the company. In his current role, he worksglobally with industrial collaborations and as a Business Advisor for the Life Sciences R&D and Business team. He is alsoco-author of Process Chromatography, a recent handbook for the biopharmaceutical industry.

Crowded markets� Fewer blockbusters� Redundant

medicines

Cost pressure� Reimbursement� Biogenerics

Maturing industry� 14% annual growth� More than 25

approved MAbs

New opportunities � New molecules� Vaccines

Figure 1: Biopharmaceutical trends – pressures and opportunities

PhRMA report 2006

Biotechnology medicines in development – by product category

Antisense 20Cellular therapy 21

Gene therapy 46Growth factors 16

Immune-based therapy 9Interferons 18

Interleukins 8Monoclonal antibodies 160

Recombinant hormones/proteins 43Vaccines 62

Others 62

Figure 2: How the industry is being driven by monoclonal antibodies (5,6)

PhRMA report 2004

Biotechnology medicines in development – by product category

Angiogenesis inhibitors 4Antisense 14

Cellular therapy 11Colony stimulating factors 2

Enzyme replacement therapy 5Gene therapy 23

Growth factors 7Immune-based therapy 10

Interferons 10Interleukins 8

Monoclonal antibodies 76Recombinant human proteins 23

Recombinant soluble receptors 2Signalling 3Vaccines 90

Others 34

IPT 26 2008 28/8/08 10:30 Page 57

EFFICIENT PRODUCTIONOperational Excellence describes the goal ofachieving superior yields, lead-time andthroughput whilst eliminating waste. It is asystematic approach to attaining world-classperformance in productivity, quality anddelivery of services and/or goods. Twoeffective tools for achieving this are Lean andSix Sigma. Along with GE Healthcare LifeSciences, one large biotech companyanticipated some of the industry pressures afew years ago and implemented a programme of Lean initiatives – recognisingthat rapid pace of growth created manyprocess improvement opportunities. Other

companies have also been heavily investing inOperational Excellence tools.

Lean can be broadly described through five principles:

� Define value – from the end-user’s point-of-view� Identify the value stream – develop an

understanding of how value flows to theproduct/service/person/object that is movingthrough the process

� Establish flow – create a situation where theproduct/service/person/object moving through theprocess does so with no interruptions or issues

� Examine ‘pull’ from the end-user – until somethingneeds doing from that perspective, don’t do it

� Describe perfection – it’s the aim that can never beachieved, but processes will be improved by trying

The main focus of Lean-enabling technologies is to reducewaste. Below is a list of the most common causes of waste,while Figure 3 shows where time and money can be savedwhen implementing Operational Excellence tools:

� Overproduction� Transportation� Motion� Waiting� Over-processing� Inventory� Defects� Under-utilisation

MANUFACTURING FLEXIBILITY AND CHROMATOGRAPHYThe industry has many tough decisions to make in theshort term to ensure its long-term viability. As cell cultureprocesses improve, the marketplace becomes morecompetitive, and we rapidly move towards multi-

product/multi-scale manufacturing at multiple sites; thereis then much speculation as to whether we need to investin new technology to meet manufacturing demands, orjust look at continuous improvement of currenttechniques and technology. Whilst we always have to beaware of possibilities through both of these routes, theconsensus in the industry suggests that much can be doneby focusing on the latter (3,8).

Over the last ten years, innovations from chromatographysuppliers such as GE Healthcare have helpedbiomanufacturers overcome many issues, providingdedicated tools for difficult challenges like mediascreening and column packing, and resins that can copewith high titres and specific purification issues – resultingin higher productivity. Looking to the future, we envisagethat in the coming decade:

� There will be an increasing need for MAbs as therapeutics

� Certain diseases, such as influenza, will necessitatespeed in development and production of vaccines

� Product titres for MAbs will reach levels of 5g/Lor greater

Figure 4 highlights our research into current industry bestperformances, summarising our findings and providingan indication of where the industry stands today and theimprovements that may be achievable.

Lean can be used to find ways to create better processflow, reduce downtime and stoppages, and reduce non-productive activities such as changeover time betweenproduction campaigns, cleaning procedures, orpreparation of equipment and process buffers. However,Lean and Lean-enabling tools can also be used in processdevelopment and optimisation. One example is HighThroughput Process Development (HTPD) on filterplates (for example, PreDictorTM plates), which allowidentification of the most appropriate chromatographyconditions for a process by running a series ofexperiments in a very short period of time, helping todefine the design space and limits of operation. Thisapproach is also used to select the most appropriatechromatography resins for each step.

Figure 5 highlights how Lean concepts can be applied toa chromatographic process step to improve on first-generation tools. In our Lean interpretation, we assumethat only the loading of material to be purified, theremoval of impurities from the bound product in severalwash and strip steps, and the elution of the purifiedintermediate product are essential value-adding activities.

58 Innovations in Pharmaceutical Technology

Before Cycle time or lead time

After Cycle time or lead time Savings

Figure 3: The benefits ofimplementing Lean and Six Sigma

Work: Wait time/waste:Value added Non-value time added time

Lean attacks waste here

Six Sigma attacks variation

IPT 26 2008 28/8/08 10:31 Page 58

The preparation of resin slurry and column packing, aswell as cleaning (CIP) and equilibration, are not.

USING MODERN RESINSHistorically, most downstream processing stems from thesuccess of first-generation chromatography products andprocess designs. In more recent years, however, a numberof improvements in resins have become available thatsignificantly impact chromatography capabilities. Onesuch example is Protein A resins where the introductionof higher capacity and more stable resins offers simplifiedCIP regimes and longer working lives. Another is novelion exchangers with higher capacities, increased volumethroughputs and multimodal selectivities that can shortena process from three to two steps.

Figure 6 (page 60) illustrates how utilising updated resinschanges the performance of the whole downstreamprocess. The assumptions and calculations used for thisgraph are from a model study performed at GE Healthcare(4). Costs relating to the facility, equipment and upstreamprocesses are excluded from the calculations, but alldownstream-related labour, resins, membranes, filters andbuffer costs are included. The absolute dollar numbers arenot directly applicable to all situations, but the relativeeffects of the changes are relevant for most cases.

When replacing the capture step (Protein A resin) in aclassic process using first-generation SepharoseTM FastFlow resins with a modern Protein A resin (MabSelectTM),process time is reduced and specific costs ($/g) arelowered by ~50%. Completely changing to recentlydeveloped resin technology reduces the process time totwo days and the costs to just above 30% of the originallevel (‘model process’). This translates into a two- tothree-fold productivity increase for the correspondingoriginal downstream process, in part by enabling manymore batches per year.

In short, Lean analysis can help manufacturers uncovermany of the real issues behind their costs and bottlenecks,such as using early generations of tools and poorlyoptimised unit operations.

TWO STEPS RATHER THAN THREEOne of the main features of Lean is to look critically atprocess steps; for example, if the same quality result canbe achieved with one less step, process engineers canimplement Lean to remove unnecessary operations. Forexample, applying at least three chromatography steps topurify protein pharmaceuticals such as antibodies is stillconsidered dogma by many. However, there are at leasttwo published variations of a two-step chromatography

process with the promiseto meet quality objectivesin many of the caseswhere they have beenapplied (9,10).

The use of a two-steppurification scheme maynot reduce the directcosts by much relative toother improvements, but smaller buffer volumes will berequired and consequently buffer preparation and storagerequirements will be reduced. This increasesmanufacturing flexibility, for example, by enabling aprocess to be located within the floor space of a smallexisting facility.

DISPOSABLE TECHNOLOGYOne of the key issues observed by those who have startedto promote disposable chromatography (for example,using membrane adsorbers instead of packed bedcolumns with beaded matrices) is the time consumedbefore and after the purification – that is, for columnpacking, cleaning and eventually storage.

Prepacked columns for fairly large-scale operation havebecome available (ReadyToProcessTM, GE Healthcare).With these pre-tested and pre-sanitised devices, resinslurry preparation and column packing is not necessaryand the corresponding waste is removed. The columns aremade of construction materials that can be disposed of byincineration, and are ready for use in single or campaignmode. However, since the devices are also stable in typicalCIP regimes, it is up to the user to decide whethermultiple-use is the preferred, more economical option.

CONTINUOUS VERSUS BATCH MODEThe idea of operating chromatography in continuous modeis not new but, in contrast to continuous processing ofbioreactors, it has never become widely established in large-

59Innovations in Pharmaceutical Technology

General best in class metrics Specific best in class metrics

� Facility utilisation = 80-100% Upstream� Product titer 5g/L

� Cost = 15% CoS (100 USD/g) � Product manufacturability checked atcell line selection level

� Process yield = 70-80%Downstream

� Time = 10 days upstream � Capacity on Protein A resin 40 to 50g/L2 days downstream � Capacity on IEC resin 80 to 100g/LLess than 7 days change � Number of production runs on Proteinover A resin: 300

� Bottlenecks removed = analytical support � Cleaning agent NaOH (even for Protein A)� Buffers produced in-line, fewer different

buffers

Figure 4: Best-in-class performance

Figure 5: Lean applied to chromatography, comparing a classic step with first-generation tools and a modern step with current tools

Classic step

Resin preparation*)Packing*)

Run

Equil. Load

Run

Equil. Load Wash Elution Strip

Wash Elution Strip

Excessive equilibration volume

Improved step

Resin preparation

and packing*)

*) Required only prior to 1st run

*) Further shortening with ready-to-process devices

CIP and storage #)

CIP and storage #)

Excessive elution volumeExcessive process time &)

Use of old tools &)

#) Optional with ready-to-process devices

#) Storage only post the last run

Value adding work

Non-value adding work

Required non-value adding work

IPT 26 2008 28/8/08 10:32 Page 59

scale biopharmaceuticalprotein manufacturing.From the processing of smallmolecules, it is known thatthis mode can increaseproductivity and reduce theconsumption of buffers, aswell as enabling high-resolution separations inisocratic mode. At GEHealthcare, we have tested a

specific variant of continuous processing referred to as 3-Column Periodic Countercurrent Chromatography (3C-PCC). We have used the capture step with a Protein A resinto establish the usefulness of this method, and havedeveloped a small-scale fully automated custom ÄKTATM

system to support those who wish to test the approach.

In our experience, the 3C-PCC system is a very simplehardware set-up that allows almost complete utilisation ofthe packed resin; since all product that breaks throughduring loading is caught by the next column in thesystem, less waste is generated throughout the operationand buffer consumption is reduced. In our labs, we have found up to a 30 per cent improvement inproductivity (g /L resin /hour), and a reduction in bufferconsumption by up to 30 per cent.

CONCLUSIONSManufacturers must carefully consider their processesand ascertain whether they are running optimally. As theindustry swings to mass production of monoclonalantibodies, it is now more important than ever to ensurethat development and manufacture is as efficient as itcan be.

We can see a few trends worth watching over the comingdecade; the market can be expected to develop in two areas:

� Large-scale, efficient manufacturing for blockbusters

� Small-scale, flexible technologies for rapidexperimentation and efficient production ofsmall volumes used in testing, clinical trials andsmall patient populations

Disposable solutions are in vogue, although ‘ready-to-use’is the relevant feature, offering the required manufacturingagility. Upstream and downstream processes are becomingmore integrated and more industry-friendly, instead ofremaining as separate unit operations with awkwardinterfaces. Continued improvements in chromatography(capacity, throughput and quality) and membranes will be

complemented by numerous approaches to improve ease-of-use and increase efficiency.

Suppliers to the biopharmaceutical industry will need totake these developments into account, and GEHealthcare plans to work closely with the industry toprovide the necessary tools for Operational Excellence.But we are not just looking at the short-term issues. It isimportant to recognise the challenges facingbiopharmaceutical production during the course of thenext decade – including flexible development andmanufacturing, multi-product facilities and tackling theissue of biosimilars. Regardless of whether a company isa contract manufacturer or produces its ownbiopharmaceuticals, there are huge pressures on theindustry to develop Operational Excellence. On top ofthis, we must never lose sight of the over-riding need toprovide innovative vaccines, MAbs and therapeutics thatcan improve health on a global scale.

The author can be contacted at gunter.jagschies@ge.com

NoteCapto, MabSelect, MabSelect SuRe, PreDictor,ReadyToProcess and Sepharose are trademarks of GEHealthcare companies.

References

1. Bornsztejn V and Ebbens M, Leaning the way

in Bioprocess, www1.gelifesciences.com

2. http://en.wikipedia.org/wiki/Sakichi_Toyoda

3. Kelley B, Very Large Scale Monoclonal Antibody

Purification: The Case for Conventional Unit

Operations, Biotechnol Prog, 23, pp995-1,008, 2007

4. Jagschies Q, Quo vadis? – Where are we heading

biopharma?, BioPharm International, in print

5. Biotechnology Medicines Survey, 2004,

http://www.phrma.org/

6. Biotechnology Medicines Survey, 2006,

http://www.phrma.org/

7. Driscoll C, A Lean Version of Lean,

www.contractpharma.com, June 2006

8. Process Chromatography Handbook, 2nd edition, eds

Hagel L, Jagschies G and SoferG, Elsevier, 2007

9. Vunnum S, Anion Exchange Purification of MAbs:

Weak Partitioning Chromatography, presented at ACS,

BIOT Division, San Francisco, September 2006

10. Johansson HJ, Current and future advances in

development of downstream processes for purification

of monoclonal antibodies, presented at IBC

Bioprocess International Conference, San Francisco,

November 2006

60 Innovations in Pharmaceutical Technology

Figure 6: Process economy gains through the use of modernresins: the combined effect of capacity, flow and lifetime

Classic process

(SepharoseTM

Fast Flow

resins)

Protein A update(MabSelect

SuReTM/SepharoseTM

Fast Flow)

Protein A 3-stepmodel process

(MabSelectSuReTM/ CaptoTMS/

CaptoTMQ)

Protein A 2-stepmodel process

(MabSelectSuReTM/

CaptoTMadhere)

DSP

proc

ess

time

[h]

Dow

nstr

eam

cos

t con

trib

utio

n [U

SD/g

]

40

35

30

25

20

15

10

5

0

37.4

19.2

11.810.9

Working volume 10,000LTiter 5g/LYield 79%

120

100

80

60

40

20

0

IPT 26 2008 28/8/08 10:32 Page 60