Unveiling the Future of Bioprocessing: Cell and Gene Therapy Purification Strategies
Table of Contents Letter of Welcome. . . . . . . . . . . . . . . . . . . . 2 Terry Pizzie
A Novel Resin for HCP Clearance: Bioprocess Intensification Through Mixed-Mode Chromatography. . . . . . . . . . 3 Thomas Valorose
Enhanced Efficiency in Cell and Gene Therapy Manufacturing: The Role of Plasmid DNA Purification . . . . . . . . . . . . . . 4
Scalable Downstream Purification: Safeguarding Yields of Lentiviral Vectors Expressed By Adherent- and SuspensionCell Systems. . . . . . . . . . . . . . . . . . . . . . . . . 6 Ian Scanlon
Cell and Gene Therapy: Triumphs, Trials, and the Road Ahead. . . . . . . . . . . . . . . . . . . 8 Marc Hummersone
Daniella Steel
T
his year has been an exceptional one for Astrea Bioseparations. We continue to surpass our own expectations thanks to highly dedicated teams working across all our company functions. In January 2023, we acquired Delta Precision. That move is expanding our offerings to include high-performance chromatography columns for biomanufacturing applications. Delta Precision is renowned for providing equipment of high quality and usability for protein purification, vaccine production, and drug development. In June, we marked an exciting new chapter in our history by joining forces with Biotage. By combining the expertise of the Delta Precision and Biotage teams with our innovative products, Astrea Bioseparations has become well positioned to drive novel breakthroughs in the chromatography sector. Following the successful 2022 launch of our Nereus LentiHERO solution for lentivirus-vector purification, which is based on our proprietary AstreAdept nanofiber technology, we have swiftly expanded our product line to include the ExoHERO and pDNAHERO tools for purification of exosomes and plasmid DNA, respectively. We also released larger formats for the LentiHERO technology. Such additions are meant to address growing demand for advanced tools in process development. The products also mark significant progress in our efforts to accelerate toward manufacturing scale. The HERO line of purification products is designed to improve upon traditional purification approaches, which often do not preserve their structural integrity while accommodating the large 2
BioProcess International
21(11–12)si
sizes of cell and gene therapy (CGT) products. We believe that the HERO method of rapid but gentle purification will maximize productivity in CGT biomanufacturing and that the technology’s adoption will empower users to accelerate workflows, streamline processes, and enhance process economics. Together, such features could provide a fast, environmentally conscious, and costeffective purification process that aligns perfectly with the needs of today’s therapeutic innovations. In addition to releasing cutting-edge nanofiber purification products, we have continued to expand our portfolio of resin-based separation media with the introduction of a mixed-mode adsorbent for host cell protein clearance that simplifies removal of impurities from monoclonal antibody products. Our new 12,000-ft2 facility in Canton, MA, opened in September 2023. We expect the site to increase our US manufacturing capacity significantly, including space for offices, cleanroom laboratories for column-packing services, increased warehousing, and dedicated column-assembly areas. I am proud of the entire Astrea Bioseparations team for their hard work and dedication to making this a year of significant growth and innovation. With much more to come for 2024, I wish you all the best in the new year. Terry Pizzie is chief executive officer of Astrea Bioseparations, Ltd., The Freeport, Ballasalla, Isle of Man, IM9 2AP, British Isles, UK; https:// www.astreabioseparations.com.
November–December 2023
Sponsored
A Novel Resin for HCP Clearance Bioprocess Intensification Through Mixed-Mode Chromatography Thomas Valorose
S
ince the introduction of recombinant biotherapeutics in the 1980s, many such products have gone on to become “blockbuster drugs.” Thanks to advancements in biotechnology, an increasing number of proteinbased biologics are addressing previously untreatable diseases. Therapeutic formats also have expanded beyond conventional monoclonal antibodies (MAbs) to include bispecific antibodies (bsAbs) and antibody–drug conjugates (ADCs). The emergence of such modalities has highlighted the need to improve biomanufacturing processes, especially considering increases in patient demand, product complexity, and industry competition. Many drug companies are intensifying methods to improve existing processes and meet new demands. Bioprocess intensification begins with examining all aspects of a manufacturing process to find strategies for increasing productivity, decreasing timelines, and reducing cost of goods (CoG). Applying process-intensification principles, for instance, a typical two-step polishing process could be condensed into a single step using mixedmode chromatography. Converting to a single mixed-mode chromatography step has many advantages. By leveraging multiple binding interactions, a mixedmode adsorbent can provide greater salt tolerance than ion exchangers and higher specificity than a hydrophobic-interaction medium. In turn, mixedmode media can maintain high yields at high levels of purity. Using mixed-mode chromatography enables significant cost savings associated with purchasing and maintaining requisite columns, resins, and buffers. Removing a purification step also moves products through a manufacturing process faster and eliminates labor costs associated with additional column packing and sample processing. One of the most important functions of a polishing process is to remove host cell proteins Sponsored
(HCPs), host cell DNA (hcDNA), and protein aggregates. To perform this critical step, Astrea Bioseparations has developed HCPure resin, which is designed to remove host-derived contaminants from a breadth of starting feedstocks, including those from Chinese hamster ovary (CHO), Escherichia coli, human embryonic kidney (HEK), and Pichia pastoris cell lines. HCPure resin can serve as the foundation of a flexible platform technology for achieving highly purified end products with minimal process adjustments. HCPure resin offers a distinctive binding profile compared with other mixed-mode media available today. Rather than being an ion exchanger with some hydrophobic properties, the resin works primarily through hydrophobic interaction, with hydrogen bonding serving as a secondary interaction chemistry. Such features give HCPure resin an advantage over other mixed-mode media and traditional ion exchangers during polishing steps: It retains its functionality at highconductivity load conditions, whereas ion-exchange media would require sample dilution to maintain their functionality. That advantage could prevent bottlenecks during manufacturing operations that require high-volume in-process hold tanks for dilution of protein A eluate. Astrea Bioseparations’ HCPure mixed-mode chromatography resin can be incorporated into a breadth of workflows. By being flexible enough to be effective, independent of starting feedstock and primary-capture method, the resin can serve as the foundation of an intensified purification process that can bring products to market quickly and costeffectively. Thomas Valorose is a product manager at Astrea Bioseparations; t.valorose@astrea-bio.com.
November–December 2023
21(11–12)si
BioProcess International
3
Enhanced Efficiency in Cell and Gene Therapy Manufacturing The Role of Plasmid DNA Purification Daniella Steel
Absorbance (mAU)
1,400 900 400 –100
4
Absorbance
1,900
Conductivity 0
20
40
60
Volume (mL)
BioProcess International
80
21(11–12)si
100
90 80 70 60 50 40 30 20 10 0
A
1,190 Conductivity
250
990
200
790
B
November–December 2023
150
590
100
390 190 –90
Conductivity (mS/cm)
2,400
Figure 2: (a) PuraBead hexyl adsorbent effectively removes open-circular DNA (ocDNA) contamination at process scale. (b) Agarose gel electrophoresis (AGE, 1% agarose) of pooled desalted fractions shows removal of ocDNA impurities from the elution pool (lane EL).
50
Absorbance 0
2
M
4
6
8
10
0 12 14 16 18 20
Conductivity (mS/cm)
Figure 1: Conditional screening of plasmid DNA (pDNA) capture from E. coli lysate onto PuraBead HF DEAE adsorbent using an automated liquid handler
uniform beads with nearly monodisperse particle sizes. The beads also show excellent porosity and flow properties. PuraBead HF adsorbents functionalized with diethylaminoethyl cellulose (DEAE) for capture (Figure 1) and hexyl ligands for polishing (Figure 2) are effective solutions to the purification and scalability challenges noted above. For scientists working at laboratory scale, both PuraBead HF adsorbent options (with DEAE or hexyl) are available in prepacked columns for automated chromatography processes and in small-volume prepacked columns for conventional workflows. For larger scales, such as for manufacturing applications, both adsorbent types are available in large-volume slurries. Available from small to large scale and in many formats, PuraBead HF adsorbents with DEAE or hexyl ligands enable effective capture and polishing of pDNA. Building on those proven technologies, Astrea Bioseparations has developed its AstreAdept next-
Absorbance (mAU)
P
lasmid DNA (pDNA) is a key intermediate for multiple stages of cell and gene therapy (CGT) manufacturing. Plasmids can serve as therapeutic payloads, building blocks for viral vectors, and templates for mRNA or closed linear-DNA sequences. Plasmid production by Escherichia coli fermentation also is common throughout the biotechnology industry, including at large scales. But process-related impurities in E. coli lysate must be removed before pDNA can be used in subsequent manufacturing steps. Such impurities include genomic DNA and RNA, host cell proteins (HCPs), and endotoxin. Manufacturers of pDNA also must separate out degraded products. “Nicking” of the DNA causes contamination of supercoiled (sc) sequences with opencircular (oc) forms, making a pDNA product less effective for transfection. Thus, regulatory agencies specify limits for ocDNA contamination in pDNA products. Removing ocDNA also increases plasmidproduct biological activity. Purification at laboratory scale is a familiar process to many biopharmaceutical scientists, but as pDNA purity specifications become more stringent, limitations of kit-based approaches are becoming clear. Industry needs a purification solution that scales from laboratory work to production. Adsorbents based on agarose beads remain the “workhorse” solution for pDNA purification. PuraBead technology from Astrea Bioseparations consists of
Column Volumes (CV) L
NB
EL
ocDNA scDNA
Sponsored
Figure 3: The open structure of AstreAdept nanofiber technology provides immediate access to a large surface area for lentivirus-vector binding.
Figure 4: Breakthrough capacity of a 4.7-kb plasmid construct on pDNAHERO 1 adsorbent for polish steps
Absorbance (mAU)
0.3
Absorbance
0.25
600
0.2 0.15
400
0.1
200 0
Pressure 0
2
4
6
Unit switched in-line
8
10
0.05 12
14
Volume (mL)
16
18
0 20
Delta Column Pressure (MPa)
By-pass (full-scale UV deflection)
800
Figure 5: Binding and elution of a 34.7-kb plasmid DNA construct with pDNAHERO 1 adsorbent for polish steps (EGFP = enhanced green fluorescent protein) Absorbance (mAU)
600
0.3
Absorbance
500
0.25 0.2
400 Conductivity
300
0.15 0.1
200 Pressure
100 0
0
0.05 50
100
Volume (mL)
150
generation chromatography matrix. The powerful convective-nanofiber adsorbent is characterized by an extremely high surface-area-to-volume ratio, so its binding capacity is large even at high flow rates. Concomitantly, the large flowpath diameter (~860 nm on average) maintains reduced backpressure and offers a low shear-flow environment (Figure 3). Such attributes address many difficulties associated with purifying large and fragile modalities — particularly CGTs — with convective chromatographic media. The open structure of the AstreAdept matrix nanofiber network circumvents issues such as inaccessibility of binding sites to large-sequence pDNA and product loss in dead-end channels. With such advantages, the AstreAdept matrix can increase purification efficiency significantly. Recently, Astrea Bioseparations has developed precision radial-flow bioprocessing devices incorporating AstreAdept technology. The pDNAHERO matrix for polish of pDNA products has a capacity of 4 mg/mL for a 4.7-kb plasmid at a residence time of 12 seconds (Figure 4). Of interest is that pDNAHERO adsorbent performance is not limited by plasmid size. Experiments have demonstrated successful binding and elution of Sponsored
0
Delta Pressure (MPa)
A
B
1: Ladder 2: EGFP strip 2 3: EGFP strip 1 4: EGFP elution tail 5: EGFP elution peak 6: EGFP flowthrough Most material is recovered in the 60% step elution (lane 5)
plasmids up to 34.7 kb in size (Figure 5) at fast flow rates (5 bed volumes/min on a 1-mL bed-volume device) while maintaining pressure at <1 bar. That makes feasible the use of a polish process across plasmid sizes without sacrificing productivity. The low operating pressure enables integration into single-use flowpaths without needing to reduce flow rate. Whether by using diffusive media such as PuraBead technologies or AstreAdept convective adsorbents, Astrea Bioseparations’ customers can address bioseparation challenges from laboratory to commercial scales. PuraBead resins provide effective solutions that are readily adaptable to specific purification needs, including flexible bed heights and characteristics that meet large-volume requirements for manufacturing scales. Meanwhile, AstreAdept technology addresses a critical unmet need in downstream processing of long plasmid sequences and fragile modalities by providing a next-generation, prepacked adsorbent that reduces processing time and increases process efficiency. Daniella Steel is senior manager for product strategy at Astrea Bioseparations; d.steel@astrea-bio.com.
November–December 2023
21(11–12)si
BioProcess International
5
Scalable Downstream Purification Safeguarding Yields of Lentiviral Vectors Expressed By Adherent- and Suspension-Cell Systems Ian Scanlon
6
BioProcess International
21(11–12)si
Absorbance (mAU)
120
2,000
80
Absorbance 1,000 0 0
40
Conductivity 50
100
150
200
Volume (mL)
250
0 300
3,500
160
Absorbance
3,000
120
2,500
80
2,000 1,500 1,000 0 0
40
Conductivity 50
100
150
Volume (mL)
200
0
Conductivity (mS/cm)
As large viruses (80–100 nm) with enveloped structures, LVVs pose problems for bind–elute chromatography based on conventional beaded adsorbents. The binding capacity of such adsorbents is limited to the outside surfaces of their beads. LVVs also are unstable in elevated salt concentrations. Exposure to concentrations >300 mM leads to particle aggregation and envelope breakdown. Historically, adsorbents for scalable bind–elute chromatography of LVVs have been functionalized with quaternized amines, relying on principles of strong anion exchange. That approach introduces complications during elution, necessitating use of high salt concentrations, often exceeding 1.2 M NaCl. Consequently, elution pools must undergo significant dilution, typically four- to fivefold, before proceeding to diafiltration and concentration steps. The Solution — LentiHERO Technology: Following successful commercialization of the Nereus LentiHERO device, a laboratory-scale purification device incorporating an AstreAdept adsorbent functionalized for weak anion exchange, Astrea designed the LentiHERO 1 and LentiHERO 10 prepacked devices for scalable radial-
160
3,000
Figure 2: LentiHERO 1 purification of suspensionexpressed lentivirus vectors (LVVs), from load to breakthrough Absorbance (mAU)
Addressing LVV Sensitivity
Figure 1: LentiHERO 1 purification of adherentexpressed lentivirus vectors (LVVs), from load to breakthrough; functional-vector recovery was 54%, as measured by green fluorescent protein (GFP) integration in Jurkat cells. Conductivity (mS/cm)
A
s demand for large-scale production of lentiviral vectors (LVVs) for chimeric antigen receptor (CAR) T-cell and other cell and gene therapies (CGTs) continues to rise, the need for a highly productive purification platform for highly sensitive vector targets becomes increasingly important. Thus, Astrea Bioseparations presents a novel radial-chromatography device incorporating AstreAdept functionalized-nanofiber technology for weak anion exchange. The Nereus LentiHERO device enables development of scalable purification processes that operate at relatively low salt concentrations and with minimal shear forces while maintaining a low delta column pressure (Δp) across the device bed.
chromatography processes. Their precisionengineered housing directs even flow through the functionalized adsorbent, facilitating LVV purification in a fast-protein liquid-chromatography (FPLC) system. LentiHERO technology is readily scalable for large-scale manufacturing processes. The following example demonstrates the potential workflow efficiencies to be gained.
Case Study in LVV Purification Using a LentiHERO 1 Device
Preloading Protocol: Feedstocks were processed and loaded according to the following conditions. Material from adherent-culture harvests was
November–December 2023
Sponsored
Figure 3: Clearance of human embryonic kidney (HEK293) host cell proteins (HCPs) and DNA (hcDNA) after purification with a LentiHERO 1 device (TFF = tangential-flow filtration)
Impurity Clearance (%)
Adherent-Culture Feed 100 90 80 70 60 50 40 30 20 10 0
100
Suspension-Culture Feed
100
100
HEK293 HCPs
hcDNA
77
39
3
Harvest
18
15 2
Elution
Harvest
Post-TFF
Elution
Table 1: Load and recovery of lentivirus vectors (LVVs) using a LentiHERO 1 device (p24 = viral capsid protein 24, TU = transduction units)
Purification with LentiHERO 1 technology recovered >50%
FUNCTIONAL LVV PARTICLES from both the
adherent- and suspension-based expression systems.
clarified using centrifugation at 3,000g and processed through a 0.45-μm filter. Then, 242 mL of each sample was loaded onto a LentiHERO 1 device for purification. Harvests from suspension-cell cultures underwent centrifugation at 3,000g, 0.45-μm filtration, and diafiltration with a 100-kDa molecular-weight–cutoff membrane. LVV solution was buffer-exchanged with 1 volume of 20 mM tris(hydroxymethyl)aminomethane (pH 7), 100 mM NaCl, and 5 mM MgCl2. Before loading onto a LentiHERO device, 200 mL of pretreated harvest was concentrated to <150 mL. Process data (not shown) indicated that clarification of the suspension feed caused no loss of functional LVV particles. Purification Protocol: Before the run, a LentiHERO 1 device was sanitized with 0.5 M NaOH at 1 bed volume (BV)/min. Equilibration, loading, postload washing, elution, and stripping all ran at 5 BV/min with a residence time of 12 seconds. Buffers for equilibration, loading, and postload washing all contained 20 mM tris and 5 mM MgCl2 (pH 7). Elution involved a step to 0.6 M NaCl in 20 mM tris (pH 7) and 5 mM MgCl2, and the elution pool was diluted immediately in an equal volume of equilibration buffer. Stripping was performed with 2 M NaCl. Finally, the LentiHERO 1 device underwent further sanitization in 0.5 M NaOH at 1 BV/min. Sponsored
100
Culture Format
Harvest Load
Load (p24/mL)
Load (TU)
Recovery (TU)
Adherent
242 mL
4.5 � 1011
4.7 � 109
2.6 � 109 (54%)
Suspension
200 mL
1.75 � 1012
2.5 � 109
1.3 � 109 (53%)
Key Findings
In the case study, purification with LentiHERO 1 technology recovered >50% functional LVV particles from both the adherent- and suspension-based expression systems (Figures 1–3). Those outcomes were achieved using a low-salt elution buffer. Operators also maintained a consistent residence time of 12 seconds throughout the loading and elution steps, resulting in a fast process. Of note is that 200 mL of suspension feed was processed using only 1 mL of adsorbent. Extrapolating from that result, a 250-mL radial unit should be sufficient for effective purification of a 50-L (clinical-scale) LVV batch. Remarkably, the unit could process that material in less than an hour, providing exceptional efficiency in large-scale production. Ian Scanlon is downstream processing lead for nanofiber technology at Astrea Bioseparations; i.scanlon@astrea-bio.com.
November–December 2023
21(11–12)si
BioProcess International
7
Cell and Gene Therapy Triumphs, Trials, and the Road Ahead Marc Hummersone
D
rug discovery and development have advanced significantly over the past 10 to 15 years. Cell and gene therapy (CGT) has become a central theme for scientists and investors alike. Many such therapies program patients’ immune systems to address disease, thus eliminating side effects and other problems that often hinder the utility of conventional small-molecule therapies. However, as our optimism for CGT grows, we should recognize that the processes we use to develop and manufacture such therapeutics need significant improvement to make them less costprohibitive and improve patient access. While celebrating the ability to treat and sometimes cure otherwise untreatable diseases, we have yet to consider fully how we can generate such medicines repeatedly and sustainably at acceptable prices for both patients and the planet. When the world needed a rapid response, “moonshot” projects for COVID-19 vaccine development pushed through in record times. That result highlights how we can quickly identify and optimize upstream processes for biopharmaceutical production. During public-health emergencies, drug makers can provide products that, compared with marketed biologicals, are relatively crude and unpurified; however, the industry still has a long way to go in terms of yield and reproducibility when considering the purification and isolation of novel therapeutics. If we scrutinize the processes and equipment that are used to discover and develop next-generation CGTs, we often find that platforms for previous drug modalities (such as monoclonal antibody therapies) simply have been repurposed. Although those repurposed technologies can work, they often are not fit for purpose or optimized for effective product delivery. Our next goal should be democratization of CGTs — not simply producing them, but rather making such medicines accessible to everyone who needs them. Although upstream processes receive considerable attention in the biotechnology and pharmaceutical industries, downstream processes tend to be overlooked. Purification is an area that is ripe for improvement and one that often adds significant costs
8
BioProcess International
21(11–12)si
to marketed drugs. Low yields and high costs associated with ineffective purification processes are commonplace, especially for viral-vector therapies. We can make a significant impact by focusing on such processes. But we can achieve improvements only through developing tools such as the HCPure resin and platforms such as the AstreAdept nanofiber technology, empowering users to discover and develop quality products with precision and purity while compressing processes and thereby reducing associated time and costs. AstreAdept technology represents the next generation of purification sorbents. The solution is designed specifically for therapeutics that are too large to permeate into the pores of conventional purification resins. Its expansive flowpath offers a low-shear environment that minimizes damage to the delicate viral vectors that offer hope to patients. Meanwhile, HCPure resin eliminates difficult-to-remove host cell proteins that can cause side effects in patients. These new bioseparation products exemplify our approach at Astrea Bioseparations. We refuse to accept that discovery is slow and that drug development must be expensive. Our products are designed to provide users with tools that are fit for purpose rather than repurposed, and we approach solution design emphasizing speed, accessibility, and quality for all. All companies should be doing the same to benefit both patients and the discoverers and developers of truly amazing CGTs. Marc Hummersone is senior R&D director at Astrea Bioseparations; m.hummersone@astrea-bio.com.
November–December 2023
Sponsored