Monday, September 21, 2020

During this session, leaders from the Biopharma community will discuss their approach to creating, promoting, investing in, and accessing innovation, and ultimately, coming together to bring cures to patients.

In an effort to access cutting edge research and stay ahead of the curve during this unprecedented time, investors and pharma companies are taking less risks. What long term trends have fizzled out and what is on the rise to increase innovation? Are new strategies sustainable as increasing investment rounds requiring more and more funds to obtain new treatments, therapies and technologies?

Leading biopharma experts and VCs will share their insights on how to engage investors, which trends to watch for, how to position technologies and discoveries to attract investors, and navigate this evolving landscape.

An acceleration in the speed at which therapies are breaking through from basic scientific discovery to development is placing a premium on innovative business development models. Biotech innovators, Pharma, and VCs alike are competing for successful win-win deal-making models in the face of novel personalized applications of gene therapies, gene editing, regenerative medicines, vaccines, therapies, and companion biomarkers. This session will cover best practices and strategies in finding the right partnerships to engage.

Whether they’re public-private partnerships, CROs and biotechs or incubators and pharma companies, novel agreements between stakeholders are leveraging the resources and unique expertise to deliver previously only theoretical contributions to science. This session will walk through review successful and impactful partnerships that are shaping the future of drug development.

This past decade has seen an explosion of technologies that are transforming healthcare. From AI to cell and gene therapies to immunotherapies, we’ve seen significant advances in life sciences and the impact on patients, but what can we expect in the next decade? What are investors looking for in the next wave of breakthroughs? What areas of development excite them most? What new business models enable riskier investments to support the development of these novel methods and technologies?

One-to-one meetings will be available to schedule online throughout the duration of the digital event.

• Meetings on this day will start at 8:00am EST.

1:00 - 1:10 Praetego Inc

1:20- 1:30 Peri-Nuc Labs

1:30 - 1:40 Adamas Nanotechnologies, Inc.

The Startup Spotlight is a pitch competition featuring the most innovative startup biotech companies. This live pitch competition will give a group of hand selected startups the opportunity to pitch in front of the BioPharm America audience.

Startup Spotlight Presenters:

Atom Bioworks Inc.

Enrich Biosystems Inc.

GeneVentiv Therapeutics, Inc.

Opus Genetics

Qprotyn Inc.

SelSym Biotech

• Scaling In Vivo Therapies: Balancing proof of concept with manufacturability and addressing production challenges for clinical and patient needs
• Overcoming Delivery Challenges: Strategies to optimize delivery beyond the liver while reducing off-target effects
• Manufacturing and Quality Control: Tackling production, purification, CMC and analytical challenges unique to in vivo CAR-T therapies
• Understanding the “drug product” concept for in vivo applications
• Regulatory and Future Perspectives: Key regulatory considerations, emerging technologies, and lessons from current in vivo CAR-T programs
• What is the realistic timeline for in vivo therapies to become mainstream, and could they make ex vivo approaches obsolete?

Through a series of case studies this session will focus on the latest advances in next-generation therapies, moving beyond rare diseases to larger patient populations.

• Explore cutting-edge preclinical and clinical case studies driving the future of cell and gene therapy.
• Discover how data-driven innovation is transforming therapy development, from control and targeting to tackling undruggable diseases.
• Unpack the challenges vs. traditional approaches, advancements in new modalities, and the manufacturing needs shaping commercialization of future therapies.
  
  Key Case Studies Areas Include:
• Solid Tumours
• Beyond Oncology, Expanding Disease Areas: Autoimmune, Chronic Diseases, Neurology, Cardiometabolic
• Multiplex & Combination Products involving Cell Therapies
• Regenerative Therapies
• mRNA, RNA, & NK Therapies

• Starting with the End in Mind: Defining the Target Product Profile (TPP), Quality Target Product Profile (QTPP) and CQAs for an early-phase CGT
• Designing therapies with long-term success in mind—beyond Phase 1
• What Phase-Appropriate GMP really means for processes and documentation?
• Phase appropriate development strategies
• Accelerating INDs without compromising commercial viability
• Meeting regulatory expectation whilst building a robust data package for future commercialisation – What data is essential, what can wait? Minimum expectations for Phase 1, 2, and pivotal studies
• Platform approaches and building flexibility within your organisation to adapt to product changes
• Avoiding common early-stage CMC pitfalls
• GMP flexibility across development phases
• How to define the right commercial scale early in development?
• Designing for manufacturability and building scalability into processes from the start
• Key considerations for commercial scale up, reimbursement, raw material availability, and regulatory approvals
• Pathways to developing safe, reimbursable therapies with broad patient access

The panel discussion will focus on addressing the critical challenges of regulatory control, site management, system integration, oversight, training, and raw material handling to ensure the successful implementation of decentralized manufacturing in cell therapy.

• Key elements enabling scalable and consistent decentralized production of cell therapies at scale
• Strategies to establish robust regulatory controls, site management practices, and system integration for decentralized manufacturing.
• Multi centre coordination – Strategies for showing CGT product comparability across a network of decentral manufacturing sites
• Streamlining site training, raw material handling, and oversight to ensure consistency and compliance across multiple locations.
• Overcoming obstacles to decentralised manufacturing to enable access to approved CGTs
• Innovative patient-centric solutions to improve accessibility and reduce travel burdens for follow-ups e.g. mobile testing units and digital data transfer
• Addressing key hurdles in decentralization, including regulatory compliance, system interoperability, and maintaining quality standards across diverse sites
• Real world experience with a decentralized cell therapy manufacturing approach

• Process optimisation to increase manufacturing efficiency, scalability for ex vivo cell therapy
• Cutting costs at scale: Strategies to lower manufacturing expenses.
• Workflow simplification strategies
• Maximizing production: Best practices for commercial-scale manufacturing.
• Application of automation at scale: Where does automation meaningfully change cost, scale, and reliability versus simply improving internal operations?
• Novel reagents, manufacturing inputs, payload, delivery technologies, editing technologies to increase manufacturing efficiency
• Real-world insights: Case studies on successful scale-up strategies for ex vivo cell therapies
• How to make ex vivo cell therapies more efficient and cheaper?

Spotlight Presentation – Calling all Technology Thought Leaders!

This session is your platform to demonstrate leadership and technical expertise. Present your data, case studies, and innovative solutions to a targeted audience of senior scientists and engineers looking to solve hurdles in their production processes.

To learn more about sponsoring this session, contact us at Partners@informaconnectls.com

The presentation will discuss key challenges and solutions in scaling next-generation allogenic cell therapies. Topics include process intensification to increase doses per batch, advances in bioreactor platforms, strategies to manage donor variability, approaches to improving scalability while reducing manufacturing costs.

Spotlight Presentation – Calling all Technology Thought Leaders!

This session is your platform to demonstrate leadership and technical expertise. Present your data, case studies, and innovative solutions to a targeted audience of senior scientists and engineers looking to solve hurdles in their production processes.

To learn more about sponsoring this session, contact us at Partners@informaconnectls.com

This presentation will illustrate an enhanced analytical strategy execution roadmap with a focus on analytical automation and method performance optimization throughout product development and commercialization. After reviewing two critical guiding principles and a execution “roadmap”, we will do a deep-dive into an integrated and enhanced analytical development program and state-of-control process. Short case studies will be reviewed to illustrate the practical application of the guiding principles and the associated benefits.

Lessons learnt from recent approvals / late-stage products – How to approach regulatory guidelines for gene therapiesFeedback from FDA on expectations, common questions asked – What do the regulators need to know?Commercial product case study examples and strategies for BLA filingCommercial release packages – How to ensure right first time?Lessons learnt from recent CMC delays

• Opportunities for the integration of AI and machine learning in gene therapy process development, manufacturing and CMC
• Where can we use and integrate data?
• How to use statistics and modelling software for predicting performance in AAV processes?
• Combining mechanistic modelling with high-volume manufacturing data to address real process variability
• Modelling tools to enhance manufacturing success and reduce failure rates
• Integration to control process robustness and variability for application in gene therapy commercial phases
• Are we there yet to incorporate yet for gene therapies?
• Case study

The rapid advancement of gene therapy programs toward clinical and commercial stages requires robust analytical frameworks that support successful method transfer to commercial testing laboratories. Unlike traditional biologics, gene therapy analytical methods — particularly cell-based potency assays and molecular methods — are inherently complex and difficult to transfer without a well-defined analytical control strategy, where failures can delay clinical timelines and jeopardize commercial readiness. An effective analytical control strategy goes beyond method validation and encompasses a holistic framework including critical reagent qualification and lot-bridging, reference standard management, system and sample suitability criteria, statistical process control, risk assessment, and a comprehensive training and knowledge transfer strategy. Our experience with the transfer of expression and functional potency assays for AAV-based gene therapy programs demonstrates how pre-defined system suitability criteria, critical reagent qualification packages, and fit-for-purpose acceptance criteria serve as the foundation for a structured transfer process. A phased training approach combined with real-time data trending enabled early detection of performance gaps and facilitated timely corrective actions, while a risk-based prioritization of high-complexity assays such as functional potency allowed for efficient resource allocation and proactive risk mitigation. In conclusion, a well-defined analytical control strategy is the critical bridge between method development and commercial testing, providing the scientific and operational framework necessary to ensure assay integrity, regulatory compliance, and commercial readiness across all stages of gene therapy development.

• Do Multispecifics require new CLD strategies?
• Ideas for purpose-built CLD strategies for complex molecules
• Is there a need for new hosts beyond CHO and HEK?

• How are groups going from pools/clones -> scale-up-> first GMP/Clinical Supply
• What are the bottlenecks? Vector build, pool generation, analytics turnaround, comparability?

Cell line development and cell banking are frequently on the critical path for early clinical timelines, particularly as biologics increase in molecular complexity. This presentation highlights an integrated cell line development approach that improves speed and predictability while maintaining desired product quality. Targeted integration is used to reduce expression variability and screening burden, enabling faster and more predictable generation of productive and stable clones. Image-based machine learning models trained on early colony data reduce bench time for clone screening, shortening development timelines by approximately three weeks. In addition, next-generation sequencing is applied for broad-range viral detection to reduce testing turnaround time and accelerate conditional release of master cell banks to the manufacturing facility. Beyond acceleration, targeted genetic engineering strategies are employed to modulate product quality attributes, such as glycosylation profiles, without compromising productivity. Together, these case studies illustrate how an integrated toolbox combining molecular biology, genetic engineering, and data-driven modeling can accelerate timelines to first-in-human while delivering predictable, high-quality cell lines.

Most cell line development (CLD) groups are shaped by legacy platforms, data structures, and practices. Here, a CLD capability was established from a clean slate while pipeline delivery expectations, compressed timelines, and an evolving modality landscape were being addressed. Key early decisions will be described, including selection of host(s) and expression systems, definition of platform elements versus program-specific customization, and prioritization of investments in automation and in-process analytics. Management of speed-versus-quality trade-offs, and their impact on titer, timelines, and tech transfer risk, will be discussed. A key focus is building the right data foundation. The minimal day-one data model and its impact on workflows and on Upstream and Analytical interfaces will be outlined. Compromises, deferred capabilities, and lessons learned will be shared to offer a practical blueprint for (re)building CLD under real-world constraints.

• What workflows are people trying to push for speed whilst taking risks and how are those risks mitigated?
• What steps are people taking to ensure processes are aligned, consistent, and fast?
• How fast is “responsibly fast”?
• What are others doing to push that boundary?
• How can you shorten timelines but not compromise on quality?

• How AI and ML are being integrated into CLD workflows
• Examples of predictive models for clone performance, stability or product quality
• Where AI-driven CLD is already delivering measurable titre gains in the industry
• How can AI tools reduce the experimental load during clone selection?
• How to train your AI model – what data sets matter most?
• What tools and guidelines are still missing?
• How do we move from isolated success to standard practise?

The success of antibody therapeutics depends not only on progress in drug discovery research and antibody engineering but also critically on the development of antibody manufacturing processes and the establishment of fast-and-steady supply systems to produce high-quality products consistently. Daiichi Sankyo has recently established a novel CHO cell line, CHO-MK derivative (Masuda et al., J Biosci Bioeng. 2024;137(6):471-479), which differs fundamentally from conventional CHO cells according to our research in the MAB (Manufacturing technology Association of Biologics). This cell line exhibits approximately twofold faster growth rate and extremely high viable cell density in comparison with an existing CHO cell line, suggesting its promising application might provide opportunities to be a game changer in a biologics manufacturing area. This presentation will introduce the host expression system, cell line development and upstream, and downstream process development using a novel CHO cell line.

This work presents a new model for distributed biologics manufacturing, centred on an automated cell free platform that enables rapid, on demand production of VLP vaccines directly from DNA. Using Pichia pastoris lysates integrated into the MANGO device, the system eliminates reliance on cell culture and complex infrastructure, allowing vaccine candidates to be generated within hours and deployed in any setting. Early demonstrations across multiple VLPs establish the platform’s technical breadth, while advances in lysate analytics, DNA sourcing, reagent lyophilisation, and streamlined purification are being unified into a scalable workflow. Together, these innovations create a robust, decentralised manufacturing solution aligned with CEPI’s 100 day mission and offer a practical path toward globally accessible, rapid-response vaccine production

We have built a functional genomics platform supported by AI and ML that enables a high throughput combinatorial approach identifying multiple synergistic target genes that enhance cellular phenotypes such as for instance titer, doubling time, metabolism or PTMs. We will show a case study where CHO cell lines with single copy insertions of the coding information for an antibody could be optimized more than 5-fold for productivity parameters by manipulating multiple genes simultaneousl

• CRISPR/Cas9, MAD7, and proprietary platforms for precision engineering in biologics and cell therapy.
• Multi-transposon strategies to improve expression balance for complex biologics and cell therapy applications.
• Impact of vector topology, promoter engineering, and codon optimization on gene expression and product quality.
• Predictive analytics for high-performing clones and integration of omics data into cell line development workflows.
• Advancements in non-viral transduction technologies and their scalability for adherent mammalian cell culture systems.

• Examples of hosts with faster doubling time
• Does the quality and stability remain consistent?

We identified a novel transposase and characterized its activity in CHO cells and optimized the transposition conditions. Our results demonstrated higher integration efficiency and product titers compared to popular wild-type transposons. We have then demonstrated the transposase this Jump-In transposase is effective for forming high titer mini-pools and clones for monoclonal antibodies and bispecific antibodies. Additionally, we will report progress towards synthetic control systems for bispecific antibody chain ratio.

Therapeutic proteins undergo various post-translational modifications during manufacturing that can lead to undesired effects, such as increased immunogenicity, loss of efficacy, and decreased stability. As a proof of concept, we applied a multiplex gene knockout strategy to simultaneously disrupt multiple enzymes responsible for an undesired post-translational modification. By performing these edits directly in an enriched bulk CHO cell population expressing a bispecific antibody, followed by simultaneously high-throughput screening of gene knockout genotype and product production in the clones during standard cell line development workflow, effective PTM risk mitigation was achieved on a fast-tracked timeline.

An effective hepatitis C virus (HCV) vaccine must reproducibly elicit the production of broadly neutralizing antibodies (bnAbs). The HCV envelope E1E2 glycoprotein complex is a natural target of bnAbs but presents biomanufacturing challenges. First, it is critical to produce E1E2 with reproducible glycan content, as glycans regulate the exposure of epitopes that give rise to a bnAb response. Second, the expression host must consistently produce functional E1E2 glycoforms. We addressed these challenges using glycoengineered CHO (geCHO) cell lines that impart nearly homogeneous glycans to examine how controlling glycan content influences antigenicity and immunogenicity. We found that glycan content impacts both biochemical properties of antigens and the quality and uniformity of immune responses. This demonstrates the utility of the geCHO system as a biomanufacturing platform

- Ultra-high cell density (UHCD) cell banking presents a transformative opportunity to intensify seed trains by enabling direct inoculation into production-relevant cultures, potentially eliminating multiple expansion passages.

- The objective of this work was to generate a data-driven proof-of-concept demonstrating the feasibility of UHCD-enabled seed train intensification, while systematically evaluating the impact of seeding density and Di-Methyl Sulfoxide (DMSO) removal on growth performance, viability, and productivity for a fusion protein.

- Two exploratory shake flask studies and one statistically powered Ambr® 250 bioreactor study were conducted using a proprietary mammalian cell line expressing a fusion protein. Key response metrics included cell viability, metabolic profiles, viable cell density, product titer, and specific productivity to enable direct process comparability assessment.

Biopharmaceutical cell culture processes generate vast amounts of heterogeneous data spanning materials, process parameters, daily sampling, controller variables, and product quality. Traditionally, scientists spend significant time manually parsing, cleaning, and reformatting such data, limiting the speed of decision-making and process optimization. To address this challenge, we developed an automated data integration and workflow framework for upstream bioreactor processes. Through centralized data pipelines, process variables, including time-series measurements, and analytical outputs are automatically retrieved, reformatted, and connected to contextual metadata in a central database. This framework enables 80% automation in data transfer and processing, reduces manual effort by more than ten hours per scientist per week, and supports unified dashboards for visualization and analysis. By establishing seamless connections across experimental and pilot-scale runs, the platform provides an end-to-end digital paradigm that accelerates insight generation, enhances reproducibility, and facilitates stage-wise decision making in process development. Powered by this automated data foundation, we further developed machine learning models for predictive control of glucose in fed-batch bioreactors. Maintaining optimal glucose levels is critical for robust cell growth, productivity, and product quality, yet traditional bolus feeding strategies often lead to wide fluctuations and metabolic byproduct accumulation. Using historical datasets across multiple molecules and scales in the database, we trained regression-based models capable of predicting next-day glucose consumption with minimal errors. Implemented as a continuous feed calculator, this approach eliminates reliance on costly online sensors and enables more controlled metabolic environments without major infrastructure changes. Moreover, the model continuously improves as new process data are incorporated. Together, automated data workflows and predictive glucose control exemplify how digitalization and machine learning can transform upstream cell culture operations, providing scalable and data-driven solutions for accelerating biologics development.

Late-stage ADC programs may balance speed with closing process gaps and ensuring conjugation-ready material. This case study shows how a few lab bioreactors tested media and feeding to lock an upstream process—high titer, viability, and low HCP. Cleaner upstream output enabled a simplified downstream: only one purification post-conjugation and PS‑80 only added in ADC DS, reducing complexity, risk, cost, and time to clinic.

L-tyrosine's limited aqueous solubility complicates reproducible delivery in biologics manufacturing, driving operational complexity via separate bolus additions. This work evaluates Glycyl-L-tyrosine dipeptide incorporated directly into standard feed solutions evaluated in high-throughput microbioreactor platforms spanning multiple recombinant protein modalities. Optimized conditions yielded higher titers, improved bioreactor pH control, and comparable critical quality attributes versus bolus controls, while offering improved media stability, operational simplicity, and cost-competitiveness at manufacturing scale.

Bioprocessing diversion strategies have typically been very straightforwards: there are no diversions in fed-batch and diversions in “steady-state” perfusion don’t impact product quality. However, with the gradual adoption of integrated dynamic continuous processes there are now opportunities for diversions where product quality is also changing. How can changes in product quality be accounted for throughout the drug substance process, and how can the impact of diversions be predicted?

Offline sampling remains prevalent in mammalian cell culture due to unresolved limitations in inline sensor reliability and verification, which constrain fully automated control. This presentation explores Raman‑based pH modeling as a real‑time PAT to mitigate inline probe drift and reduce reliance on offline pH sampling. The chemometric model performance demonstrates feasibility and supports the long‑term objective of sample‑free bioreactor operations with improved efficiency and greater confidence in control.

• What does a “useful” sensor mean to you for upstream processing?
• How do you balance needing the sensors for online analytics and adding complexity and fragility to the process?
• Which sensor signals have improved upstream process robustness in your experience?
• What are the most common reasons sensor data gets ignored by operators or scientists?
• What decisions can be informed by sensors? How do you decide when to intervene and adjust the process?

• How can bottlenecks in downstream processes be addressed to handle high-titer and high-cell-density outputs from upstream operations?
• What strategies can mitigate the challenges of product aggregation and purity issues during the transition from upstream to downstream processes?
• How can variability in upstream culture conditions impact downstream purification steps, and what measures can be taken to minimize these effects?

• What are the practical steps to implement Raman in GMP (pre-processing choices, model lifecycle, drift handling)
• How can Raman spectroscopy be effectively implemented for GMP process monitoring?
• Example of robust, simplified control strategies that survive later stage/ commercial constraints

Excessive foam in bioreactors poses significant challenges in bioprocessing. Driven by gas sparging and stabilized by surfactants, foam and the resulting addition of antifoam agents can negatively impact oxygen transfer and cell viability, particularly in perfusion processes with high cell densities and filter-based cell retention devices. However, foam control is often inefficient as foam is mostly monitored manually, making antifoam dosage insufficient in times of sudden foam proliferation. To address these issues, we applied an AI-based foam detection system to enable real-time foam monitoring in benchtop bioreactors. Once validated, continuous foam monitoring facilitated closed-loop foam control. Further optimization of antifoam feeding strategies reduced antifoam usage without experiencing uncontrolled foaming. This technology also has potential for proactive foam control, where antifoam is added before a foam cap forms, and for scale-up in larger bioreactors. Overall, this study demonstrates that continuous foam monitoring, enabled by AI vision, paves the way for more effective foam control and improved bioprocess outcomes

• What can be utilized now that actually assists with scale up?
• What technologies are providing measurable improvements?
• What didn’t work?
• Are there any trade-offs or constraints?
  
  

Process intensification is a key enabler to maximize the output of our manufacturing networks. In this talk, we will describe the integrated continuous purification (ICP) platform and the Pionic equipment platform via several case studies. Together, these technologies will enable process intensification to solve real-world manufacturing challenges at Sanofi and, ultimately, industry-wide.

This talk explores the purification of increasingly complex multispecific biotherapeutics. Purification scientists face challenges in finding scalable, efficient methods as therapies advance. Capture redox adds a powerful technique to the purification toolkit, enabling the generation of multispecific antibodies through electro-steering heterodimerization during affinity chromatography—streamlining production and supporting the next generation of innovative medicines.

• Identify why standard mAb DSP platforms fail for ADCs and which assumptions break post-conjugation
• Understand which ADC attributes most strongly drive purification complexity (DAR distribution, hydrophobicity, linker chemistry)
• Evaluate practical purification options for removing unconjugated antibody, high-DAR species, aggregates, and free payload
• Decide where platform approaches are realistic vs where customization is unavoidable
• Anticipate scale-up and GMP risks earlier to avoid late-stage re-engineering

Protein therapeutics represent a major class of biopharmaceuticals; however, downstream process development has become increasingly challenging due to accelerated timelines and the growing complexity of molecular designs. To address these challenges, we developed an automated high throughput screening platform to optimize polishing chromatography conditions. Using two challenging bispecific antibodies, the capability of rapidly resolving the issues was demonstrated. Implementation of RoboColumn studies, combined with experimental designs guided by predictor plate screening results, enabled efficient optimization of step elution conditions, leading to improved yield and impurity clearance.

Downstream-Specific Challenges:

• Variability in impurity profiles during scale-up (e.g., Host Cell Proteins, DNA, and polysorbate degradation).
• Managing hold-up volumes and flow inconsistencies in large-scale downstream operations.
• Addressing precipitate formation during purification steps (e.g., low pH and high DNA content).

Process Robustness:

• Strategies for ensuring consistent column loading and resin performance during tech transfer.
• Evaluating operational ranges for downstream unit operations like chromatography, filtration, and ultrafiltration/diafiltration (UFDF).
• Mitigating risks associated with resin aging and membrane degradation during scale-up.

Best Practices for Tech Transfer:

• Pooling material from bioreactor runs to account for variability in impurity profiles.
• Leveraging digital twins and PAT tools to monitor downstream processes in real-time.
• Ensuring alignment between bench-scale and large-scale downstream workflows to minimize variability.

Regeneron Pharmaceuticals, Inc. has successfully integrated Raman Spectroscopy as a Process Analytical Technology (PAT) tool to enhance the robustness of unit operations including Ultrafiltration/Diafiltration (UF/DF) during the purification of protein-based therapeutics. This innovative approach enables real-time, in-line monitoring of critical quality attributes through universal chemometric models, streamlining the transition from process development to GMP implementation.

• Material Selection and Optimization
• Process Intensification
• Automation and Digitalization

This presentation introduces a digital twin framework for robust mAb purification that integrates mechanistic modeling, online HPLC, and real-time automated control. By coupling real-time analytical feedback with calibrated simulations, we demonstrate proactive management of process variability. Case studies show that even under extreme conditions, consistent product quality is maintained, supporting the implementation of model-based control strategies in downstream development.

• What are the key challenges in scaling downstream processes?
• How can process robustness be ensured during scale-up?
• What role does digitalization play in scaling downstream processes?

Host cell protein (HCP) persistence poses major purification challenges in biomanufacturing, affecting product quality, safety, and compliance. Recent mechanistic insights explain why some HCPs resist conventional removal, enabling a shift from reactive control to proactive, design-based mitigation. This review summarizes current knowledge, highlights implications for upstream and downstream process development, and outlines integrated, science-driven strategies for improved HCP risk management.

Due to their platformability, safety, and efficacy, messenger RNA (mRNA) technologies have shown promise for use in prophylactic vaccines and therapeutic medicines. However, chromatographic methods to purify mRNA products are challenging due to their large size and invariable chemical structure, which limit the effectiveness of diffusive resin-based approaches and result in co-purification of molecularly similar, but shorter RNA byproducts. Therapeutically relevant mRNA often contains a 3′-polyadenine tail that promotes in vivo stability and can be utilized in affinity chromatography with complementary oligo-deoxythymidine (OdT) ligands coupled to either resin beads or monoliths. However, factors such as product length and sequence can have an impact on chromatographic performance. Furthermore, identifying optimal process parameters for each target molecule requires the ability to quickly and reliably screen a wide range of chromatography conditions. Currently, high-throughput (HT) technologies to screen OdT-based conditions utilize multi-well plates containing beaded resin media or monoliths, but such batch chromatography plates do not provide insights on the impact of flow dynamics on chromatographic separations. In this study, we evaluate a new technology from Sartorius, which utilizes a small volume convective monolith chromatography media technology (CIMOcta®) under flow to perform HT chromatography experiments with a liquid handler. RNA size and process parameters including residence time, feed concentration, and salt concentration were evaluated to assess their impact on dynamic binding capacity, product yield and integrity. Best screening conditions were then validated for scalability utilizing 1 mL CIMmultus OdT monoliths. The results of this work demonstrate how CIMOcta® offers a scalable platform for HT screening studies, which consider the impact of flow dynamics. Hence, this technology is a valuable tool for developing affinity-based purification processes of mRNA targets in a rapid and cost-effective fashion.

Downstream biologics process development generates complex, heterogeneous data that are often stored without sufficient context to enable reuse, integration, or advanced analysis. Project Lexy addresses this challenge by establishing a structured, context-driven data mapping framework that links process metadata, unit operations, and parameter-level information across downstream workflows. A standardized schema was developed to capture molecule identity, process intent, unit operation sequencing, and detailed parameter metadata, including control type, targets, limits, and units. This approach preserves both experimental values and their operational context. Mapped data are persisted in a relational database to support structured storage and querying, while a complementary graph-based data model captures relationships between molecules, processes, and parameters to enable contextual navigation and cross-process analysis. A lightweight frontend enforces schema-compliant data entry and supports consistent data ingestion with minimal user burden when paired with an integrated backend database. Together, these components transform fragmented downstream data into a reusable, extensible knowledge asset. Project Lexy demonstrates how deliberate data mapping and contextual modeling can support scalable process understanding and lay the foundation for advanced analytics and decision support in biologics development.

• What are the emerging technologies for impurity removal?
• How can impurity clearance be optimized for complex biologics?
• What role does process modeling play in impurity management?

As biologics and advanced therapies grow in complexity, traditional analytical paradigms are no longer sufficient. This presentation explores how AI-enabled approaches are transforming biopharmaceutical characterization—integrating multi-attribute methods (MAM), advanced analytics, and real-time data interpretation. We will discuss how digital and AI-driven frameworks are enabling faster decision-making, improving product understanding, and supporting regulatory convergence across the therapeutic lifecycle.

Process Analytical Technology (PAT) enables real-time and near real-time decisions in biopharmaceutical manufacturing. This presentation introduces a structured PAT roadmap, demonstrated via an at-line protein analyzer case study. It addresses technical, regulatory, and business aspects, emphasizing value proposition, method development, manufacturing feasibility assessment, and change management helping determine PAT's role as an informational or decision-support tool

• Quantifying avoided deviations and comparability work
• Linking digital tools to PPQ success
• Funding digital beyond pilots

• When you look back, what was your biggest misconception about digital biomanufacturing when you started?
• Why do digital pilots look successful but not make it to routine manufacturing?
• Everyone talks about data-driven manufacturing – What data actually matters when a batch is at risk?
• What capabilities had to exist in manufacturing before digital tools started delivering value?
• What one capability should every biologics site build in the next 2-3 years?
• If you were in the audience – what would you be trying to learn this week?

Seventy percent of digital transformation initiatives in regulated manufacturing fail to achieve their stated objectives, yet the conversation remains focused on tool selection and AI deployment. This presentation argues that the root cause is architectural, not technological: manufacturing intent, execution context, and quality evidence are lost at every lifecycle handoff — from process development to manufacturing, from site to site, from internal to CDMO. Drawing on work within the Pistoia Alliance CMC Process Ontology project and the BioPhorum Data Enablement Program, the talk introduces the concept of an Intent-Driven Digital Core — an architectural framework that preserves meaning, context, and evidence across the product lifecycle. It then demonstrates how this foundation changes what is possible: technology transfer timelines compress because receiving sites instantiate rather than reinterpret processes; batch disposition shifts from retrospective reconstruction to continuous evidence evaluation; and AI models earn regulatory trust because the data they consume has preserved semantics and provenance. The presentation closes with practical implications for MSAT and manufacturing leaders: what to invest in before investing in analytics, and how to recognize when your architecture is the bottleneck.

Digital twins are transforming biomanufacturing from development to production. In a joint session, six industrial showcases address real-world challenges: accelerated biosimilar development, predictive media optimization, smarter scale-up, UF/DF process design, advanced viral vector manufacturing, and integrated end-to-end process control. The presentation guides the audience from experimental planning and execution through automated raw data preprocessing to model-based decision-making—demonstrating how digital twins enable faster development, reduced risk, and more intelligent, data-driven manufacturing strategies.

How digital tools change tech transfer processes?

• Processes that didn’t benefit from digitalisation
• Where manual expertise still wins
• Risks of over-automation · Decision framework

• Data and PAT requirements
• Regulatory hurdles and successes
• Lessons learned from early adopters

1. AI-Driven Bioprocess Optimization:

o How can AI models evolve to support real-time prescriptive process adjustments in biologics manufacturing?

o What are the unique challenges of applying AI to complex biologics processes, such as upstream and downstream bioprocessing?

2. Technological Foundations for Autonomous Bioprocessing:

o What advancements in data integration, process analytics, and control systems are necessary to enable autonomous manufacturing systems?

o How can bioprocessing facilities leverage digital twins, predictive modeling, and advanced sensors to achieve this vision?

3. Scalability and Flexibility in Autonomous Systems:

o How can facilities ensure that autonomous systems are scalable and adaptable to meet the demands of both clinical and commercial biologics production?

o What role does modular facility design play in supporting the transition to autonomous bioprocessing?

4. Overcoming Barriers to Adoption:

o What are the cultural and technical barriers to implementing autonomous systems in bioprocessing facilities?

o How can organizations foster cross-functional collaboration between process scientists, engineers, and IT teams to drive adoption?

5. Regulatory and Quality Considerations:

o How can autonomous systems align with regulatory requirements for biologics manufacturing, ensuring compliance and product quality?

o What role do AI and automation play in enhancing process robustness and reducing variability in biologics production?

6. The Future of Bioprocessing Facilities:

o What does the next generation of biologics manufacturing facilities look like with autonomous bioprocessing at the core?

o How can facilities balance innovation with operational efficiency to remain competitive in a rapidly evolving industry?

Models, digital shadows, digital twins are receiving increased attention from industry and regulatory bodies. But is the maturity level still rather academic? What are the hurdles to get them deployed in real business cases for biologics to create impact? This contribution demonstrates the potentials of the digital twins along the product life cycle with industrual case studies:

• Being quicker in process development and chracterization

• Being more predictive in tech transfer and validation

• Allow for process robustness and optimization in manufacturing using digital twin based (predictive) control

• Real time release using end to end solutions

• What specific examples demonstrate how AI-driven modeling and simulation have successfully reduced timelines and improved outcomes in upstream process development?
• What real-world case studies highlight how predictive modeling has uncovered actionable insights that traditional methods failed to identify?

Tech transfer and new product introduction (TT/NPI) teams often run robust risk processes, but the underlying data is still fragmented across disconnected tools, making it hard to reuse prior learnings, coordinate mitigations across sites/functions, and detect issues early enough to protect timelines and supply. This presentation shares a practical blueprint for an Integrated Risk Tracking & Scenario Simulation Platform that modernizes TT/NPI risk management without changing the underlying SOP-governed process. The approach: (1) standardize and digitize risk capture so entries are consistent and lifecycle-managed, (2) connect risk registers across programs/sites and enrich them with operational signals, and (3) layer AI to turn connected risk data into role-based, action-oriented insights. A centrepiece of the vision is an AI agent designed to surface potential risks early—before they materialize—by combining historical risk patterns with current signals, and prompting teams with targeted questions, likely failure modes, and recommended mitigations drawn from prior programs. We will walk through high-impact applications enabled by this architecture, including a pre-TT/NPI risk scan, recommended mitigations (owners/timing informed by prior outcomes), and exception-based alerts driven by continuous monitoring of evolving risk signals. Attendees will leave with a repeatable implementation pattern for moving from static registers to proactive, AI-supported risk monitoring—helping teams reduce rework, improve right-first-time execution, and strengthen governance decisions across the bioprocessing lifecycle.

• Where complex formats break traditional assumptions?
• What manufacturability data CMC wishes they had earlier?
• Cost and timeline impact of late fixes
• Examples of re-design versus rescu

Critical quality attributes (CQAs) are the scientific cornerstone of risk-based CMC strategies, directly linking molecular characteristics to clinical safety and efficacy. For monoclonal antibodies, CQA frameworks are well-established — encompassing glycosylation, charge variants, aggregation, and post-translational modifications — with robust analytical platforms and regulatory precedent under ICH Q8/Q9/Q11 and Q6B. The landscape is rapidly evolving. Bispecific antibodies, ADCs, mRNA therapeutics, and cell and gene therapies introduce structural complexity that challenges conventional CQA paradigms, demanding orthogonal analytical strategies, deeper process understanding, and earlier risk assessment. This presentation will walk through the systematic process of CQA identification and risk ranking — from initial risk assessments and criticality scoring to control strategy development — before examining how these approaches must adapt across modalities. Key themes include leveraging mAb platform knowledge to accelerate newer modality development, integrating advanced analytics (multi-attribute monitoring, HOS characterization, real-time release testing) to drive both quality and efficiency, and anticipating regulatory expectations where clinical correlates for novel CQAs remain emerging. A fit-for-purpose, modality-informed CQA framework built on mechanistic understanding and analytical rigor is essential to delivering safe, efficacious biologics at pace.

• Assays that correlate with aggregation, stability and heterogeneity?
• How to build a minimal, decision enabling assay panel for complex proteins?

Primary containers, such as glass vials, play a critical role in ensuring the stability, sterility, and proper administration of a drug product (DP). However, challenges such as glass flake formation and vial breakage can compromise product quality and render the product unusable in the clinic. To mitigate these risks, manufacturers have introduced innovations in glass composition and surface coatings. A workflow using simple analytical techniques was developed to evaluate the performance of several specialty vials across buffer systems and processing conditions. Overall, the results showed limited performance differences between specialty vials and standard glass vials. The streamlined workflow proved effective for assessing vial robustness and can be extended to evaluate other primary container types or handling conditions. Opportunities for further refinement of the workflow will also be discussed

• What are the key strategies for implementing continuous biomanufacturing in commercial manufacturing?
• How can upstream and downstream operations be effectively integrated for seamless processing?
• What metrics define success when transitioning from fed-batch to continuous processes?
• What are the expectations for ICH Q13 and what do you need to define upfront.
• Examples of changes in documentation and operations when moving from batch to hybrid/continuous.

• Strategies to control aggregates without destroying yield?
• Trade-offs between purity, recovery and robustness
• Regulatory considerations when “perfect” is not achievable

Repeated validation cycles create cost and time burdens, raising questions about regulatory flexibility and risk-based approaches.

• Current FDA and EMA expectations for process validation
• Re-validation for well-established, industry-standard processes
• Balancing robustness and flexibility
• Any harmonization strategies?

Developing a complex molecule like bispecific antibody is very challenging due to its asymmetric nature. Thorough and effective development strategies are vital to the success of the bispecific antibody development. This presentation will focus on the various development strategies critical to advance bispecific molecules from biological design to clinical development, including case studies on bispecific technology selection, drug developability assessment, purification and analytical development to address mispaired species issue to ensure product quality and supply clinical trials.

Complex molecules can fall outside standard downstream platform processes when molecule-specific attributes disrupt individual unit operations. This presentation highlights common reasons for platform mismatch and presents a practical approach to diagnosing underlying gaps and selecting effective solutions. Real case examples will illustrate root-cause identification, the modifications implemented, and the outcomes achieved—providing a practical toolbox for tailoring downstream processes for non-platform molecules.

• Which analytics predict PPQ risk?
• Early warning signals MSAT can act on?
• Real examples of avoided delays

• What approaches can ensure scalability and robustness of upstream processes to meet the demands of large-scale production?
• How can continuous manufacturing technologies be integrated across upstream and downstream operations to enhance efficiency and reduce costs?
• What role does regulatory guidance (e.g., FDA and ICHQ13) play in designing future-proof manufacturing processes?
• How can advanced modeling tools, such as digital twins and AI, be leveraged to predict and optimize large-scale production outcomes
• Which new analytical technologies are worth the CMC risk?
• What would you design differently today?

• What are early signals that the molecule is going to be a problem?
• Which decision would you reverse if you could?
• What data was mossing – and how can this be generated next time?
• What advice would you give a team starting a complex molecule today?

• Review of the main streams and initial status
• Broad requirements for Tox and for First in Human Studies
• Cell banking
• Other Raw materials
• Drug Substance Process and Manufacture
• Minimum requirements
• Understanding your process :
• Impurities : identity, clearance, control
• First steps towards a control strategy
• Adventitious contamination and Viral Clearance Studies
• Drug formulation and Drug Product Processing
• Analytical package
• Release methods definition and development
• From method performance to method validation
• In Process Controls (else cover under process?)
• Batch data in the submission
• Product Characterisation and Reference standard
• Stability ( DS and DP)
• Forced degradation studies : necessity and importance
• Why is stability important ?
• Different type of stability studies and typical package for PhI
• Shelf life assignment

We demonstrate the first end-to-end concept downstream continuous processing train for AAV vectors at pilot 50 L scale, addressing key bottlenecks in traditional batch processing. The process utilizes novel enabling technologies for each unit operation, including multi-column affinity capture with fast-flow loading, dual-tank low pH viral-inactivation, rapid-cycling anion exchange chromatography using weak partitioning AEX for full-empty separations, and countercurrent hollow fiber filtration for single-pass UFDF. The continuous train was run for 72 hours of continuous operation and reduced resin volumes by >90% while matching or exceeding batch process CQAs. The continuous purification process as demonstrated can be linearly scaled up from 50 L to be compatible with 500-2,000 L scale staggered batch harvests as well as upstream perfusion systems.

With pressure to reduce COGs and time-to-market, process intensification has re-emerged as a strategic priority.

• Practical strategies to intensify processes without increasing risk
• Understanding where intensification truly delivers value
• Insight into future manufacturing models for biologics

• What problems AI is actually good at solving?
• Pattern recognition versus mechanistic understanding
• Why AI does not replace process understanding
• The three types of AI: Descriptive, Predictive, Prescriptive – Where does bioprocessing mostly sit today?
• What AI can realistically deliver in the next 12-24 months? (where it is working, where it is still experimental, where expectations need resetting)

TBC