Bispecific antibodies (BsAbs) are promising therapeutics due to their dual-targeting capability. However, BsAbs present greater purification challenges than traditional monoclonal antibodies, owing to their high format diversity and structural complexity. This presentation will outline distinct considerations in BsAb purification, each of which represent common bottlenecks in BsAb CMC. Case studies will be provided to illustrate strategies used to address these challenges.

Early-stage development of complex biologics is constrained by limited material and aggressive timelines. An integrated mechanistic modelling strategy for the cation-exchange (CIEX) step enabled rapid, data-driven downstream development of a complex multi-specific, advancing to GMP in six months despite evolving upstream conditions. Clone-specific models and targeted simulations defined operating optima and a multidimensional design space, revealing critical pH sensitivity. Modelling insights delivered a 19% increase in yield, enhanced impurity clearance, and reduced risk—achieved faster and at lower cost than traditional DoE, accelerating progress to GMP.

Accelerated Seamless Antibody Purification (ASAP) was developed at Sanofi as a simplified, continuous purification strategy to enhance process efficiency and deliver a direct productivity impact. Initial demonstrations showed that ASAP can reduce mAb purification time by up to threefold. The next step is to transition this approach from a laboratory proof of concept to GMP implementation. Ongoing efforts across Sanofi’s mammalian platform aim to make this innovation a reality for all Phase 1 programs.

Bioseparation is a critical step in biopharmaceutical manufacturing. It employs polymeric adsorbents decorated with affinity ligands to increase the selective capture of target biological products. The EU-funded PURE project aims to develop precisely functionalised adsorbents. The project has brought together an interdisciplinary team of leading scientists from academia and industry with expertise in biology, chemistry, computational modelling, materials science, bioprocess engineering and applied social sciences.

Downstream API production steps frequently suffer from high solvent and resin consumption, as well as yield losses. Internal recycling of product and solvent streams can provide a way to improve these metrics. Understanding which recycling technology is preferred is challenging due to the high-dimensional space of objectives, constraints, and system properties. We propose a model-based methodology to systematically explore this space to enable faster and better decision-making.

The current challenge is to produce antibodies at a COGs > 10 $/g. One approach to achieving this goal is through economies of scale. Another is to implement hybrid processes that integrate alternative capture methods or continuous biomanufacturing with rapid cycling or a combination. The downstream processing design for these three scenarios will be compared, with an evaluation of pros and cons.

We present an approach to maximise volumetric productivity in chromatography. Optimisation was achieved by balancing flow rates with dynamic binding capacity, and by reducing bed height and process volumes. In the second part, we compare Protein A resin attributes across different manufacturing scenarios, calculating their impact on productivity, cycle count, and resin costs. Finally, we propose an outlook on future strategies to further enhance productivity beyond current benchmarks.

Upstream advances now enable titers above 10 g/L, boosting productivity but creating downstream bottlenecks. Larger intermediates exceed tank capacity, and chromatography, filtration, and concentration steps take longer. To address this, we are testing high-capacity resins, flow-through chromatography, and single-pass TFF to reduce volumes and streamline processing. New virus filters with higher membrane loading also enhance throughput, helping downstream operations keep pace with upstream gains for efficient, high-titer biotherapeutic production.

The position of virus filters in the DSP process was evaluated. The currently employed Planova 15N filter was compared to the new Planova filters—i.e., BioEx, S20N, and FG-1. Several (modified) mAbs were selected and a comparison was made of the virus filter step between the usual location halfway through the purification process and after polishing. Process performance was shown to be dependent on mAb type, with the modified, larger mAbs being more challenging to filter. Regardless, viral clearance was excellent, with effective removal observed under all conditions tested. Finally, MockV was evaluated as an indicator for VC studies.

3D printed chromatography columns provide advantages over conventional packed bed resins, in particular, control over the multiscale structure that can be tailored to the needs of the separation. However, possible disparities between design and fabricated columns lessens this benefit. We apply 3D imaging to evaluate and compare printed structures to original CAD files to refine the design process, creating more effective separation capabilities for emergent biological products.