De-risking next-generation bispecific antibodies: a target-centric approach to safety intelligence

The rapid expansion of bispecific and multispecific antibody development is transforming the therapeutic landscape. By simultaneously engaging multiple targets, these modalities offer powerful new ways to modulate disease biology—particularly in oncology and immunology. Yet, with this added complexity comes an equally significant challenge: understanding and managing safety risk early and effectively.

Traditional safety assessment strategies are no longer sufficient to address the unique risks posed by multi-target biologics. Instead, drug developers need a more integrated, target-centric view of safety—one that connects preclinical findings, clinical outcomes, and real-world evidence across the full development lifecycle.

This blog explores how such an approach can help de-risk bispecific antibody programs and accelerate confident decision-making.

The emerging complexity of bispecific antibody safety

Bispecific and multispecific antibodies present unique safety challenges that differ fundamentally from those of single-target therapies.

Several key risks must be considered early in development:

• Target crosslinking and unintended signaling, which can activate pathways not anticipated during design
• Complex target expression profiles, increasing the likelihood of off-tumor or systemic effects
• Incomplete understanding of mechanism of action, particularly when targets interact within broader biological networks
• Organ-specific toxicities, driven by overlapping or combinatorial biology
• Uncertainty in starting dose selection, due to limited precedent or comparable molecules

These challenges underscore a core issue: safety can no longer be evaluated in isolation at the molecule level. Instead, it must be understood in the context of the underlying biology of each target—and critically, how those targets interact.

Moving beyond molecules: The value of a target-centric perspective

A central theme in the presentation is the importance of shifting from a molecule-centric to a target-centric safety strategy.

In practice, this means:

1. Understanding the safety profile of each individual target
   Before combining targets in a bispecific, developers must characterize known adverse events associated with each target independently.
2. Evaluating combinations of targets
   Risks may not simply add—they may amplify or interact in unexpected ways.
3. Mapping safety signals across modalities and programs
   Information from monoclonal antibodies, small molecules and other modalities targeting the same biological pathways can provide valuable context.

This approach becomes particularly powerful when supported by a comprehensive and continuously updated safety intelligence platform.

Integrating safety intelligence across the development lifecycle

Key characteristics include:

• Broad data integration, spanning preclinical, clinical, and post-marketing sources
• Coverage across all drug modalities, enabling cross-comparison between biologics, small molecules, and combination therapies
• Daily updates, ensuring access to the most current safety information
• Expert manual curation, supported by advanced analytics

The scale of this data integration is notable:

• Millions of curated safety alerts
• Tens of thousands of drugs, biologics, and targets
• Coverage of thousands of adverse event endpoints

By consolidating evidence from literature, regulatory documents, clinical trials and real-world data sources, such platforms provide a comprehensive view of safety signals that would otherwise require significant manual effort to assemble.

Case study: Learning from established bispecifics

One of the most practical applications of this approach is benchmarking against known bispecific antibodies.

For example, a CD3/CD19-targeting bispecific can be analyzed by:

• Reviewing the safety profile of CD3 independently, including known immune-related adverse events
• Evaluating safety findings across existing therapies targeting CD3
• Comparing across drugs that engage both CD3 and CD19, where data exists

Using comparative analysis tools, developers can quantify and visualize safety signals across system organ classes, identifying areas of elevated risk such as immune system disorders or neurological effects.

This kind of analysis enables early-stage programs to benefit from the collective experience of prior therapies—helping to anticipate issues before they emerge in the clinic.

Navigating novel target combinations with limited data

The value of a target-centric approach becomes even more evident when examining novel bispecific combinations with little or no direct precedent.

The presentation highlights an example of a CD28/PD-1 bispecific candidate for which no direct safety data exist for the specific combination. In such cases, developers can:

• Examine all available data for CD28-targeting agents
• Analyze safety profiles of PD-1 inhibitors, which are well-characterized in clinical settings
• Expand analysis to include related pathways and interacting targets

Comparative tools allow these datasets to be combined into a single view, enabling:

• Identification of shared adverse event patterns
• Detection of potential additive or synergistic toxicities
• Prioritization of risk areas for further investigation

This approach effectively bridges the gap between known biology and unknown combinations.

Translational insights: Connecting preclinical and clinical data

A persistent challenge in drug development is translating findings from preclinical studies into clinical expectations.

The presentation emphasizes the importance of translational safety intelligence, which connects:

• Experimental assay results
• Animal model data
• Clinical trial outcomes
• Post-marketing evidence

By aligning safety signals across these domains, developers can better understand:

• Which preclinical findings are most predictive of clinical risk
• How toxicity profiles evolve across development stages
• Where additional experimental validation is needed

This integrated view supports more informed decision-making and strengthens confidence in go/no-go decisions.

From data to action: Supporting toxicology and regulatory strategy

Beyond early discovery, comprehensive safety intelligence supports several downstream use cases:

• Toxicology program design, including hazard identification and prioritization
• Regulatory submissions, by enabling robust responses to health authority questions
• Safety monitoring, through continuous tracking of emerging adverse events
• Comparative benchmarking, against competitor products and drug classes

Importantly, this approach also supports modern New Approach Methods (NAMs). Survey data referenced in the presentation indicate that in silico approaches are widely used in nonclinical programs, reinforcing the growing role of data-driven safety prediction.

The future: AI-driven safety insight generation

As the volume and complexity of safety data continue to grow, there is a clear need for more efficient ways to access and interpret it.

The presentation outlines the upcoming integration of an AI assistant within OFF-X, designed to:

• Reduce manual search time
• Accelerate insight generation
• Enable more informed decision-making through conversational queries

This represents a natural evolution—from data aggregation to intelligent, user-guided insight extraction.

A proactive framework for de-risking innovation

The development of bispecific and multispecific antibodies is advancing rapidly—but so too must the strategies used to assess their safety.

A proactive, target-centric framework offers several advantages:

• Early identification of potential safety liabilities
• More accurate translation of preclinical findings to the clinic
• Improved confidence in first-in-human dosing decisions
• Faster, more informed development timelines

Ultimately, the goal is not simply to react to safety signals as they arise, but to anticipate and mitigate risk before it impacts patients or programs.

As multi-target therapeutics become increasingly central to drug development pipelines, integrating comprehensive safety intelligence into decision-making workflows will be essential. The ability to connect data, biology and insight in a single, coherent framework is no longer a value-add—it is a necessity.

Explore how translational safety intelligence can support target‑specific risk assessment and help de‑risk novel therapeutic modalities throughout discovery and development: OFF-X preclinical and clinical safety data | Clarivate

Bispecific antibodies BsAb Clinical trials Pharma R&D Real-World Data Regulatory Research Safety

content_copy Copy link Copied