Signal-to-Noise: Why Immunogenicity Testing Needs More Context, Not More Cut Points

Immunogenicity to biologic therapies is, by definition, a biomarker, a measurable biological response to therapeutic exposure. However, industry practice often treats immunogenicity as a uniquely separate scientific area, relying on a standardized three-tier testing paradigm applied broadly across programs, regardless of the context of use.  This one-size-fits-all approach arose from a reasonable desire to prevent rare but severe outcomes. Over time, however, it has hardened into a routine practice that is expensive in time, labor, and cost, and paradoxically can deliver an inferior understanding of clinically relevant immunogenicity.

The three-tier construct was not designed to maximize actionable insight. It was created to efficiently triage samples for titration. Operational cut points and confirmatory steps serve as process tools, not as reflections of biological reality. This framework introduces an artificial binary that can mask the nuanced data required for both scientific understanding and informed program strategy.

Beyond Binary: The Signal-to-Noise Perspective

The traditional three-tier immunogenicity testing paradigm, including screening, confirmation, and titration, has served the industry for decades. It emerged from reasonable intentions: ensure consistency, minimize risk, and establish a framework regulators could evaluate. And for many programs, particularly higher-risk molecules, this comprehensive approach remains appropriate and valuable. Our experience shows that relying on statistical cut points to create binary positive or negative classifications can sometimes obscure the insights sponsors need most.

Cut points weren’t designed to identify clinically relevant responses. They are operational tools, statistical thresholds based on negative control samples, created primarily to triage samples to proceed to the next tier. By classifying everything below these thresholds as negative, we discard the contextual information that helps distinguish background noise from meaningful biological responses.

As a result, incidence rates frequently do not correlate with clinical impact. Teams often address stakeholder concerns about the percentage of ADA-positive cases before determining whether these responses affect drug exposure, efficacy, or safety.

For programs seeking informed decision-making, incidence data without clinical context does not yield clear, actionable insights. This matters because cut point-driven incidence commonly inflates apparent immunogenicity without correlating to impact. A sponsor might see high ADA incidence and immediately raise red flags internally triggering cross-functional meetings, risk assessments, and stakeholder management, only to find that those responses didn’t affect drug exposure or clinical outcomes.

Conversely, clinically meaningful responses can be masked when pooled with large numbers of low-level, cut point-positive but irrelevant responses, causing teams to miss real signals in the noise of statistical artifacts. The problem compounds when we look at how response magnitude itself gets measured.

Understanding Response Magnitude

Titers have been the field’s standard measure of response magnitude for good reason. They provide a familiar, quasi-quantifiable metric with decades of precedent, particularly when we look at vaccine development.

Drug development programs, however, face different questions than vaccine programs. Vaccines are designed to elicit strong immune responses, with success measured by robust antibody production. In therapeutic development, the focus is on characterizing unexpected responses across a spectrum of magnitudes, from low-level background reactivity to moderate responses that may affect exposure, and robust responses that could impact outcomes.

Standard titer schemes using two-fold dilutions reliably distinguish large fold-changes, typically greater than four-fold, and are effective for high-magnitude responses. However, many programs require more granular characterization, such as differentiating modest responses with no PK impact and moderate responses that begin to affect drug exposure or tracking subtle changes over time.

Continuous S/N provides equivalent or even superior information. S/N ratios preserve the full range of assay responses, enabling correlation analyses across the entire spectrum rather than only at the high end.

When Context Drives Strategy

This scientific philosophy guides our practice at Immunologix. We offer a full spectrum of immunogenicity testing approaches, including traditional three-tier strategies when appropriate. We also prioritize beginning each engagement with a context-driven discussion.

Risk levels vary by molecule. For example, a fully human mAb targeting a non-essential protein has different immunogenicity implications than a replacement therapy with potential endogenous cross-reactivity. Therapies dosed at nanogram levels, where modest ADA can impact exposure, require different assessment sensitivity than those dosed at higher levels, where only robust responses are clinically significant.

Examining complete response profiles, such as continuous S/N ratios across timepoints rather than cut point-censored positives, reveals meaningful patterns. True immune responses rise above biological noise, persist over time, and correlate with changes in drug levels or therapeutic effect.

These patterns are visible only when full context is retained, including baseline variability, placebo population data, and the complete distribution of responses that distinguish biological noise from clinically relevant signals.

This underscores the importance of a thoughtful upfront strategy. Some programs benefit from comprehensive three-tier testing with titers for every confirmed positive, while others gain clearer insights from S/N-based approaches that preserve continuous data. Many programs use a combination, starting with traditional paradigms and adapting as data emerges.

The key consideration is not which approach is universally better, but which best serves the specific decision-making needs of the program.

What Thoughtful Strategy Delivers

Regardless of whether a program uses the traditional three-tier testing or S/N-based approach, the goal is to generate data that genuinely informs decisions.

The most valuable immunogenicity datasets, regardless of testing approach, share certain characteristics:

• Clinical correlation: Response measurements linked to PK, PD, safety, or efficacy outcomes, not just statistical thresholds

• Biological context: Understanding of baseline variability and placebo patterns that help distinguish noise from signal

• Individual profile clarity: Subject-level response trajectories that reveal patterns masked in population averages

• Transparent communication: Data presentations that stakeholders and regulators can interpret intuitively

The most important factor is not whether titers or S/N ratios are reported, but whether the immunogenicity data addresses the program’s key questions:

• Are responses affecting drug exposure?

• Is there an impact on efficacy?

• Do safety signals correlate with immunogenicity?

The path to those answers can vary. Sometimes traditional three-tier testing provides adequate data when teams look beyond cut point incidence to examine actual clinical correlations, connecting specific response patterns to PK or efficacy outcomes. Other programs benefit from preserving continuous S/N data from the outset, particularly when granular characterization of lower-magnitude responses is critical to understanding clinical impact. The right approach depends on the program’s specific needs.

The Scientific Partnership Approach

The traditional immunogenicity paradigm was developed in response to significant safety concerns during biotherapeutic development. Decades of experience have clarified both its strengths and its limitations in meeting program needs.

At Immunologix Laboratories, we serve as scientific partners, guiding sponsors through the complexities of immunogenicity assessment. Our team offers both proven, regulatory-supported solutions, such as traditional three-tier testing, and forward-thinking methodologies, including S/N-based approaches, tailored to the unique needs and goals of each program.

Ultimately, the best immunogenicity data, whether generated through traditional or innovative approaches, does more than demonstrate compliance. It builds a genuine understanding of the program and enables confident decision-making.

This is the partnership we aim to provide.