In this article:
This article is part 3 of a series about specificity testing. Be sure to read part 2, about the available tools and when to use them.
Summary
The short answer is no, tissue cross-reactivity (TCR) studies are not sufficient for biotherapeutic specificity testing. Even though TCR is widely used and has long been recommended by the FDA, it has a long and much-discussed list of limitations. Importantly, TCR results correlate poorly with clinical outcomes, and toxicologists don’t often use TCR data to drive drug development decisions. There is a clear need for a shift to newer specificity testing approaches that are quantitative, objective, validated, and better correlated with patient safety outcomes.
Why are conventional specificity testing methods insufficient?
As described in the previous article in this series, What tools are available for specificity testing during drug development?, tissue cross-reactivity (TCR) studies were the first specificity testing method to be required for biotherapeutic IND submissions beginning in the 1980s. TCR was the best specificity testing tool available at the time, and it quickly became the standard. But as most people familiar with TCR will tell you, the technology has inherent limitations. Importantly, its results correlate poorly with patient safety outcomes.
This article breaks down those limitations, presents some information about how toxicologists use (or don’t use) TCR data, and points to several resources where you can learn more about these topics.
What does the literature say about the limitations of TCR studies?
Several published reviews (Cunningham et al., 2021; Li et al., 2020; MacLachlan et al., 2021 ) and case studies (Brennan et al., 2018; Leach et al., 2010) walk through the limitations of using TCR studies for predicting in vivo toxicity and safety.
The following list, pulled from the resources above and our own publications and experiences, summarizes TCR’s most concerning limitations.
- TCR cannot identify target proteins. The primary limitation of TCR studies is that they can identify only binding locations; they cannot reveal the identity of specific target or off-target proteins. That means if TCR reveals unexpected binding patterns, the options for follow-up studies to understand the nature of that binding are severely limited. Moreover, on-target tissue staining may provide false reassurance when a target and off-target are expressed within the same tissue or when the off-target is expressed at low levels.
- In vivo protein expression is highly variable. Protein expression levels vary within and between tissues, between individuals, and over time—and these levels are difficult to quantify. With TCR, there’s no way to determine whether lack of staining is due to lack of antibody-protein interaction or lack of protein expression.
- Processing alters protein conformations. Most TCR tissue samples are fixed or frozen, placed onto glass slides, and processed for staining. These steps can alter proteins’ conformations, potentially leading to false positives and false negatives.
- TCR has high background staining levels. Native Fc receptors and endogenous IgG present throughout human tissues frequently cause high background binding, complicating results interpretation.
- Scoring is subjective. All TCR results are based on qualitative observations. Trained pathologists score IHC staining results based on their observations and interpretations.
- Turnaround is slow. Optimizing the staining protocol and completing TCR studies typically takes 12 weeks or more.
- Secreted proteins are excluded. Secreted proteins can be a source of off-target binding, but they are washed off the tissue samples during processing. Thus, TCR cannot reveal target or off-target binding to secreted proteins.
- Some protein variations remain untested. Tissue samples from three donors are unlikely to include all possible protein variations present in the general population:
- Heterocomplex formation and multimer arrangements are often transient or disease specific.
- Many targets can be expressed as several possible isoforms and have isoform- or disease-specific cellular locations.
- Post-translational modifications are variable and they can be permanent or transient.
- TCR studies are not quantitative. With no quantification, TCR studies cannot be statistically analyzed or tracked for quality control.
- TCR studies have never been validated. There are no published analyses of measures such as reproducibility, variability, false-negatives, false-positives, and sensitivity.
- TCR studies have never been qualified by the FDA. Although TCR studies are accepted, TCR studies have never formally been qualified by the FDA, who would review the available validation data if it existed.
The long list of limitations shows that while TCR can provide useful information about biotherapeutic binding locations, other tools are needed to better detect and understand off-target binders and predict patient safety outcomes.
Do toxicologists trust TCR for determining specificity?
Given its long list of limitations, it shouldn’t be too surprising that toxicologists don’t trust TCR data. Perhaps the most telling indication of this viewpoint is a set of survey results that captures just how little influence TCR studies have on drug development decisions (MacLachlan et al., 2021). In this survey, industry experts, mostly biotechnology and pharmaceutical toxicologists, answered a series of questions about how they perceive the utility and value of TCR studies.
The results indicate that the vast majority of toxicologists believe TCR results are not predictive of in vivo toxicity, and that TCR results are not actually used in practice to make critical decisions. We encourage you to read the paper.
Cell-based protein arrays fill the gaps
TCR studies leave large information—and trust—gaps. These gaps could be filled by an objective, quantitative, and consistent approach that identifies target proteins and enables statistical analysis between studies.
Fortunately, such tools already exist, and they are increasingly being adopted. Cell-based protein arrays, such as the Membrane Proteome Array (MPA), provide essential information beyond what TCR can deliver.
As this series continues, we’ll share more about cell-based protein arrays and the evolving specificity testing landscape, including what it looks like to qualify a New Approach Methodology (NAM) with the FDA, how MPA and TCR compare in performance, and methods for minimizing false positives and false negatives.
Looking ahead
In the next article, we’ll dig into how the MPA addresses TCR limitations, and how it is being qualified through the FDA’s ISTAND program.