Depending on the year, biologic drugs comprise between one quarter and one half of all approved pharmaceuticals. Small molecule drug discovery has benefited from numerous novel discovery initiatives, for example computational chemistry, x-ray crystallography, combinatorial chemistry, and automated, ultra-high-throughput biochemical assays.
The molecular complexity and difficulty in generating candidate monoclonal antibodies (mAbs) has made the application of high-throughput approaches more difficult for this important class of biopharmaceutical.
Epitope binning, a process through which mAbs that bind to an antigen are grouped according to their affinity to specific epitopes on that antigen, is arguably developers’ best bet for both screening and characterizing antibody libraries. The problem is that standard biochemical binning methods (e.g. ELISA) are slow, and biophysical assays like SPR lack throughput or, when automated at all, are only marginal improvements over ELISA assays.
That is beginning to change, as SPR systems begin to catch up with other assay types in terms of throughput.
Surface plasmon resonance
During target-based mAb discovery, investigators generate hundreds or thousands of antibodies that bind to the disease-modulating antigen. But since typical antigens incorporate so much chemical diversity, not all mAbs that bind do so through the same epitope. Through epitope binning, mAbs with specificity to the same antigen are tested pairwise, against all other antibodies within a given set, in a binding assay. mAbs that mutually interfere by competing for the same epitope are grouped together, or binned.
“Epitope binning narrows the candidate field,” says Christina Burtsoff Asp, global product marketing manager at GE Healthcare Life Sciences. “At the same time it uncovers mAbs that bind to the same antigen but do so through different mechanisms of action.” Having two or more antibodies that bind to target antigens through diverse mechanisms could be helpful in treating diseases like cancers and/or infections that adapt to and eventually overcome pharmacology. “Epitope diversity is therefore also a means for obtaining broader intellectual property (IP) protection and identifying binding pairs for diagnostic testing,” she adds.
GE’s contribution to epitope binning is the Biacore analysis system which, combined with Biacore Insight Epitope Binning Extension software, provides assay setup, data analysis, and visualization. According to the company, the system screens up to 2,300 antibody molecules per day, characterizes 64 molecular interactions simultaneously, and runs for 72 hours unattended.
Next-generation screening and characterization
“Epitope binning screens and characterizes antibodies in one step,” says Chris Silva, vice president of marketing at Carterra. “Studying kinetics, affinity, and epitope relationships in high throughput allows characterization of antibody libraries generated against a specific antigen through B cell cloning, phage display, or hybridoma technologies. Screening large libraries and exploiting their diversity demands greater throughput than is currently generally available.”
Traditionally, discovery groups screen libraries using low-resolution, endpoint ELISA methods, then follow up on a subset of molecules via low-throughput SPR or less-reliable label-free methods. “But understanding epitope diversity along with kinetics and affinity for every clone is impossible in a single, conventional SPR or biolayer interferometry experiment,” Silva adds. “Epitope binning via high-throughput SPR identifies optimal candidates plus a portfolio of backups to support more-comprehensive intellectual property coverage.”
At a recent symposium hosted by Carterra, Guangwei Yang from Boehringer Ingelheim discussed how the Carterra LSA instrument allowed his group to assess both epitope diversity and binding kinetics for a large antibody panel. Yang concluded that, combined with sequence and epitope information, HT-SPR assays “enable the fast and optimized selection of antibody panels for MOA [mechanism of action] studies, while simultaneously providing a crucial evaluation of antibody-generation technologies.”
Similarly, a presentation by researchers from Bristol-Myers Squibb outlined how capturing the full diversity of antibody panels numbering in the hundreds-to-thousands without creating redundancy “is critical to finding the optimal therapeutic candidate and using resources efficiently.” Using epitope binning as the initial screen identifies and captures “the full epitope diversity generated by different transgenic animals and different antibody-generation techniques” while enabling rapid selection of antibody panels for mechanism studies.
What this means, says Silva, is “if you’re not binning or just binning tens of antibodies at a time, you’re not finding out much about your epitope. Using HT-SPR, it’s possible to get the ultimate resolution on epitope coverage, to know an antibody’s mechanism of action, reduce failures in development and in the clinic, and discover and lock down IP that differentiates your antibody from competitors.”
HT-SPR replaces sandwich and competitive ELISA assays requiring the integration of multiple liquid handlers, washers, readers, and microplate handlers. ELISA’s shortcomings include somewhat lower reliability, depending on the assay, compared with SPR and more false positives. Additionally, ELISA often misses low-affinity candidates, and its use of detection tags presents challenges when measuring kinetics or affinity.
Flow cytometry is frequently used for screening projects involving intact cells, but flow methods are generally better-suited to discrete, “yes/no” cellular events than to affinity and kinetics measurements. Flow methods are expensive, stress cells (requiring culturing for downstream steps), and introduce yield and purity issues with rare cells.
Carterra’s LSA platform, based on HT-SPR technology, consumes an extremely low amount of sample and enables epitope coverage of up to 384 x 384 interactions, or the potential to analyze nearly 150,000 sensorgrams of label-free data. Epitope binning based on traditional SPR or BLI platforms is significantly lower-throughput, uses large amounts of sample, and is unsuited for large antibody panels.
Epitope binning informs mAb discovery groups with more information in less time than approaches that rely on simple antigen binding as the primary affinity event. Whether this benefit translates to a higher number of approvals for biologic drugs depends on several factors, however. The possession of kinetics, affinity, and competitive affinity data is one thing; having the fortitude (and finances) to explore epitope bin-mates after a Phase 3 debacle is quite another. And, as we’ve seen with combinatory chemistry and million-compound small molecule libraries, more molecules does not always translate to more drugs.