Reply to Dimitrov et al.: VelociSuite technologies are a foundation for rapid therapeutic antibody development

We appreciate the interest of Dimitrov et al. (1) in our work and welcome the opportunity to respond. The authors conclude that our report provides “No evidence for a superior platform to develop therapeutic antibodies rapidly…” and focus their points on our lead antibodies and mouse model developed using VelocImmune and VelociGene technologies. However, as indicated in our Significance statement (2), a platform approach should not only be based on rapid lead antibody discovery (for which several rapid methods exist), but should encompass the entire development process from discovery and preclinical validation through clinical material production. To this end, our report addresses all three components: (i) rapid discovery of potent lead antibodies; (ii) rapid generation of a small animal model; and (iii) antibody isolation compatible with rapid production for clinical use. Our antibody development process links “direct isolation of antibodies from B cells to development of isogenic cell lines,” which “can immediately be used for production of clinical-grade antibody material.” Given that the industry standard for manufacturing cell line development is 6–9 mo, we demonstrated gram quantity production of purified material weeks after lead selection. Such rapid identification, in vivo testing, and scale-up capacity are critical for a timely response to urgent public health threats.

The VelocImmune approach to rapid discovery of potent, fully human antibodies is a cornerstone of our platform. VelocImmune technology and the advantages of in vivo selection have been described (3, 4) and are properly referenced. Our antibodies are fully human in that they contain human variable and human constant regions. The fact that our antibodies are not down-selected on human tissue does not mean they are not fully human. These antibodies are essentially indistinguishable from naturally occurring human antibodies, supported by the fact that hundreds of thousands of doses of VelocImmune-derived antibodies have been safely administered to humans.

Our antibodies were compared with all Middle East respiratory syndrome (MERS)-neutralizing antibodies with publically available sequences. All antibodies were produced using the same methods to enable direct comparison of their inherent properties, avoiding potentially confounding data due to differences in glycosylation, purification, etc. that could occur by obtaining antibodies from other sources. Interest in standardizing antibody research in this manner is growing (5).

In our report, we describe the utility of our VelociGene DPP4 humanized mice for MERS coronavirus infection in comparison to the two previous mouse models, neither of which accurately recapitulates human disease. In contrast, MERS infection of our humanized DPP4 mice results in interstitial lung infiltration, alveolar thickening, and other manifestations consistent with the radiographic findings of significant lung disease in infected humans. More importantly, we demonstrate how our technology avoids time-consuming breeding to generate animal models: “Using the VelociMouse method, F0 humanized mice fully derived from ES cells were available for analysis within 4 mo.”

Taken together, a superior platform for rapid therapeutic antibody discovery and development must address hurdles spanning from discovery and preclinical validation through clinical material production, and the data described in our report clearly demonstrate that VelocImmune and VelociGene are effective foundations of this rapid response platform.