PO.IM01.14 · 免疫学
GPRC5D多特异性抗体:针对多发性骨髓瘤的亲和力、跨物种反应性和双/三特异性衔接的工程化设计
GPRC5D multispecific antibodies: Engineering for affinity, cross-species reactivity, and bi/trispecific engagement in multiple myeloma
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
背景:GPRC5D是一种G蛋白偶联受体,表达于多发性骨髓瘤细胞,但在健康组织中大部分缺失,使其成为T细胞衔接多特异性抗体的有吸引力的靶点。预计同时衔接GPRC5D和其他分子(例如BCMA和CD3)以激活T细胞的抗体作为单药疗法可提供优异的疗效。多次跨膜蛋白是有价值的治疗靶点,但由于结构复杂性和高度保守性,通常难以用于抗体发现。利用我们的膜蛋白优化发现平台(MPS),我们生成了一组GPRC5D抗体,并将其工程化为双特异性和三特异性分子。我们还开发了一个平台,对我们的先导抗体进行全面的CDR扫描,以进行亲和力成熟、可开发性和用于临床前测试的跨物种反应性评估。
方法:分离针对GPRC5D的抗体,并构建为GPRC5D×CD3和GPRC5D×BCMA×CD3形式。这些分子的体外和体内效力、细胞因子释放和结合特异性均得到测试。为进一步优化我们的先导物,我们开发并优化了全面CDR扫描的方法,其中6个CDR区域的每个残基被替换为所有其他19种氨基酸变体,生成986个单残基变体。使用ELISA对变体进行单独的表达和结合评估,并评估其针对人和食蟹猴GPRC5D的表达和反应性。使用结构指导和AI/ML指导的设计策略将有益的替换(>200%野生型结合)组合起来进行进一步验证。
结果:我们鉴定出的多特异性抗体在多发性骨髓瘤细胞系上表现出皮摩尔级效力的强效T细胞介导的细胞毒性,细胞杀伤与细胞因子释放之间存在较大的窗口,并在人源化小鼠模型中对多发性骨髓瘤异种移植具有稳健的肿瘤控制。抗体特异性分析证实了对预期靶点的高特异性结合,在整个膜蛋白组中未检测到脱靶相互作用。对先导抗体GPRC5D结合部分的CDR区域进行饱和突变,鉴定出14个对人GPRC5D结合升高>200% WT的变体,以及10个对食蟹猴GPRC5D结合升高的变体。组合产生了20多个结合>4,000% WT的变体。
结论:GPRC5D多特异性抗体作为多发性骨髓瘤强效且安全的疗法具有前景。通过CDR扫描,我们正在推进新一代具有增强特性的GPRC5D抗体,包括亲和力提高、pH敏感性、多反应性降低、半衰期延长以及用于临床前开发的食蟹猴交叉反应性。
查看英文原文 English abstract
Background: GPRC5D is a G protein-coupled receptor expressed on multiple myeloma cells but largely absent from healthy tissues, making it an attractive target for T cell-engaging multispecific antibodies. Antibodies that simultaneously engage GPRC5D and other molecules (e.g. BCMA and CD3) to activate T cells are anticipated to provide excellent efficacy as monotherapies. Multipass membrane proteins are valuable therapeutic targets but often inaccessible for antibody discovery due to structural complexity and high conservation. Using our membrane protein-optimized discovery platform (MPS), we generated a panel of GPRC5D antibodies and engineered them as bi- and trispecific molecules. We also developed a platform for comprehensive CDR-Scanning of our lead antibody for affinity maturation, developability, and cross-species reactivity for preclinical testing.
Methods: Antibodies against GPRC5D were isolated and formatted as GPRC5D×CD3 and GPRC5D×BCMA×CD3. These molecules were tested for in vitro and in vivo potency, cytokine release, and binding specificity. To further optimize our lead, we developed and optimized method for comprehensive CDR-Scanning where each residue across 6 CDR regions was substituted to all 19 other amino acid variants, generating 986 single residue variants. Variants were individually evaluated for expression and binding using ELISA and assessed for expression and reactivity against both human and cynomolgus GPRC5D. Beneficial substitutions (>200% wild-type binding) were combined using structure-guided and AI/ML-informed design strategies for further validation.
Results: We identified multispecific antibodies that exhibited potent T cell-mediated cytotoxicity with picomolar potency on multiple myeloma cell lines, a large window between cell killing and cytokine release and robust tumor control against multiple myeloma xenografts in a humanized mouse model. Antibody specificity profiling confirmed high specificity binding to the intended targets, without detectable off-target interactions across the membrane proteome. Saturating mutagenesis of the CDR regions of the GPRC5D-binding moiety of the lead antibody identified 14 variants with elevated binding to the human GPRC5D >200% WT, and 10 variants with elevated binding to cynomolgus GPRC5D. Combination yielded more than 20 variants with binding >4,000% WT.
Conclusions: GPRC5D multispecific antibodies hold promise as potent and safe therapeutics for multiple myeloma. Through CDR-Scanning, we are advancing a new generation of GPRC5D antibodies with enhanced attributes, including affinity improvements, pH sensitivity, reduced polyreactivity, improved half-life, and cross-reactivity in cynomolgus for preclinical development.
利益披露 Disclosure
H. Roth,
Integral Molecular Employment.
T. Barnes,
Integral Molecular Employment.
D. Rogers,
Integral Molecular Employment.
I. Sanchez,
Integral Molecular Employment.
L. J. Stafford,
Integral Molecular Employment.
V. Firers,
Integral Molecular Employment.
B. Tyrell,
Integral Molecular Employment.
A. M. Snyder,
Integral Molecular Employment.
K. Doolan,
Integral Molecular Employment.
B. J. Doranz,
Integral Molecular Employment.
R. Chambers,
Integral Molecular Employment.
J. B. Rucker,
Integral Molecular Employment.