Jintang He

Understanding the target engagement of covalent inhibitors, such as KRAS G12C inhibitors, is essential for evaluating their mechanisms of action and efficacy, particularly given the KRAS G12C mutation's significant role in cancer progression. We previously developed a method to directly evaluate KRAS G12C target engagement by measuring both free and drug-bound proteins in frozen tissues without the need for control samples. Since frozen tissues may not be readily available, we aimed to extend our method to formalin-fixed, paraffin-embedded (FFPE) tissues, which are the standard clinical specimens due to their superior preservation and long-term storage capabilities. This adaptation required addressing the analytical challenges associated with fixation-induced protein cross-linking and structural complexity. Here, we report an optimized workflow for quantification of both free and drug-bound KRAS G12C in FFPE tissues. The workflow integrates a specialized pretreatment process including deparaffinization, heat-induced antigen retrieval, and protein-level immunoaffinity enrichment with a targeted 2D-LC-MS/MS to enable a precise and reproducible determination of KRAS G12C target engagement in complex FFPE matrices. The method typically requires a minimum of 5 μg of total protein input, making it suitable for clinical applications with limited sample quantities such as core needle biopsies. A direct comparison of KRAS G12C target engagement measured in both FFPE and corresponding frozen tissues from a xenograft mouse study demonstrated a strong correlation between the two tissue formats. Overall, the reported workflow may serve as a sensitive and reliable tool for assessing the KRAS G12C engagement and potentially that of other covalent inhibitors.

KRAS is one of the most frequently mutated oncogenes, with KRAS G12C recently becoming an actionable target for small molecule intervention. GDC-6036 is an investigational KRAS G12C inhibitor that acts by irreversibly binding to the switch II pocket of KRAS G12C when in the inactive GDP-bound state, thereby blocking GTP binding and activation. Assessing target engagement is an essential component of clinical drug development, helping to demonstrate mechanistic activity, guide dose selection, understand pharmacodynamics as it relates to clinical response, and explore resistance. Here, we report the development of an ultra-sensitive approach for assessing KRAS G12C engagement. Immunoaffinity enrichment with a commercially available anti-RAS antibody was combined with a targeted 2D-LC-MS/MS technique to quantify both free and GDC-6036-bound KRAS G12C proteins. A KRAS G12C-positive non-small cell lung cancer xenograft model was dosed with GDC-6036 to assess the feasibility of this assay for analyzing small core needle biopsies. As predicted, dose-dependent KRAS G12C engagement was observed. To date, a sensitivity of 0.08 fmol/μg of total protein has been achieved for both free and GDC-6036-bound KRAS G12C with as little as 4 μg of total protein extracted from human tumor samples. This sub-fmol/μg level of sensitivity provides a powerful potential approach to assess covalent inhibitor target engagement at the site of action using core needle tumor biopsies from clinical studies.

Mass spectrometry has recently emerged as a powerful analytical tool for the assessment of pharmacokinetics and biomarkers in drug development. Compared with ligand binding assays, a major advantage of mass spectrometry-based assays is that they are less dependent on high quality binding reagents, while a key limitation is the relatively lower sensitivity. To address the sensitivity issue, we have developed a generic reagent, ultratargeted two-dimensional liquid chromatography-tandem mass spectrometry (2D-LC-MS/MS) method which combines commercially available protein A affinity capture, targeted analyte isolation by 2D-LC, and targeted detection by multiple reaction monitoring (MRM). A targeted-2D-with-dilution configuration was designed to automate 2D-LC-MS/MS. This method was systematically evaluated using an anti-CD22 monoclonal antibody spiked into monkey and human serum, where lower limits of quantification (LLOQ) of 0.78 and 1.56 ng/mL were achieved, respectively. This represents an over 100-fold improvement in assay sensitivity compared to the conventional LC-MS/MS method. The performance of the method was further confirmed by analyzing another monoclonal antibody, bevacizumab, as well as a soluble antigen, circulating PD-L1. The results indicate that our method enables quantification of antibody therapeutics and antigen biomarkers in both clinical and nonclinical samples in the pg/mL to low ng/mL range. Protein A affinity capture was employed as a universal sample preparation procedure applicable to both full-length antibody therapeutics and antibody–antigen complexes. This novel method is also fully automated and proven to be highly robust for routine bioanalysis in drug development.

Antibody-drug conjugates (ADCs) have complex molecular structures as they are composed of both small and large molecules, and they often undergo biotransformation over time in circulation. Here we describe a high-resolution Orbitrap MS approach for the characterization of ADC biotransformation and stability. Compared with conventional approach by Q-TOF MS, the method described here significantly improved the mass resolution and enabled more comprehensive characterization of ADC catabolites. It is particularly beneficial for characterizing ADC biotransformations with small mass changes.

This work discloses the first examples of antibody-drug conjugates (ADCs) that are constructed from linker-drugs bearing dimeric seco-CBI payloads (duocarmycin analogs). Several homogeneous, CD22-targeting THIOMABTM antibody-drug conjugates (TDCs) containing the dimeric seco-CBI entities are shown to be highly efficacious in the WSU-DLCL2 and BJAB mouse xenograft models. Surprisingly, the seco-CBI-containing conjugates are also observed to undergo significant biotransformation in vivo in mice, rats, and monkeys and thereby form 1:1 adducts with the A1M plasma protein from these species. Variation of both the payload mAb attachment site and length of the linker-drug is shown to alter the rates of adduct formation. Subsequent experiments demonstrated that adduct formation attenuates the in vitro antiproliferation activity of the affected seco-CBI-dimer TDCs, but does not significantly impact the in vivo efficacy of the conjugates. In vitro assays employing phosphatase-treated whole blood suggest that A1M adduct formation is likely to occur if the seco-CBI-dimer TDCs are administered to humans. Importantly, protein adduct formation leads to the underestimation of total antibody (Tab) concentrations using an ELISA assay but does not affect Tab values determined via an orthogonal LC-MS/MS method. Several recommendations regarding bioanalysis of future in vivo studies involving related seco-CBI-containing ADCs are provided based on these collective findings.

Few treatment options are available for acute myeloid leukemia (AML) patients. DCLL9718A is an antibody-drug conjugate that targets C-type lectin-like molecule-1 (CLL-1). This receptor is prevalent on monocytes, neutrophils, and AML blast cells, and unlike CD33, is not expressed on hematopoietic stem cells, thus providing possible hematopoietic recovery. DCLL9718A comprises an anti-CLL-1 IgG1 antibody (MCLL0517A) linked to a pyrrolobenzodiazepine (PBD) dimer payload, via a cleavable disulfide-labile linker. Here, we characterize the in vitro and in vivo stability, the pharmacokinetics (PK) and pharmacodynamics (PD) of DCLL9718A and MCLL0517A in rodents and cynomolgus monkeys. Three key PK analytes were measured in these studies: total antibody, antibody-conjugated PBD dimer and unconjugated PBD dimer. In vitro, DCLL9718A, was stable with most (> 80%) of the PBD dimer payload remaining conjugated to the antibody over 96 hours. This was recapitulated in vivo with antibody-conjugated PBD dimer clearance estimates similar to DCLL9718A total antibody clearance. Both DCLL9718A and MCLL0517A showed linear PK in the non-binding rodent species, and non-linear PK in cynomolgus monkeys, a binding species. The PK data indicated minimal impact of conjugation on the disposition of DCLL9718A total antibody. Finally, in cynomolgus monkey, MCLL0517A showed target engagement at all doses tested (0.5 and 20 mg/kg) as measured by receptor occupancy, and DCLL9718A (at doses of 0.05, 0.1 and 0.2 mg/kg) showed strong PD activity as evidenced by notable reduction in monocytes and neutrophils.

Trastuzumab emtansine (T-DM1) is an antibody-drug conjugate (ADC) designed for the treatment of HER2-positive cancers. T-DM1 is composed of the humanized monoclonal antibody trastuzumab connected to a maytansine derivative cytotoxic drug, via a nonreducible thioether linker at random lysine residues, and therefore has a very complex molecular structure. It was anticipated that T-DM1 undergoes biotransformations in circulation. However, there was limited knowledge on these structural changes due to bioanalytical challenges. Here, we have investigated the in vivo biotransformations of T-DM1 using a high-resolution accurate-mass (HR/AM) mass spectrometry approach. Three types of biotransformations were characterized for T-DM1 in circulation in tumor-bearing mice, including cysteine or glutathione adduct formation via maleimide exchange, loss of maytansinol via ester hydrolysis, as well as addition of H₂O via linker-drug hydrolysis. These results provide new insights into in vivo catabolism of T-DM1.

A novel strategy to attach indole‐containing payloads to antibodies through a carbamate moiety and a self‐immolating, disulfide‐based linker is described. This new strategy was employed to connect a selective estrogen receptor down‐regulator (SERD) to various antibodies in a site‐selective manner. The resulting conjugates displayed potent, antigen‐dependent down‐regulation of estrogen receptor levels in MCF7‐neo/HER2 and MCF7‐hB7H4 cells. They also exhibited similar antigen‐dependent modulation of the estrogen receptor in tumors when administered intravenously to mice bearing MCF7‐neo/HER2 tumor xenografts. The indole‐carbamate moiety present in the new linker was stable in whole blood from various species and also exhibited good in vivo stability properties in mice.

Previous investigations on antibody-drug conjugate (ADC) stability have focused on drug release by linker-deconjugation due to the relatively stable payloads such as maytansines. Recent development of ADCs has been focused on exploring technologies to produce homogeneous ADCs and new classes of payloads to expand the mechanisms of action of the delivered drugs. Certain new ADC payloads could undergo metabolism in circulation while attached to antibodies and thus affect ADC stability, pharmacokinetics, and efficacy and toxicity profiles. Herein, we investigate payload stability specifically and seek general guidelines to address payload metabolism and therefore increase the overall ADC stability. Investigation was performed on various payloads with different functionalities (e.g., PNU-159682 analog, tubulysin, cryptophycin, and taxoid) using different conjugation sites (HC-A118C, LC-K149C, and HC-A140C) on THIOMAB antibodies. We were able to reduce metabolism and inactivation of a broad range of payloads of THIOMAB antibody-drug conjugates by employing optimal conjugation sites (LC-K149C and HC-A140C). Additionally, further payload stability was achieved by optimizing the linkers. Coupling relatively stable sites with optimized linkers provided optimal stability and reduction of payloads metabolism in circulation in vivo.

The incorporation of cysteines into antibodies by mutagenesis allows for the direct conjugation of small molecules to specific sites on the antibody via disulfide bonds. The stability of the disulfide bond linkage between the small molecule and the antibody is highly dependent on the location of the engineered cysteine in either the heavy chain (HC) or the light chain (LC) of the antibody. Here, we explore the basis for this site-dependent stability. We evaluated the in vivo efficacy and pharmacokinetics of five different cysteine mutants of trastuzumab conjugated to a pyrrolobenzodiazepine (PBD) via disulfide bonds. A significant correlation was observed between disulfide stability and efficacy for the conjugates. We hypothesized that the observed site-dependent stability of the disulfide-linked conjugates could be due to differences in the attachment site cysteine thiol pK_a. We measured the cysteine thiol pK_a using isothermal titration calorimetry (ITC) and found that the variants with the highest thiol pK_a (LC K149C and HC A140C) were found to yield the conjugates with the greatest in vivo stability. Guided by homology modeling, we identified several mutations adjacent to LC K149C that reduced the cysteine thiol pK_a and, thus, decreased the in vivo stability of the disulfide-linked PBD conjugated to LC K149C. We also present results suggesting that the high thiol pK_a of LC K149C is responsible for the sustained circulation stability of LC K149C TDCs utilizing a maleimide-based linker. Taken together, our results provide evidence that the site-dependent stability of cys-engineered antibody-drug conjugates may be explained by interactions between the engineered cysteine and the local protein environment that serves to modulate the side-chain thiol pK_a. The influence of cysteine thiol pK_a on stability and efficacy offers a new parameter for the optimization of ADCs that utilize cysteine engineering.