Evaluation of Anti-CAR Linker mAbs for CAR T Monitoring after BiTEs/bsAbs and CAR T-Cell Pretreatment

For the monitoring of chimeric antigen receptor (CAR) T-cell therapies, antigen-based CAR detection methods are usually applied. However, for each target-antigen, a separate detection system is required. Furthermore, when monitored CAR T-cells in the blood of patients treated with bispecific antibodies or T-cell engagers (bsAbs/BiTEs) recognize the same antigen, these methods produce false-positive results in clinical diagnostics. Anti-CAR-linker monoclonal antibodies (mAbs) targeting the linker sequence between the variable domains of the antigen binding CAR fragment promise a universal and unbiased CAR detection. To test this, we analyzed clinical specimens of all BCMA- and CD19-targeting CAR T-cell products currently approved for clinical use. We found a highly specific and sensitive CAR detection using anti-CAR-linker mAb in blood cells from patients treated with Ide-cel, Tisa-cel, Axi-cel, Brexu-cel, and Liso-cel. For Ide-cel and Tisa-cel, the sensitivity was significantly lower compared to that for antigen-based CAR detection assays. Strikingly, the specificity of anti-CAR linker mAb was not affected by the simultaneous presence of bispecific blinatumomab or teclistamab for Axi-cel, Brexu-cel, Liso-cel, or Ide-cel, respectively. Cilta-cel (containing a monomeric G4S-CAR linker) could not be detected by anti-CAR linker mAb. In conclusion, anti-CAR-linker mAbs are highly specific and useful for CAR T-cell monitoring but are not universally applicable.

The manufacturing process of autologous CAR T-cell products includes lymphocyte apheresis, lentiviral transfection of the CAR into the T-cells, followed by cell expansion and subsequent infusion into the patient [11].Monitoring the course of CAR T-cell treatment numbers and frequencies of CAR T-cells in individual patients is pivotal for diagnosis [1,12,13].Therefore, antigen-based flow cytometric CAR T-cell detection methods are feasible for clinical routine diagnostics [12,14].These assays recognize the binding of a biotin-labeled recombinant antigen to the antigen-binding site of the CAR, detected by a fluorescent labeled anti-biotin antibody [12,15].
One arm of a BiTE/bsAb binds to a surface antigen of a tumor cell (e.g., CD19 or BCMA), whereas the other arm binds to the CD3 receptor of a T-cell initiating a T-cell receptor-independent killing of cancer cells.As the combination of BiTE/bsAb and CAR T-cells has been demonstrated to improve the effectiveness of CAR T-cell therapy, either as bridging or as combined therapeutic approaches, both are now increasingly used in practice or in clinical trials [19], respectively.
Previously, we have demonstrated that teclistamab, a BCMA×CD3 bsAb [20], interferes with flow cytometry-based BCMA CAR T-cell detection in the blood of MM patients, thus leading to false positive results [12].Teclistamab binds to all T-cells via CD3, irrespective of the presence of a CAR, while the other arm (even in the absence of a CAR) binds the soluble biotinylated BCMA used for CAR detection.More generally, CAR monitoring in the presence of BiTEs/bsAbs is not possible for CARs/BiTEs/bsAbs combinations recognizing the same antigen (e.g., anti-CD19 CARs/CD19×CD3 BiTEs), if antigen-based CAR detection methods are applied [12].
To circumvent this critical diagnostic problem, we here evaluate a new detection strategy using anti-CAR linker monoclonal antibodies (mAbs) [13,21,22], targeting the linker sequence between the variable heavy and variable light domains of the scFv.
To activate the isolated T-lymphocytes, cells were transferred to a culture plate precoated with anti-human CD3 (5 µg/mL) and costimulated with soluble anti-CD28 Ab (2 µg/mL).After 2-3 days, cells were transferred to X-VIVO 10™ media (supplemented with human AB-sera and human IL2) without stimulus.All cell cultures steps were performed at 37 • C, 5% CO 2 .

Bispecific Antibody Treatment
Blood samples from CAR T-cell treated patients (1 mL) or T-cells subcultured from patients' blood (1 × 10 6 cells in 1 mL PBS) were incubated with or without teclistamab (10 µg/mL) for 10 min at RT. Cells were washed twice with 1 mL PBS containing 2% heatinactivated fetal calf serum (FCS; Gibco distributed via Thermo Fisher Scientific, Waltham, MA, USA) and analyzed by flow cytometry.
In assays using blood specimens, 2 mL of BD FACS™ lysing solution (Becton Dickinson, Heidelberg, Germany) or 2 mL BD Pharm Lyse™ lysing buffer (Becton Dickinson, Heidelberg, Germany) was added for 10 min or 15 min, respectively.After washing and fixation, the cells were analyzed using a FACSLyric flow cytometer (Becton Dickinson).BD FACSuite™ software 1.6 (Becton Dickinson) was used for data analysis.
Cultured T-cells were subsequently washed twice and resuspended in 100 µL PBS with 5 µL of 7-AAD.After 10 min at RT, the cells were analyzed using a FACSLyric flow cytometer (Becton Dickinson).BD FACSuite™ software 1.6 (Becton Dickinson) was used for data analysis.
In assays using blood specimens, 2 mL of BD FACS™ lysing solution (Becton Dickinson, Heidelberg, Germany) or 2 mL BD Pharm Lyse™ lysing buffer (Becton Dickinson, Heidelberg, Germany) was added for 10 min or 15 min, respectively.After washing and fixation, the cells were analyzed using a FACSLyric flow cytometer (Becton Dickinson).BD FACSuite™ software 1.6 (Becton Dickinson) was used for data analysis.
Cultured T-cells were subsequently washed twice and resuspended in 100 µL PBS with 5 µL of 7-AAD.After 10 min at RT, the cells were analyzed using a FACSLyric flow cytometer (Becton Dickinson).BD FACSuite™ software 1.6 (Becton Dickinson) was used for data analysis.

Results
Antigen-based flow cytometric CAR T-cell detection methods are most frequently used for clinical routine diagnostics.These methods are based on the binding of a biotin-labeled recombinant ligand to the antigen-binding site of the CAR, detected by a fluorescent labeled anti-biotin antibody.For each CAR recognizing a different target-antigen, a separate biotinlabeled recombinant antigen is required (Figure 1A).Most scFv-based CARs contain either a Whitlow linker (GSTSGSGKPGSGEGSTKG) or a (G 4 S) 3 linker (GGGGSGGGGSGGGGS), targeted by anti-Whitlow mAb or by anti-G 4 S mAb, respectively [13,21,22] (Figure 1B).To test whether these antibodies allow for a universal CAR detection in real life, we performed flow cytometry CAR detection analyses of all approved CD19-and BCMA-targeting CAR T-cell products, using isolated lymphocyte specimens from patients' blood (Figure 2A).
Furthermore, we performed flow cytometry CAR detection analyses of all approved BCMA CAR T-cell products (Ide-cel and Cilta-cel) (Figure 2A-C).As result, anti-BCMA CAR T-cells were also successfully detected by anti-Whitlow mAb from myeloma patients treated with Ide-cel.In comparison to the results obtained with BCMA antigen-based CAR detection assays, the percentage of detected CAR T-cells was reduced (69 ± 10% and 53 ± 10% for biotinylated and Alexa 647-conjugated anti-Whitlow mAb, respectively).However, the detection of Cilta-cel (featuring two single domain antibodies, connected by a single G 4 S linker) was not possible by any CAR linker mAb.
From a technical perspective, CAR detection with Alexa-conjugated anti-G 4 S mAb leads to false positive results in whole blood specimens, if red blood cells are lysed by BD FACS™ Lysing solution (Figure S1).This phenomenon occurs in all analyzed cells, but not with the corresponding Alexa-conjugated isotype, anti-CD3 Ab and anti-CD45 Ab.We speculate that an artificial binding of that antibody by stickiness may have occurred through simultaneous fixation by the lysing buffer in the diagnostic routine protocol.This artifact can be avoided using BD Pharm Lyse™ lysing solution, devoid of fixation reagent (Figure S1).
We next determined whether CAR linker mAb allows for the specific detection o CAR T-cells by the simultaneous presence of bsAbs/BiTE recognizing the same tumor antigen (e.g., CD19×CD3 BiTE blinatumomab and anti-CD19 CAR T-cells) (Figure 3A-C)    Here, we show that the detection of anti-CD19 CAR T-cells with biotinylated anti-Whitlow mAbs in blood specimens of lymphoma patients treated with Brexu-cel is unaffected by the presence or absence of blinatumomab (41% versus 42%, respectively), in contrast to the CD19-based CAR detection assay (artificially 99% versus 48%, respectively) (Figure 3D).Similar results were obtained using Liso-cel and Axi-cel anti-CD19 CAR T-cells and blinatumomab (Figure 3D).Furthermore, the detection of BCMA specific CAR T-cells after prior treatment with BCMA×CD3 bsAb teclistamab is also impaired if BCMA antigenbased CAR detection reagents are used (Figure 3A-C).We show that Ide-cel (anti-BCMA) CAR T-cell detection by biotinylated anti-Whitlow Abs is also unaffected in the presence or absence of teclistamab (7% versus 6%, respectively), in contrast to the BCMA-based CAR detection assay (artificially 100% versus 10%, respectively) (Figure 3D).
Finally, the real-world feasibility for anti-CAR linker antibodies is identical to classical CAR detection, but the costs are considerably lower.In our laboratory, the costs per application are € 40.80 and € 30.90 for the classical CD19-and BCMA-based CAR detection assays, respectively, in contrast to € 23.00 and € 21.83 for the anti-Whitlow and anti-G 4 S linker mAb-based CAR detection assays, respectively.

Discussion
In the domain of cellular immunotherapy, the concurrent administration of bispecific T-cell engagers (BiTEs) or bispecific antibodies (bsAbs) and chimeric antigen receptor (CAR) T-cell therapies targeting the same antigen presents a significant challenge for the accurate monitoring of CAR T-cell populations.The monitoring of a CAR T-cell therapy in cases where BiTEs/bsAbs recognizes the same antigen is not possible by antigen-based CAR detection methods.The BiTEs/bsAbs bind to CD3 on all T-cells irrespective of whether they carry a CAR or not.In this case, antigen-based CAR detection does not only bind to the CAR but also to the antigen-specific free arm of the bound BiTEs/bsAbs.This non-selective binding results in the misidentification of non-CAR-bearing T-cells as CAR-positive, thereby producing false-positive results in clinical therapy monitoring.
For clinical diagnostics, this is a serious problem, as a combination of BiTE/bsAbs and CAR T-cells is employed either as a bridging therapy or as a combined therapeutic approach in clinical practice or clinical trials, respectively [19].We still observed in our clinical routine, diagnostic false-positive results on day 9 after the last teclistamab treatment.
The removal of the BiTEs/bsAbs from the cells before detecting the CAR might solve that problem.However, the in vitro application of reducing agent dithiothreitol [23] failed to completely disrupt teclistamab binding during CAR detection [12].Alternative CARdetection strategies based on Protein L, polyclonal anti-F(ab) fragment Abs (Figure S2), and anti-isotype antibodies are less specific [24], and would also label BiTEs/bsAbs bound to all T-cells in most cases.In addition, anti-idiotype antibodies are only available for the detection of a limited number of CARs [12,24].

Figure 1 .
Figure 1.Schematical representation of CAR T-cell detection with the classical ligand-based assay and anti-CAR linker mAb.(A) A schematical representation of CD19 and BCMA tumor antigens and their targeting CARs are shown.The recombinantly produced and biotinylated ligands and APC-conjugated anti-biotin Abs are used to visualize CAR expression on the surface of T-cells.(B) The general structure of a chimeric antigen receptor (CAR) is depicted as an illustration.The single-chain variable fragment (scFv) of CARs contains either a Whitlow linker (GSTSGSGKPGSGEGSTKG) or a (G 4 S) 3 linker (GGGGSGGGGSGGGGS), targeted by anti-Whitlow mAb or by anti-G 4 S monoclonal antibody (mAb), respectively.Biotinylated CAR-linker mABs and APC-conjugated anti-biotin Abs are used for the universal visualization of CAR expression on the surface of T-cells.

Figure 2 .
Figure 2. Anti-CAR linker mAbs allows for specific and sensitive CAR T monitoring in patients treated with any clinically approved CAR T therapies except Cilta-cel.(A) The structures of BCMAand CD19-specific CARs are depicted in the illustration.CD19 antigen-specific CARs: Brexu-cel, Liso-cel, Axi-cel, and Tisa-cel; BCMA-specific CARs: Ide-cel and Cilta-cel.(B) Flow cytometry-based CAR detection analyses are shown, performed with biotinylated anti-CAR linker mAbs and APCconjugated anti-biotin Abs to all approved BCMA and CD19 CAR T-cell products using isolated

Figure 2 . 12 Figure 3 .
Figure 2. Anti-CAR linker mAbs allows for specific and sensitive CAR T monitoring in patients treated with any clinically approved CAR T therapies except Cilta-cel.(A) The structures of and CD19-specific CARs are depicted in the illustration.CD19 antigen-specific CARs: Brexu-cel, Liso-cel, Axi-cel, and Tisa-cel; BCMA-specific CARs: Ide-cel and Cilta-cel.(B) Flow cytometry-based CAR detection analyses are shown, performed with biotinylated anti-CAR linker mAbs and APCconjugated anti-biotin Abs to all approved BCMA and CD19 CAR T-cell products using isolated lymphocyte specimens from patients' blood (n = 3).(C) Flow cytometry-based CAR detection analyses are shown, performed with Alexa 647-conjugated anti-CAR linker mAbs to all approved BCMA and CD19 CAR T-cell products using isolated lymphocyte specimens from patients' blood (n = 3).The CAR-positive cells were determined from the CD3-positive cells and significances calculated based on BCMA-and CD19-antigen-based CAR detection reagent (Miltenyi = 100%).Statistical analysis was performed using Student's t-test.* p ≤ 0.05%; ** p ≤ 0.01%; *** p ≤ 0.001%.

Figure 3 .
Figure 3. Anti-CAR linker mAbs allows for specific CAR T monitoring in patients pretreated with BiTE/bsAb, targeting the same antigen as the CAR.(A) The structure of the blinatumomab (CD19×CD3 BiTE) and teclistamab (BCMA×CD3 bsAb) are illustrated.(B) Shown is a schematic overview of CAR T-cell detection strategies based on the CD19 or BCMA CAR detection reagent (Miltenyi) that leads to a false positive staining in patients pretreated with BiTE/bsAb targeting the same antigen as the CAR.The structure of CD19-or BCMA-specific CAR and CD3 on the cell surface of a T-cell is depicted as an illustration.(C) Shown is a schematic overview of CAR T-cell