Screen the best scfv to construct a clinically suitable CART cell vector

Screen the best scfv to construct a clinically suitable CART cell vector Summary.   Researchers screened the scFv phage display library derived from human B cells and identified a set of BCMA-specific clones from which the human CAR was modified.

Although the affinity range for BCMA is very narrow, there are significant differences in CAR T cell expansion between unique scFvs in repeated antigen stimulation experiments. These results were confirmed by screening in a MM xenograft model, in which the survival of CAR in vivo can be prefabricated from repeated antigen stimulation assays alone, eliminating the disease and prolonging survival.

The results of the screening identified a highly effective CAR T cell therapy with characteristics including rapid expansion in the body (>10,000 times, on the 6th day), elimination of the huge tumor burden and long-lasting protection against tumor re-attack.

We have generated the second-generation CAR, and the CAR T cell vector derived from this work is under clinical research.

Introduction

By designing a fully human CAR that integrates scFv derived from human B cells, it is possible to potentially avoid the patient’s anti-CAR immunity. In this article, we report the development of a human scFv-derived CAR targeting BCMA. We describe the strategies and principles used to distinguish CARs containing unique human scFvs, and to determine scFv-mediated amplification differences after repeated antigen stimulation, as a particularly useful detection method in optimizing CAR design. In addition, we have characterized the kinetics of expansion and accumulation of T cells genetically modified with our lead CAR in vivo. The highly active human BCMA targeting CAR construct derived from this work is currently under clinical evaluation.
The specific process is as follows:

01 Identify human anti-BCMA scFv and integrate it into the CAR vector

CAR vector We screened a human B cell-derived scFv phage display library (Figure S1), which contains 6×1010 scFvs with recombinant human BCMA extracellular domain-immunoglobulin G1 (IgG1) Fc fusion protein (BCMA-Fc) , To identify BCMA-specific human scFvs. After DNA sequencing, 57 unique and diverse BCMA-specific clones were identified, which contained light chain and heavy chain CDRs, and each clone covered 6 subfamilies with HCDR3 ranging from 5 to 18 amino acids in length. Flow cytometry analysis confirmed the binding specificity of the unique clone to NIH3T3 murine fibroblast artificial antigen presenting cells (BCMA-aAPC) expressing full-length human BCMA. The combination of 17 clones with human MM cell line was further confirmed by flow cytometry, and a part of these scFvs were directly cloned into the second-generation CAR vector of retroviral plasmid.

Flow cytometry analysis after staining with BCMA-Fc confirmed the expression of CAR on the cell surface of the donor T cells. We consistently achieved similar retroviral transduction efficiency (50%-60%) on the scFv studied (Figure 1A). Most of the scFvs studied for BCMA have similar single-digit nanomolar affinities for BCMA (Table S1).

Figure (1 A) Retroviral transduction efficiency;
image

02 After repeated antigen stimulation between scFv clones, CAR T cells expanded

Like B-cell ALL [1], in the CAR T cell clinical trial of MM [10], the expansion of CAR T cells in patients seems to be related to clinical efficacy. In vivo, CAR T cells need to be expanded under continuous exposure to target antigens to eliminate the large tumor burden. For this reason, we tried to evaluate the in vitro expansion potential of CAR T cells in multiple cycles of antigen stimulation. This repeated antigen stimulation test revealed a substantial difference in the amplification of the new scFv incorporated into our CD28-containing CAR construct, which is different from the scFv clones 125 and 171 incorporated into other scFv clones [BCMA (125) and 171 respectively). Compared to BCMA (171)], they are superior amplicon scFvs. For example, in an experiment using three independent donors, BCMA (171) and BCMA (125) CAR T cells uniquely continued to expand after four stimulations, and expanded between 115 times and 158 times on average ( Compared with any other, p <0.005 other CAR). No other scFvs study continued to extend beyond the fourth stimulus, and the peak extension of the fourth stimulus was limited to between 2 and 30 times (Figure 1B).

To analyze the cytotoxicity, we quantified the ATP-dependent bioluminescence of the target of the OMP2MM cell line transduced with luciferase (ffLuc) after 4 hours of co-cultivation with CART cells [18]. The cytotoxicity to OPM2-ffLuc cells showed that CART cells incorporated any of the following anti-BCMAscFv to study the lysis of MM cell lines in a dose-dependent manner. However, it cannot distinguish most CARs, and BCMA (183), BCMA (171), BCMA (130) and BCMA (125) all produce statistically equivalent cell-to-target ratios. Only BCMA (137) has lower cell killing ability than other BCMA.

(B) Repeat the antigen stimulation test;

CART cells (CD28 costimulatory domain) containing the indicated scFv were placed on the BCMA-aAPC or CD19-aAPC monolayer. CART cells were counted every 4 days, and the same number of CAR+ T cells were re-plated on a new aAPC monolayer (arrow). CARs containing human anti-BCMA scFv 171 and 125 can show excellent amplification ability when plated on BCMA-aAPC;

(C) Cytotoxicity analysis;

CAR+ T cells transduced with the same construct as 1b were co-cultured with the OPM2 human myeloma cell line at an increased E:T ratio. All scFv lysed OPM2 cells in a dose-dependent manner. CAR T cells spiked with scFv183, 171, 130 and 125 are better than 137;

03 In vivo evaluation of human anti-BCMA CAR confirmed the results of repeated antigen stimulation tests in vitro to distinguish anti-BCMA scFvs

NSG mice were implanted with bone marrow-tropic human MM cell line OPM2.19, and a single injection of donor T cells modified with the same CAR as described above. In order to confirm the predictive value of repeated antigen stimulation tests, only BCMA (125) and BCMA (171) containing CAR T cells can improve the survival rate of mice. Bioluminescence imaging (BLI) showed that eradication of OPM2-ffLuc cells was also restricted to mice treated with BCMA (125) and BCMA (171) CAR T cells (Figure 2B). Although the two scFvs are equivalent in various assays, we chose BCMA (171) to further characterize and evaluate its clinical transcriptional effects.

By injecting OPM2-ffLuc cells for the second time in mice, it was evaluated that human anti-BCMACARs incorporating clones with CD28 or 4-1BB costimulatory domains [171] provided re-attack on endogenous BCMA-expressing MM cell lines The ability to last for protection.

Figure 2. BCMA (171) and BCMA (125) CAR T cells in the MMMA high tumor xenograft model prolonged survival time and eradicated tumors.

The human bone marrow MM cell line OPM2-ffLuc was injected into NSG mice (1×106 OPM2 cells) through the tail vein, and then implanted and expanded for 21 days. Mice were imaged by bioluminescence (BLI); those patients with high tumor burden were randomly assigned to receive no treatment or were treated with (3×106 CAR+) genetically modified donor T cells, these donor T cells The same as the CD28 (SJ25C1) or BCMA-targeted CD28 shown in Figure 1 (A) demonstrated long-term survival in all mice that received a single dose of BCMA-targeted CART cell therapy spiked with scFv 171 or 125. (B) CART cells against BCMA incorporating scFv 171 and 125 uniquely eliminate the high burden of MM determined by BLI.

Figure 3. Regardless of the CAR costimulatory domain OPM2-ffLuc cells, CAR T cells targeting BCMA (171) can provide long-term immunity (1×106) against human MM cell lines and then injected into NSG mice through the tail vein in. On day 4, mice were treated with a single injection of CAR T cells containing CD28 or 4-1BB costimulatory domain (1×106CAR+) targeting CD19 (SJ25C1) or BCMA (171). (A) Compared with CD19 (SJ25C1) CAR T cell treatment, BCMA (171) CAR T cell treatment can significantly prolong survival. The survival rate among mice treated with BCMA (171) CAR T cells with different costimulatory domains was not significant. (B) On day 21, representative ventral and dorsal BLIs showed that the two groups of BCMA (171) CAR T cells eliminated OPM2 MM cells. (C) 98 days after the initial injection of OPM2 (94 days after CAR T cell injection), long-term surviving mice from 4a were challenged again with OMP2 cells. At the same time, OPM2 cells were injected into naive mice with CAR T cells in the control group. The mice were not retreated with CAR T cells. Long-term surviving mice previously treated with BCMA(171) CAR T cells are protected from OPM2 re-attack and continue to have long-term survival. (D) The imaging on the 34th day (130 days in total) after OPM2 re-attack showed that there was a large MM burden in mice that were not injected with CAR T cells, while there was no OPM2 BLI signal in mice that were previously treated (* p < 0.001).

04 BCMA (171) CAR T cells quickly homing in the established MM model, accumulate and eradicate tumors

We designed a bicistronic retroviral vector, which includes the second-generation BCMA(171)/4-1BBz CAR and external Gaussian luciferase (extGLuc) to monitor the distribution and expansion of CAR T cells in the body over time. increase. We demonstrated that extGLuc/CAR T cells accumulate at the site of OPM2-ffLuc tumor within 24 hours after injection. In the first 6 days after injection, the accumulation of ExtGluc/CAR T cells rapidly expanded at this site, thereby eliminating MM xenografts within this time frame. After the antigen is cleared, CAR T cell fluorescence decreases (Figure 4A). On the 6th day after the CAR T cell injection, the BLI of the MM in the femoral region reached its peak. From the 24-hour imaging to the peak signal, the BLI signal increased by an average of 14,610 times (n=6; range increased by 9,593-19,993 times). Compared with CAR T cells targeted by extGLuc CD19 (SJ25C1), the average BLI signal of CAR T cells targeted by extGLuc BCMA (171) reached a peak of 1.5×109 photons/sec, while CAR T cells targeted by EXTGLuc CD19 (SJ25C1) The average BLI signal of T cells was 5.3×105 photons/s (p <0.0001; Figures 4B and 4C).

Figure 4. The kinetics of BCMA-targeted CAR T cell accumulation and tumor elimination. OPM2-ffLuc cells (1 106) were injected into NSG mice through the tail vein to produce high tumor burden within 24 days until a single dose Of people were treated for T cells (3106; day 0).

05 Human scFv is specific to BCMA

The verification of target specificity is crucial for new scFvs with potential clinical applications. An automated flow cytometry cell-cell binding assay was developed in which a population of cells expressing a query molecule (ie, anti-BCMA scFv) is mixed with cells expressing a library of cell surface molecules to identify queries for potential target interactions.

The advantage of this method is that it can present query and library composition targets in the context of mammalian plasma membranes, and has close to natural post-translational modifications and interactions. To this end, we transiently expressed our anti-BCMA scFv sequence in HEK293 using a cell surface display construct containing the cytoplasmic mCherry adjacent to the murine PD-L1 transmembrane domain.

At the same time, we transiently expressed a library consisting of 395 TNFRSF and immunoglobulin superfamily members, expressed as C-terminal GFP fusions. The library only contains targets that are pre-screened for high protein expression and correct membrane localization, and represents a set of functionally important immune targets.

The mixed scFv query and bank cell populations were analyzed by flow cytometry to identify double-positive GFP; mCherry events, representing scFvtarget-mediated cell-cell interactions. In our scFv panel, two different isoforms of scFv 130 and signal regulatory protein (SIRP) beta-1 showed off-target interactions (Figure S2A), while there were no other scFvs screened, including 171 (Figure 5). ) Shows off-target non-target genes.

We further evaluated the non-specific binding potential of our lead scFvs through the co-culture assay of CAR T cells and isolated primary cells from a variety of basic normal tissue types. Using interferon γ (IFN-γ) release as a substitute for CAR participation through scFv, we found that IFN-γ release only occurs when CAR T cells are co-cultured with the positive control human MM cell line, and the signal-to-noise ratio is> 10,000:1 (Figure S2B).

Figure 5. The specificity of BCMA (171) was transiently transfected into different populations of HEK293s cells. The potential scFv/target antigen interaction was identified as a double positive signal by automatic flow cytometry. The peak labeled “B” represents the interaction with the BCMA+ control. Unlike the scFv clone 130 that interacts with the two isoforms of SIRP-B1 (Figure S2A), all other scFvs including clone 171 (shown) only interact with BCMA.

06 Production of bicistronic CAR vectors including EGFRt for clinical transcription

The truncated epidermal growth factor receptor (EGFRt) has previously been developed as a selectable marker and potential elimination gene, 21 in the case of refractory clinical toxicity. We generated a bicistronic construct including human BCMA(171)/4-1BBz CAR separated by a P2A element with EGFRt for clinical translation(Figure 5A). The co-expression of CAR and EGFRt was verified by flow cytometry analysis (Figure 5B). In the presence of cetuximab, the CAR T cells were co-cultured with the natural killer cell line NK-92 (CD16a+), which proved that the EGFR antibody cetuximab promotes the antibody-dependent cytotoxicity of CAR T cells (ADCC ) Ability [23]. CAR T cells expressing EGFRt, co-cultured with NK-92 cells in the presence of cetuximab, can specifically induce ADCC (Figure 5C).

Sum up:

After screening in vitro to eliminate high-volume MM, we found that T cells transduced by CAR combined with the leading scFv (Figure 2) delivered long-lasting protection (Figure 3), and proved that the MM site is powerful in vivo Amplification and accumulation (Figure 4). EGFRt/BCMA(171)-41BBz (Figure 6) CAR T cells are undergoing clinical evaluation in a phase I trial, which allows repeated dosing and will monitor their immune response to CAR (NCT03070327). Other anti-BCMA CAR T cell vectors derived from this work have entered two separate clinical trials, namely the ICTII/IIII multi-institutional study NCT03338972 and NCT03430011.

Figure 6. Generation and validation of EGFRt/BCMA(171)-41BBz containing CAR T cell therapy

(A) A bicistronic vector containing EGFRt and BCMA(171)/4-1BBz separated by a P2A element engineered to be clinically translated. (B) FACS shows that EGFRt and CAR are co-expressed on the transduced donor T cells (quadrant V++). (C) In the presence or absence of EGFRt-specific antibody cetuximab, CART cells were co-cultured with CD16+ NK-92 cells to evaluate the potential elimination function of EGFRt. Only when both EGFRt are contained in the transgene and cetuximab is present in the co-culture, CAR+ T cells have obvious ADCC.

(source:internet, reference only)

Disclaimer of medicaltrend.org