Basic And Clinical Study Of BCMA/CD38 And BCMA/CS1 Bispecific CAR-T Cells In Treatment Of Multiple Myeloma

[Background]Anti-B-cell maturation antigen(BCMA)chimeric antigen receptor(CAR)-T cell therapy has shown significant efficacy in patients with refractory or relapsed multiple myeloma(RRMM),but the duration of remission needs to be improved.BCMA downregulation or loss on MM cells was seen in some relapsed patients.Single-cell sequencing showed that despite BCMA loss in one recurrent patient after BCMA-target CAR-T cell therapy,CD38 and CS1 were still highly expressed on MM cells.[Objective]To enhance the function of CAR-T cells,reduce single-target antigen escape,and provide more treatment options for patients with RRMM,this study was designed to investigate the feasibility,preliminary safety,and efficacy of bispecific CAR-T cells targeting BCMA plus CD3 8 or CS1 in treatment of MM.[Method]1.Construct bispecific CAR-T cells targeting BCMA and CD38 and explore their anti-MM activity at the cellular and animal levels and in the phase I clinical trial.1.1 During the design of bispecific CAR targeting BCMA and CD38,the affinity of antiCD38 single chain fragment variable(scFv)was optimized to reduce its potential ontarget off-tumor toxicity and the high affinity of anti-BCMA scFv was maintained to keep its effective cytotoxicity.BCMA CAR,CD38 CAR,BM38 CAR and 38BM CAR contain 4-1BB as the costimulatory domain.CAR-T cell manufacture process includes plasmid transformation,extraction,identification and expansion culture;lentivirus packaging,concentration,and purification;determination of physical and active titres of lentivirus;transfection of activated T cells;expansion and culture in vitro.1.2 The proliferation curves of four kinds of CAR-T and non-transduced T(NT)cells were drawn by trypan blue counting.The transfection rate,CD4/CD8,differentiation phenotype(CD45RA and CD62L)and surface CD38 expression were identified by flow cytometry.1.3 The cytotoxicity of four kinds of CAR-T and NT cells on MM cell lines MM.Is,U266 and RPMI-8266 with heterogeneous expression of BCMA and CD38,BCMA+CD38K562-BCMA cells,BCMA-CD38+ Raji cells,and BCMA-CD38-K562 cells was detected by using the LDH kit.The effector-target ratio was 1:1,5:1 and 10:1,and the co-culture time was 16 hours.Flow CBA method was used to detect interferon γ release after co-culture of four kinds of CAR-T and NT cells with the six target cells in 1:1 for 16 hours.1.4 B-NDG mice were injected with luciferase-expressing MM.1s cells through the tail vein to build the MM mice models.In vivo tumor growth was evaluated by luciferase live imaging.The survival was graphically represented as Kaplan-Meier curves.1.5 The dockinging mode of bispecific CAR with BCMA and CD38 on MM cell surface was simulated by computer modeling.1.6 The predetermined infused dose was 0.5,1.0,2.0,3.0 and 4.0 × 106 CAR-T cells/kg.The adjusted 3+3 dose climbing design was used to explore the suitable dose.1.7 Cytokine release syndrome(CRS)was evaluated per Lee criteria,and other adverse effects were assessed per CTCAE5.0 criteria.1.8 Efficacy was evaluated according to the 2016 consensus of the International Myeloma Working Group(IMWG).Other outcomes included overall survival(OS),progressionfree survival(PFS)and duration of response(DOR).1.9 Quantitative PCR(qPCR)was used to evaluate the expansion and persistence of CART cells in vivo.2.Investigate the suitable dose,preliminary safety,efficacy,and pharmacokinetics of CS1-BCMA CAR-T cells in RRMM,as well as analyze the phenotypes of infused CAR-T cells and the potential biomarkers related to safety and efficacy.2.1 The predetermined infusion dose was 0.75,1.5 and 3.0 × 106 CAR-T cells/kg.The standard 3+3 dose climbing design was used to explore the suitable dose.2.2 CRS and immune effector cell-associated encephalopathy syndrome(ICANS)were evaluated per ASTCT consensus,and other adverse effects were assessed per CTCAE5.0 criteria.2.3 Efficacy evaluation is as the same as 1.6.2.4 Absolute flow cytometry and digital PCR were used to evaluate the pharmacokinetics of CAR-T cells.2.5 Multiparameter flow cytometry was performed to evaluate the phenotype of infused CAR-T cells,including the expression of CS1.The possible relationships between the phenotype and efficacy and CRS were also analyzed.2.6 BCMA and CS1 can be shed from the surface of MM cells to form soluble BCMA(sBCMA)and soluble CS1(sCS1).sBCMA and sCS1 will be closely monitored longitudinally in patients5 peripheral blood and bone marrow to find the possible biomarkers during CS1-BCMA CAR-T cell therapy.[Results]1.The basic and clinical studies of bispcific CAR-T cells targeting BCMA and CD38 in MM.1.1 The four CARs were stably expressed in activated T cells and had similar in vitro expansion,CD38 expression,CD4/CD8,and differentiation phenotype as activated NT cells.Thus,anti-CD38 scFv with optimized affinity did not mediate fratricide.1.2 The four CAR-T cells killed BCMA+and(or)CD38++ tumor cells at different degrees.The cytotoxicity increased with the increase of the ratio of effect to target.Their killing effects on BCMA-CD38-K562 cells were weak.Among them,anti-CD38 CAR-T cells showed the weakest cytotoxicity,while BM38 CAR-T cells possessed the strongest killing effects.Four kinds of CAR-T cells released different levels of interferon γ in cocultured with six kinds of target cells.There was no difference of interferon γ release in co-cultured with K562 cells.1.3 In the MM.1s mice model,CAR-T cells could effectively killed MM cells.Compared with NT cells,BM38 CAR-T cells displayed the strongest anti-MM effect in vivo.1.4 Computer modeling indicated that BM38 CAR had three classical docking modes with MM cells.In view of the length of BM38 CAR and BCMA and CD38 extracellular domain,CD38 binding might help anti-BCMA scFv bind to BCMA on MM cells,and BCMA binding could reversely enhance CD38 binding.1.5 As of December 24,2019,23 patients received the infusion of BM38 CAR-T cells.The median line of prior treatments was 4.61%of patients were refractory to last treatment,and three patients were given autologous stem cell transplantation.Due to osteolytic injury of MM,52%of patients had an ECOG score greater than 1,74%had high-risk cytogenetic characteristics,and 39%had extramedullary disease(EMD).1.6 87%of patients experienced CRS per Lee criteria,and 65%were grade 1-2.Tocilizumab was administrated to four patients,and glucocorticoid to three patients.Four patients had degraded CRS per ASTCT consensus.No neurotoxicity was observed.1.7 The ORR was 87%,of which 12(52%)patients achieved strict complete response(sCR).Of the 9 patients with EMD,EMD was eliminated completely in 56%of patients and partially in 33%.At a median follow-up of 9.0 months,the median PFS was 17.2 months.The median DOR and OS were not reached,and the 1-year DOR and OS rates were 76%and 93%,respectively.1.8 At the dose of 4.0 ×106 BM38 CAR-T cells/kg,92%of the patients obtained clinical response,62%achieved sCR,and all the 12 responders maintained ongoing remission at a median follow-up of 218 days.The infused dose of BM38 CAR-T cells was not related to the severity of CRS.Therefore,4.0 ×106 BM38 CAR-T cells/kg was the suitable dose for expansion.1.9 The expansion and persistence of BM38 CAR-T cells in vivo were quantified as CAR copies/ug DNA.The peak values and the area in the first 28 days(AUC0-28)of CAR copies/ug DNA after infusion were related to the response and the reponsive depth,and the persistence of CAR-T cells correlated with the duration of remission.2.The phase Ⅰ clinical study of CS1-BCMA CAR-T cells in treatment of RRMM.2.1 As of October 30,2022,16 patients received the infusion of CS1-BCMA CAR-T cells,and two patients died due to disease progression during cell production.Due to the osteolytic damage of MM,50%of patients have an ECOG score greater than 1.56%of patients had high-risk cytogenetic characteristics,and six patients(38%)had EMD.Among them,three patients had only extramedullary plasmacytoma without detected MM cells in the bone marrow.The median lines of prior treatments were 4.44%of patients received autologous stem cell transplantation,13%had BCMA targeted CART cell therapy,and 38%received daratumumab.2.2 Six(38%)patients developed CRS per ASTCT consensus,and one(6%)was grade 3.Two patients were administrated with tocilizumab and six patients received glucocorticoid.ICANS was not observed.The suitable expansion dose was 3.0×106 CAR-T cells/kg.2.3 The ORR of all the 16 patients was 81%.Among them,six(38%)reached sCR.Three patients with only EMD did not respond.At a median follow-up of 246 days,the median OS was not reached.The median PFS was 269 days.The 1-year DOR and OS rates were 56.26%and 62.22%,respectively.Among 13 patients with MM cells in the bone marrow,the ORR was 100%.The sCR rate was 46%.The 1-year PFS and OS rates were 56.26%and 72.73%,respectively.2.4 The expression of BCMA and CD38 remained on EMD after infusion.Flow cytometry analysis showed that the infiltration of CAR-T cells in EMD was less than 1%.Transcriptional sequencing showed that there were significant differences between EMD and peripheral blood samples in metabolism and biosynthesis,mTOR signal pathway,Jak-STAT signal pathway and other pathways.2.5 Digital PCR and flow cytometry for absolute count were consistent in quantitative determination of CAR-T cells in peripheral blood and bone marrow.The peak expansion and AUC0-28 of CAR-T cells were related to the response,and the persistence of CART cells in vivo was related to the responsive depth of patients.2.6 CAR+T retained as similar CS1 expression as CAR-T cells.CAR+T cells have a higher proportion of CD4+T,terminal differentiation T(45RA+62L"),CCR3+CD4+T,regulatory T cells(Tregs),PD1+T,TIM3+T cells,and a lower proportion of CD8+T and effect memory T(45RA"62L")cells than CAR-T cells.The infused cell products have a higher proportion of CD4+CAR+T cells and TIM3+CAR+T cells in the responders.The occurrence of CRS was related to the proportion of CAR-Treg cells in the infused cell products.2.7 Soluble BCMA at baseline in patients’ peripheral blood and bone marrow did not affect the efficacy,but the decline rate of sBCMA in peripheral blood one month after infusion correlated with the efficacy.The sBCMA in responders showed a rapid decline after infusion,decreased and remained within the normal range in patients with sCR,and first decreased rapidly and then rose slowly and exceeded the normal range in responders without achieveing sCR.The sBCMA rose synchronously or in advance and exceeded the normal range in patients with relapse or progression.The sCS1 level were lower than the detection limit in most of samples,and further analysis was not done.[Conclusion]This study successfully constructed two kinds of single-target CAR-T cells and two kinds of bispecific CAR-T cells targeting BCMA and CD38 and explored their anti-MM function at the cellular and animal levels.BM38 CAR-T cells showed the most potent anti-MM activity and were next tested in the phase Ⅰ clinical trial.BM38 CAR-T cells displayed a manageable safety profile,good efficacy,and in vivo persistence in RRMM,especially for patients with EMD.CS1-BCMA CAR-T cells were also safe and effective in the treatment of RRMM,which could provide a therapeutic option for patients after BCMA-targeted treatment failure.Soluble BCMA can be used as a biomarker for efficacy monitoring during CS1-BCMA CAR-T cell therapy.[Future Perspective]The study investigated two kinds of bispecific CAR-T cells in MM and demonstrated that bispecific CAR-T cells targeting BCMA plus CD38 or CS1 were feasible,safe,and effective for the treatment of RRMM.Target escape was not observed in our clinical obsvervation.BM38 and CS1-BCMA CAR-T cells provide more treatment options for patients with RRMM,especially for those who have failed after single-target therapy.Future clinical trials with multi-center,large sample and long-term follow-up are needed to verify our prelimary findings.