In vivo generation of CAR T cells in the presence of human myeloid cells

Abstract

Pre-clinical humanized mouse models are a powerful tool to evaluate immunotherapies. NSG-SGM3 mice reconstituted with human stem cells (huSGM3) develop pronounced human myeloid cells due to transgenic expression of stem cell factor, granulocyte-macrophage colony-stimulating factor, and interleukin-3 (IL-3) compared with the widely used humanized NSG (huNSG) model. We assessed in vivo generation of CD19-CAR T cells in huSGM3 mice upon single intravenous injection of the T cell-specific lentiviral vectors (LVs) CD4-LV and CD8-LV. While in vivo CAR T cell generation was clearly detectable in individual mice, generation appeared less efficient than previously observed for huNSG mice. Especially for the CD4-LV group, this correlated with increased IL-15 and decreased GM-CSF levels, indicating activation of monocytes and macrophages. Co-culture assays identified macrophages as a potential barrier for gene transfer. Refining CD4-LV and CD8-LV with a less immunogenic surface by using modified packaging cells substantially improved the transduction of lymphocytes in vitro in the presence of macrophages, as well in vivo in huSGM3 mice. Notably, two mice that developed less CAR T cells showed high interferon-α or -β levels before vector injection. Our data emphasize the relevance of innate immune responses for in vivo generation of CAR T cells, which can be overcome by vector surface engineering. Pre-clinical humanized mouse models are a powerful tool to evaluate immunotherapies. NSG-SGM3 mice reconstituted with human stem cells (huSGM3) develop pronounced human myeloid cells due to transgenic expression of stem cell factor, granulocyte-macrophage colony-stimulating factor, and interleukin-3 (IL-3) compared with the widely used humanized NSG (huNSG) model. We assessed in vivo generation of CD19-CAR T cells in huSGM3 mice upon single intravenous injection of the T cell-specific lentiviral vectors (LVs) CD4-LV and CD8-LV. While in vivo CAR T cell generation was clearly detectable in individual mice, generation appeared less efficient than previously observed for huNSG mice. Especially for the CD4-LV group, this correlated with increased IL-15 and decreased GM-CSF levels, indicating activation of monocytes and macrophages. Co-culture assays identified macrophages as a potential barrier for gene transfer. Refining CD4-LV and CD8-LV with a less immunogenic surface by using modified packaging cells substantially improved the transduction of lymphocytes in vitro in the presence of macrophages, as well in vivo in huSGM3 mice. Notably, two mice that developed less CAR T cells showed high interferon-α or -β levels before vector injection. Our data emphasize the relevance of innate immune responses for in vivo generation of CAR T cells, which can be overcome by vector surface engineering. IntroductionDue to their crucial role in adaptive immunity and anti-tumor surveillance, T lymphocytes are prime targets in cancer gene therapy. One outstanding approach is equipping T cells with a tumor antigen-specific chimeric antigen receptor (CAR) facilitating TCR-independent anti-tumoral response through its intracellular signaling domains upon target binding. Meanwhile, their potential as a therapeutic regimen has been most prominently demonstrated by the approval of five CAR T cell products directed against CD19 and B cell maturation antigen, thus covering B cell malignancies and multiple myeloma, as well as many more products in clinical trials.1Neelapu S.S. Locke F.L. Bartlett N.L. Lekakis L.J. Miklos D.B. Jacobson C.A. Braunschweig I. Oluwole O.O. Siddiqi T. Lin Y. et al.Axicabtagene ciloleucel CAR T-cell therapy in refractory large B-cell lymphoma.N. Engl. J. Med. 2017; 377: 2531-2544https://doi.org/10.1056/nejmoa1707447Crossref PubMed Scopus (0) Google Scholar, 2Maude S.L. Laetsch T.W. Buechner J. Rives S. Boyer M. Bittencourt H. Bader P. Verneris M.R. Stefanski H.E. Myers G.D. et al.Tisagenlecleucel in children and young adults with B-cell lymphoblastic leukemia.N. Engl. J. Med. 2018; 378: 439-448https://doi.org/10.1056/nejmoa1709866Crossref PubMed Scopus (0) Google Scholar, 3Wang M. Munoz J. Goy A. Locke F.L. Jacobson C.A. Hill B.T. Timmerman J.M. Holmes H. Jaglowski S. Flinn I.W. et al.KTE-X19 CAR T-cell therapy in relapsed or refractory mantle-cell lymphoma.N. Engl. J. Med. 2020; 382: 1331-1342https://doi.org/10.1056/nejmoa1914347Crossref PubMed Scopus (0) Google Scholar, 4Abramson J.S. Palomba M.L. Gordon L.I. Lunning M.A. Wang M. Arnason J. Mehta A. Purev E. Maloney D.G. Andreadis C. et al.Lisocabtagene maraleucel for patients with relapsed or refractory large B-cell lymphomas (TRANSCEND NHL 001): a multicentre seamless design study.Lancet. 2020; 396: 839-852https://doi.org/10.1016/s0140-6736(20)31366-0Abstract Full Text Full Text PDF PubMed Google Scholar, 5Munshi N.C. Anderson L.D. Shah J. S. Lin Y. A. et in relapsed and refractory multiple Engl. J. Med. PubMed Scopus (0) Google CAR T cell therapy is and which its in the J. M. A. of CAR T and in Med. 2017; PubMed Scopus Google of T cells is most by or lentiviral vectors the with the to their high transduction are for in vivo gene this of the vector is this was by the human the vector surface the of human to improved gene in by M. A. T. A. S. M. I. et lentiviral vectors gene therapy in Med. PubMed Scopus Google is well for cell for T thus a potential for CAR gene M. J. M. I. et of to chimeric antigen receptor T cell therapy by transduction of a single B Med. 2018; PubMed Scopus Google this demonstrated and of a CD19-CAR or T cells using or which T cells the T cell A. S. S. C. et vivo generation of human CD19-CAR T cells in B-cell and of Med. 2018; PubMed Scopus Google S. vivo generation of CAR T cells in human the of the of 2020; Full Text Full Text PDF PubMed Scopus Google for generation of CAR T cells in vivo by single intravenous vector humanized NSG (huNSG) mice. in vivo gene in patients many for in vivo CAR T cell generation be in models that the human immune more the innate immune has a role in the against the of myeloid cells, in macrophages, is of most of the huNSG is the and human of human myeloid cells is in this and human granulocyte-macrophage colony-stimulating and interleukin-3 (IL-3) for of human stem cells myeloid M. and of of Full Text Full Text PDF PubMed Scopus Google has been by in vivo of human and in huNSG mice with a human myeloid cell M. J. of human of cells in humanized of the of of the of Scopus Google human GM-CSF and are in and NSG-SGM3 mouse in human myeloid cell upon T. I. T. T. M. H. S. C. et of a human using human PubMed Scopus Google Scholar, A. T. J. T. Y. S. et and of human innate immune cells in a humanized mouse PubMed Scopus Google Scholar, M. M. is improved in mice human GM-CSF and PubMed Scopus Google Scholar, E. M. A. of human T cells in human stem cell colony-stimulating humanized PubMed Scopus Google is the huSGM3 which is a of the huNSG and in human stem cell factor, a for cell and PubMed Google the of monocytes in this was to be the in CAR T and M. A. M. M. C. P. et and are for and due to CAR T Med. 2018; PubMed Scopus Google this a humanized immune with clinical the huSGM3 mouse for in vivo CAR T cell generation with CD4-LV and CD8-LV. We that CAR T cell generation in the presence of this humanized immune is less efficient compared with the previously huNSG model. in vivo transduction can be improved by the vectors against cell in vivo CAR T cell generation in CD19-CAR T cells in huSGM3 mice of or a of vector mice human before and vector to of the T cells and vector transduction as demonstrated previously S. S. E. activation and in vivo gene through lentiviral 2020; PubMed Scopus Google was assessed for the presence of CAR T cells as well as levels in the T cells, in injection as by of mice and of CD8-LV mice developed detectable CAR T cells was T cells in CD4-LV in for mouse a detectable CAR was and While in mice CAR T cells in and T cell CD8-LV mice developed CAR T cells, the of the of the CD19-CAR T cells was by a of B cells in While for most B cell with high CAR T cell levels, in of mice of the CD4-LV for which pronounced CAR T cells been B cell was for CD8-LV mice and mice. B cell levels the in of two and of B cells in of CAR T cell the CAR T cell in the the of the group, for the of CD4-LV mouse a of B cells was observed for mice. and CD8-LV mice showed B cell levels of cells and two CD8-LV mice a B cell in the and of mice showed a of B cells in the of cells and vector in the the observed for the CD4-LV group, the mice of this the for CAR T cells with to CAR T cell in was in the compared with the of mice with tumor cells in CAR T cell in the T cell in the of mice of the CD8-LV group, of the CD4-LV group, and two of the that CAR T cells, the in and in the of mice, of the CD4-LV group, the cells and in and cell and vector of the CAR gene in the B cell levels in the and are for the of by cell of the CAR was by for the the of the mice are with and of the and in two by with multiple and compared with the of of T as and in the mice compared with the group, was increased in the CD4-LV the with mouse in the CD4-LV group, which showed the CAR T cell a for and and as well as the showed a expression in CD4-LV mice. was to increased compared with the and was increased the the which is to myeloid was increased to and GM-CSF was decreased in CD4-LV mice compared with the be due to GM-CSF by monocytes upon PubMed Scopus Google A. S. human monocytes and PubMed Google in of huSGM3 mice vector using a for the are for mouse with and the and by with multiple and vectors in vitro T cell transduction in the presence of the of human myeloid cells the huNSG and the huSGM3 mouse the in vivo T cell transduction in a in vitro model. We a of human T cells with in vitro macrophages to the of the myeloid and human and macrophages with of macrophages high expression of and with Y. of macrophages using a PubMed Scopus Google the the or the human T cells and human myeloid cells in human and of huNSG mice of the for CAR T cell generation of for CD4-LV and for CD8-LV in the was as with of macrophages the transduction decreased T cell transduction by macrophages is by transduction of T cells with the of macrophages using CD4-LV or CD8-LV in packaging of and in T cells in a with or macrophages. with to in in by with or multiple and the to T cell this of the M. A. T. A. S. M. I. et lentiviral vectors gene therapy in Med. PubMed Scopus Google is through in cells surface and the human of thus vector that are and high of in their M. A. T. A. S. M. I. et lentiviral vectors gene therapy in Med. PubMed Scopus Google M. A. S. M. T. A. J. Holmes et for of lentiviral vectors for in vivo gene Med. 2017; PubMed Scopus Google and showed efficient transduction of the intravenous to and improved compared with M. A. T. A. S. M. I. et lentiviral vectors gene therapy in Med. PubMed Scopus Google and by the modified packaging cell with the for the vectors increased transduction in the their transduction increased for and for in vivo CAR T cell the the in vivo compared and with their upon huSGM3 mice to in vivo CD19-CAR T We two vector to the for vector and was used as before and the vectors for the of to the of and the receptor the CD8-LV in to than the vector in the in the CD8-LV and in this two in which mice of and of CAR T cell in a of T cells in the CD4-LV group, which showed CAR T cell mice, mouse of the single and developed clearly detectable levels of CAR T cells with high for T the to of T cells and to CAR T cells and of five mice the CD4-LV detectable CAR T cells in to T cells and CAR T cells and in the CD8-LV of the vector showed for the presence of CAR T cells in of the of of five mice developed CAR T cells, in T CAR T cells than the CAR T cells in the CD4-LV and the CAR T cells to to CAR T cells and CAR in the in the T cell detectable in the and vivo CAR T cell generation using and huSGM3 mice with the vectors as a single or a was the mice. of T cells in are as of T cell and cells of the the for the of B cells in mice. as in are with and and in the in the of huSGM3 mice before the cells in the of human cells mice showed levels with to cells cell showed a to with the of CAR T cells in the CD4-LV with the of CAR T cells, a of the B cell levels was observed in CD4-LV was for mice with detectable CAR T cells in the for mice with or CAR mice in the CD4-LV group, mice in the and mice in the showed substantially or B cell levels with the CAR T cells levels, the and the of B cell was less pronounced in the this group, the of B cell correlated well with the of CAR T cells mice which the of showed B cell of which mouse a to and mice a to or Notably, B cells in mouse of the group, in of the mice the CD8-LV of the B cell levels in of CAR T cells by the B cell was pronounced in CD4-LV for the the by the for vector compared with mice CD8-LV cell and in and B cell levels in and by of human of the CAR by in T cells the of the the mice are as data with and for and in by with multiple and the single and of showed a B cell which that with the the the vector clearly showed a B cell compared with the in a with the of human T cells CAR gene mice more than compared with mice of the CD4-LV was for the which in more than of mice that the of of the showed transduction in the myeloid cell of with tumor cells with cells mice the a of in CAR T cell CAR T cells with the mouse observed pronounced of T and in for the for the CD8-LV and which are by CAR T cells, increased in of mice of the observed vector was observed for and to and levels the and Notably, this was observed for the CD4-LV and for CD4-LV was a of GM-CSF by and of in huSGM3 of huSGM3 mice with the by for are for as in are in with and of for the a for interferon-α and in mice of the Notably, mice and in the and showed a for compared with the levels before vector with of and levels to mice developed CAR T cell levels in the as a this in vivo CAR T cell generation in the human immune of the huSGM3 mouse using CD4-LV and CD8-LV. We in vivo CAR T cell generation by the vector surface with a less immunogenic target for and T T cell transduction of the vectors in target of the the of CAR T cells and their target mice developed less than detectable CAR T cell most showed a pronounced and of B B cell to be a more than CAR T cell in the by was for CAR T cells in the and for CAR T cells T cell and CAR are potential S. J. S. J. Wang Y. P. et antigen receptor to and showed 2020; Full Text Full Text PDF PubMed Scopus Google B cell levels, and assays of with tumor cells the presence of CAR T cells vector of this is that CAR T cell generation was in huSGM3 mice than in huNSG mice. While data is a single injection of CD4-LV and less huSGM3 mice developed detectable CAR T cells than huNSG mice NSG-SGM3 mouse a human with a more pronounced myeloid than the NSG humanized with stem cells human M. M. is improved in mice human GM-CSF and PubMed Scopus Google E. M. A. of human T cells in human stem cell colony-stimulating humanized PubMed Scopus Google C. S. P. P. M. C. M.A. in humanized of and transgenic mouse PubMed Scopus Google the myeloid to be for the in vivo transduction in huSGM3 mice, in macrophages, which are in in of T cells with of macrophages, T cell transduction for CD4-LV and CD8-LV. in the of the of vector in the the transduction of of in vivo CAR T cell generation in huNSG and huSGM3 of are as two of five assessed of and and for the presence of cell was two of B cell cell et et and and are as two of five assessed of and and for the presence of B cell was two of B cell et S. vivo generation of CAR T cells in human the of the of 2020; Full Text Full Text PDF PubMed Scopus Google et A. S. S. C. et vivo generation of human CD19-CAR T cells in B-cell and of Med. 2018; PubMed Scopus Google and in a the of vectors and in vivo transduction in the huSGM3 the T vectors in the packaging cell modified cell been to in vivo gene of of by through the M. A. T. A. S. M. I. et lentiviral vectors gene therapy in Med. PubMed Scopus Google and with a less immunogenic surface transduction of T cells in the with macrophages in and improved in vivo gene in the huSGM3 and high T cell target was observed for in the of and B cell levels in improved gene for as CAR T cell generation was less efficient in the huSGM3 mouse in than in was most due to a vector and vector to more pronounced and in vivo CAR T cell generation in cell in the that for in vivo CAR T cell generation was with in NSG mice, CAR T cells in of CD4-LV than of S. vivo generation of CAR T cells in human the of the of 2020; Full Text Full Text PDF PubMed Scopus Google this for humanized CAR T cell generation in huSGM3 mice was pronounced for be of to monocytes and macrophages, which to of vector and in vivo gene J. T lymphocytes and monocytes the Google transduction of cells vector is CD4-LV the which is to as in myeloid cells, vector as is the in C. M. M. S. J. of of macrophages by the PubMed Scopus Google is well in with the that showed in the huSGM3 due to which in and monocytes activation and J. T lymphocytes and monocytes the Google A. S. human monocytes and PubMed Google A. in monocytes and PubMed Scopus Google is that CD4-LV a pronounced innate immune response that gene in the huSGM3 model. was in with the observed for mice. IL-15 and GM-CSF immune response of the innate immune to a immune directed against the which has been in patients to immune C. S. M. et cells of in B cell of clinical PubMed Google Scholar, J. A. A. et and of CAR T cells with human domains in patients with B-cell Med. 2020; PubMed Scopus Google Scholar, E. M.A. M. M. M. Shah et of CAR T cells in cancer PubMed Scopus Google observed a of high in the of huSGM3 mice to with CAR T cell generation in vivo of mice in the CD4-LV group, the potential of the myeloid in vivo CAR T cell of the NSG-SGM3 is the of with human is of the high of human cells can be in NSG human and thus are less to be by the innate Wang S. J.S. in of human stem PubMed Scopus Google with human myeloid cells with for in of the as this is a the are with myeloid cells of the which observed in mice, was levels of human and before mice develop pronounced levels of CAR T cells in the most due to immune of the innate immune responses in mice with increased and in the before and to A. E. A. vivo of lentiviral vectors a response that gene and vector PubMed Scopus Google et A. E. A. vivo of lentiviral vectors a response that gene and vector PubMed Scopus Google showed that the response in vivo gene and was improved in mice of and huSGM3 mice to human S. C. J. tumor human cells in humanized mouse 2020; PubMed Scopus Google of and or receptor be before injection to the immune has the of in vivo generation of CAR T cells with T by the of this in huSGM3 mice, which a human immune Our data that this many in of the human immune for in a more human immune for T in vivo gene which be in immune responses against the vectors in vivo CAR T cell generation can be overcome by improved in vivo CAR T cell generation in the huSGM3 mouse and the potential to CAR T cell or T cell gene therapy in with and and for and of CD4-LV and CD8-LV T. S. S. J. E. in vivo of human T lymphocytes in mouse PubMed Scopus Google vector the for a CD19-CAR with a and signaling of the in cells or cells in a before with of packaging of and of of for or of of for the vector in the and through a was the was in of and the vector of the vector by transduction of human or and cell in a of the vector and the of the gene in the or cell was the cells and cell human the by T cells, a using the T cell was cells in T cell of and with human and human IL-15 T cell with and to the cell for macrophages, monocytes with using a and for in with human GM-CSF and cells in with and cells in with and and cells in with and cells and cell transduction of T cells was with the of macrophages in a in with human and human IL-15 Co-culture was with the for vector and was in for vivo CAR T cell in with the of the and the NSG-SGM3 or NSG mice cells and B cell and T cell in the by before the of the and to the vector and to of mice, which to be due to the or with human and before and and vector injection. the huSGM3 in of or of CD8-LV or CD4-LV and CD8-LV the in and of CD4-LV or of CD8-LV or or as was the a of was for their CAR T cell and B cell levels in the of the mice by the and for of and and of by the through a cell by of and for in a was by a cell of and the with and in for for of the used for T cell and was with and or cells of in with human or human with human IL-15 and a two a for was in with cells with human and with mouse used for for CAR CD19 and the was in and or the and data using or of human cells was identified as the that human the human for and the cells for human and cells and in the of to the before was and cells huSGM3 mice using the and the was by using the the and the for data was using and and human and and the and as single for as well as for the in for for and for a was used to the of human cells and vector was as a for human cells the was the of mice by the for by a of the for with a the and with the using the assessed as in the by using or with or multiple was assessed by IntroductionDue to their crucial role in adaptive immunity and anti-tumor surveillance, T lymphocytes are prime targets in cancer gene therapy. One outstanding approach is equipping T cells with a tumor antigen-specific chimeric antigen receptor (CAR) facilitating TCR-independent anti-tumoral response through its intracellular signaling domains upon target binding. Meanwhile, their potential as a therapeutic regimen has been most prominently demonstrated by the approval of five CAR T cell products directed against CD19 and B cell maturation antigen, thus covering B cell malignancies and multiple myeloma, as well as many more products in clinical trials.1Neelapu S.S. Locke F.L. Bartlett N.L. Lekakis L.J. Miklos D.B. Jacobson C.A. Braunschweig I. Oluwole O.O. Siddiqi T. Lin Y. et al.Axicabtagene ciloleucel CAR T-cell therapy in refractory large B-cell lymphoma.N. Engl. J. Med. 2017; 377: 2531-2544https://doi.org/10.1056/nejmoa1707447Crossref PubMed Scopus (0) Google Scholar, 2Maude S.L. Laetsch T.W. Buechner J. Rives S. Boyer M. Bittencourt H. Bader P. Verneris M.R. Stefanski H.E. Myers G.D. et al.Tisagenlecleucel in children and young adults with B-cell lymphoblastic leukemia.N. Engl. J. Med. 2018; 378: 439-448https://doi.org/10.1056/nejmoa1709866Crossref PubMed Scopus (0) Google Scholar, 3Wang M. Munoz J. Goy A. Locke F.L. Jacobson C.A. Hill B.T. Timmerman J.M. Holmes H. Jaglowski S. Flinn I.W. et al.KTE-X19 CAR T-cell therapy in relapsed or refractory mantle-cell lymphoma.N. Engl. J. Med. 2020; 382: 1331-1342https://doi.org/10.1056/nejmoa1914347Crossref PubMed Scopus (0) Google Scholar, 4Abramson J.S. Palomba M.L. Gordon L.I. Lunning M.A. Wang M. Arnason J. Mehta A. Purev E. Maloney D.G. Andreadis C. et al.Lisocabtagene maraleucel for patients with relapsed or refractory large B-cell lymphomas (TRANSCEND NHL 001): a multicentre seamless design study.Lancet. 2020; 396: 839-852https://doi.org/10.1016/s0140-6736(20)31366-0Abstract Full Text Full Text PDF PubMed Google Scholar, 5Munshi N.C. Anderson L.D. Shah J. S. Lin Y. A. et in relapsed and refractory multiple Engl. J. Med. PubMed Scopus (0) Google CAR T cell therapy is and which its in the J. M. A. of CAR T and in Med. 2017; PubMed Scopus Google of T cells is most by or lentiviral vectors the with the to their high transduction are for in vivo gene this of the vector is this was by the human the vector surface the of human to improved gene in by M. A. T. A. S. M. I. et lentiviral vectors gene therapy in Med. PubMed Scopus Google is well for cell for T thus a potential for CAR gene M. J. M. I. et of to chimeric antigen receptor T cell therapy by transduction of a single B Med. 2018; PubMed Scopus Google this demonstrated and of a CD19-CAR or T cells using or which T cells the T cell A. S. S. C. et vivo generation of human CD19-CAR T cells in B-cell and of Med. 2018; PubMed Scopus Google S. vivo generation of CAR T cells in human the of the of 2020; Full Text Full Text PDF PubMed Scopus Google for generation of CAR T cells in vivo by single intravenous vector humanized NSG (huNSG) mice. in vivo gene in patients many for in vivo CAR T cell generation be in models that the human immune more the innate immune has a role in the against the of myeloid cells, in macrophages, is of most of the huNSG is the and human of human myeloid cells is in this and human granulocyte-macrophage colony-stimulating and interleukin-3 (IL-3) for of human stem cells myeloid M. and of of Full Text Full Text PDF PubMed Scopus Google has been by in vivo of human and in huNSG mice with a human myeloid cell M. J. of human of cells in humanized of the of of the of Scopus Google human GM-CSF and are in and NSG-SGM3 mouse in human myeloid cell upon T. I. T. T. M. H. S. C. et of a human using human PubMed Scopus Google Scholar, A. T. J. T. Y. S. et and of human innate immune cells in a humanized mouse PubMed Scopus Google Scholar, M. M. is improved in mice human GM-CSF and PubMed Scopus Google Scholar, E. M. A. of human T cells in human stem cell colony-stimulating humanized PubMed Scopus Google is the huSGM3 which is a of the huNSG and in human stem cell factor, a for cell and PubMed Google the of monocytes in this was to be the in CAR T and M. A. M. M. C. P. et and are for and due to CAR T Med. 2018; PubMed Scopus Google this a humanized immune with clinical the huSGM3 mouse for in vivo CAR T cell generation with CD4-LV and CD8-LV. We that CAR T cell generation in the presence of this humanized immune is less efficient compared with the previously huNSG model. in vivo transduction can be improved by the vectors against

Topics & Concepts