Abstract
An isolated immune cell, a method for preparing such modified immune cell, a method of treating a living being suffering or at risk of suffering from cancer or non-malignant diseases, an oligonucleotide and a use thereof.
Claims
1. A modified immune cell comprising a modified phosphoinositide 3-kinase (PI3K) pathway compared to a non-modified reference immune cell.
2. The modified immune cell of claim 1, comprising an overactive PI3K pathway compared to a non-modified reference immune cell.
3. The modified immune cell according to claim 1 comprising a modified phosphoinositide 3-kinase (PI3K) having increased activity compared to a non-modified reference PI3K.
4. The modified immune cell according to claim 3, wherein said modified PI3K comprises a point mutation.
5. The modified immune cell according to claim 4, wherein said point mutation in PI3K is at amino acid position 81.
6. The modified immune cell according to claim 5, wherein by said point mutation in PI3K a glutamic acid (E) is replaced by a lysine (K) (PI3K.sup.E81K).
7. The modified immune cell according to claim 1, which is a T cell or a NK cell.
8. The modified immune cell according to claim 1, which is a chimeric antigen receptor (CAR) T or NK cell, which CAR comprises an extracellular antigen-binding domain, a transmembrane domain, and an intracellular signaling domain.
9. The modified immune cell according to claim 8, wherein the intracellular signaling domain comprises a CD3 polypeptide or modifications thereof.
10. The modified immune cell according to claim 8, wherein the intracellular signaling domain comprises a CD137 (4-1BB) or CD 28 costimulatory polypeptide.
11. The modified immune cell according claim 8, wherein the extracellular antigen-binding domain comprises an CD19 binding polypeptide.
12. (canceled)
13. (canceled)
14. A method for preparing a modified immune cell characterized by a modified PI3K pathway compared to a non-modified reference immune cell, comprising (i) providing an immune cell, and (ii) modifying the phosphoinositide 3-kinase (PI3K) comprised by the immune cell such that its activity is modified compared to non-modified reference PI3K.
15. The method of claim 14, wherein the modified PI3K comprised by the immune cell is overactive compared to non-modified reference PI3K.
16. The method of claim 14, wherein said modification is the introduction of a point mutation in PI3K.
17. The method of claim 16, wherein said point mutation in PI3K is at amino acid position 81.
18. The method of claim 17, wherein by said point mutation in PI3K a glutamic acid (E) at position 81 is replaced by lysine (K) (PI3K.sup.E81K).
19. The method of claim 14, wherein the immune cell is a T cell or an NK cell.
20. The method of claim 14, wherein the immune cell is a chimeric antigen receptor (CAR) T or NK cell, which CAR comprises an extracellular antigen-binding domain, a transmembrane domain, and an intracellular signaling domain.
21. (canceled)
22. (canceled)
23. (canceled)
24. The method of claim 14, wherein in step (ii) the PI3K is modified by CRISPR/Cas9 editing.
25. (canceled)
26. The method of claim 24, wherein a sgRNA molecule is used comprising the nucleotide sequence of gctcttgctgctccgctgtc (SEQ ID NO: 1) or aagagctggaggacgagcaa (SEQ ID NO: 2).
27. (canceled)
28. A method of treating a living being suffering or at risk of suffering from cancer or non-malignant diseases, comprising the administration of the modified immune cell according to claim 1 into said living being.
29. The method of claim 28, wherein the modified immune cell was prepared starting from the living being's own immune cells (autologous) or from immune cells of a reference living being (allogeneic).
30. (canceled)
31. (canceled)
Description
BRIEF DESCRIPTION OF THE FIGURES
[0073] FIG. 1: Base editing in CAR T cells. [A] Schematic representation of the generation and experimental testing of base-edited CAR T cells. T cells were isolated from buffy coats and activated using CD3/CD28 beads and the cytokines IL7/IL 15 (day 0). Two days later, base editing of T cells was performed by electroporation, and the following day, viral transduction was performed to produce CAR T cells. (Re) stimulation of CAR T cells with target cells was performed at the time points shown using irradiated NALM6 cells (CD19+ B-ALL cell line) in an effector:target ratio of 2:1. At the endpoint, T cell differentiation and cytotoxic activity of CAR T cells were measured. [B] Schematic representation of the base editing technique with the point mutation obtained resulting in a conversion of a cytidine (C) to a thymidine (T) in the PI3KCD gene. The initial editing occurs on the reverse strand, resulting in a guanidine (G) to adenosine (A) mutation on the coding strand. Ultimately, a mutation is present in the PI3K protein, with a lysine (Lys, K) inserted instead of a glutamic acid (Glu, E) at position 81 (E81K). [C] Effect of E81K point mutation on PI3K downstream signaling pathways shown by Western blot using wild-type or E81K-modified SupT1 (T-LBL cell line) single cell clones as examples, tubulin serves as loading control. [D] Sequencing of wild-type control CAR T cells (WT), initially edited E81K CAR T cells (E81K) and edited E81K CAR T cells re-stimulated three times with target cells (final restim. E81K). The proportion of the mutated base in the total cell population of the gene PIK3CD at nucleotide position 242 (encoding E81) determined by TIDER (http://shinyapps.datacurators.nl/tider/) is shown after final re-stimulation (D18 final restim.). The altered base is highlighted in blue. [E] Sequencing of T cells initially and four months after genomic editing. The changed base is highlighted in grey.
[0074] FIG. 2: E81K CAR T cells show enhanced cytotoxicity compared to unmodified CAR T cells. [A] Cytotoxic activity of untransduced T cells (UT), 19BBz CAR T cells (19BBz) and E81K-modified 19BBz CAR T cells (19BBz E81K). Luciferase-expressing (ffLuc)-NALM6 cells and CAR T cells were mixed in the ratios shown and cultured for 4 h and 18h, respectively. The lysis that occurred was determined by the luminescence signal of the NALM6 cells. [B] Cytotoxic activity of untransduced T cells, 19BBz CAR T cells and 19BBz E81K CAR T cells. CAR T cells after second (left panel) and third re-stimulation (right panel) with the target cells were mixed with ffLuc-NALM6 cells in the ratios shown and cultured for 18 h. The lysis that occurred was determined by the luminescence signal of the NALM6 cells. [C] Cytotoxic activity of untransduced T cells (UT), untransduced E81K-modified T cells (UT E81K), 19BBz CAR T cells and 19BBz E81K CAR T cells. ffLuc-NALM6 cells with low CD19 expression (CD19-med) and CAR T cells were mixed in the ratios shown and cultured for 4 h and 18h, respectively. The lysis that occurred was determined by the luminescence signal of the NALM6 cells.
[0075] FIG. 3: The enhanced cytotoxic activity of CAR T cells by E81K modification persists initially and is maintained after multiple restimulations with tumor cells. [A] Overview of cytotoxic CAR T cell efficacy after 4h co-cultivation of ffLuc-NALM6 cells and 19BBz T cells (mock) or 19BBz E81K T cells (E81K) from healthy donors (n=8). Shown are results at CAR T-cell to NALM6 ratios of 2:1, 1:1 and 1:2. p-values were determined using a paired t-test. [B] Overview of cytotoxic CAR T cell efficacy 18h after co-culturing ffLuc-NALM6 cells and 19BBz T cells or 19BBz E81K T cells from healthy donors (n=8). Shown are results at CAR T-cell to NALM6 ratios of 2:1, 1:1, and 1:2. p-values were determined using a paired t-test. [C] Overview of cytotoxic CAR T cell efficacy 18h after co-culturing ffLuc-NALM6 cells and triply re-stimulated 19BBz T cells or 19BBz E81K T cells from healthy donors (n=5). Shown are results at CAR T cell to NALM6 ratios of 2:1, 1:1, and 1:2. p-values were determined using a paired t-test.
[0076] FIG. 4: E81K CAR T cells show enhanced cytotoxicity compared to unmodified CAR T cells. [A] Cytotoxic activity of untransduced T cells (UT), anti-Mesothelin-BBz CAR T cells (aMeso-BBz) and E81K-modified anti-Mesothelin-BBz CAR T cells (aMeso-BBz E81K). Luciferase-expressing (ffLuc)-NALM6 cells and CAR T cells were mixed in the ratios shown and cultured for 18 h. The lysis that occurred was determined by the luminescence signal of the NALM6 cells. [B] Cytotoxic activity of untransduced T cells (UT), CAR T cells and E81K-modified CAR T cells (E81K CAR T cells). Luciferase-expressing (ffLuc)-solid tumor cell lines and CAR T cells were mixed in the ratios shown and cultured for 18h. Lysis was determined by the luminescence signal of the tumor cells. [C] Cytotoxic activity of untransduced T cells (UT), anti-CD19-28z1xx CAR T cells (1xx) and E81K-modified anti-CD19-28z1xx CAR T cells (1xx E81K). Luciferase-expressing (ffLuc)-NALM6 cells and CAR T cells were mixed in the ratios shown and cultured for 18 h. The lysis that occurred was determined by the luminescence signal of the NALM6 cells.
[0077] FIG. 5: E81K CAR T cells exhibit increased antigen-dependent proliferative capacity but do not show increased antigen-independent proliferative potential. [A] Cell numbers of vital 19BBz CAR T cells with or without E81K point mutation (E81K or mock) from two healthy donors over 14 days after cytokine discontinuation. [B] Cell number of vital 19BBz CAR T cells with and without E81K point mutation under cytokine stimulation using a healthy donor as an example over 18 days. [C] Antigen-dependent proliferation seven days after twice re-stimulation of 19BBz CAR T cells with and without E81K point mutation. [D] Antigen-dependent proliferation of 19BBz CAR T cells with and without E81K point mutation after multiple tumor re-stimulation.
[0078] FIG. 6: Phenotypic characterization of E81K CAR T cells after chronic antigen stimulation shows no increased T cell differentiation or exhaustion. [A] Differentiation of CD8.sup.+ and CD4.sup.+ CAR T cells with (E81K) and without E81K (mock) point mutation after three times re-stimulation with the irradiated B-ALL cell line NALM6 in n=3 donors (D1, D2, D3). Using flow cytometric analysis of surface markers CCR7, CD62L, CD95 and CD45RA to determine T cell differentiation, five different T cell populations are distinguished (naive T cells: T_N; stem cell memory T cells: T_SCM; central memory T cells: T_CM; effector memory T cells: T_EM; effector T cells: T_EFF). [B] The percentage of CD57 positive T cells with (E81K) and without E81K (mock) point mutation was determined by flow cytometry. CAR T cells from donor1 after single restimulation and untransduced T cells from donor4 after four months of cultivation are shown; the results of T cells with (E81K) and without (mock) E81K point mutation are compared. [C] Schematic representation of the generation and experimental evaluation of the phenotype of base-edited CAR T cells after chronic stimulation with target cells compared to controls. Re-stimulation of CAR T cells with (E81K) and without E81K (mock) point mutation were performed at the time points shown with irradiated CD19-overexpressing 3T3 fibroblast cells in an effector to target ratio of 2:1. At the endpoint, CAR T cells were analyzed for PD-1, TIM3, and LAG3 expression by flow cytometry. [D] Proportion of PD-1 positive and PD-1, TIM3 and LAG3 triple positive CD4.sup.+ and CD8.sup.+ CAR T cells with and without E81K point mutation after the third, fourth and fifth re-stimulation according to the scheme shown in [C].
[0079] FIG. 7: NALM6 (B-ALL) tumor cells were injected i.v. into immunodeficient NSG mice and 4 days later a single therapy with CD19-specific CAR T cells with or without E81K mutation (i.v.) was administered. The tumor growth and survival of the mice was then monitored.
[0080] FIG. 8: NALM6 (B-ALL) tumor cells were injected i.v. into immunodeficient NSG mice and 4 days later a single therapy with CD19-specific CAR T cells with and without E81K mutation (i.v.) was administered. [A] The blood of CAR T cell-treated mice was analyzed for the presence and number of CAR T cells by means of flow cytometry. [B] The corresponding CAR T cells with and without E81K mutation were isolated from a group of CAR T cell-treated mice and examined ex vivo for their (preserved) cytotoxic potential against NALM6 tumor cells.
[0081] FIG. 9: NALM6 (B-ALL) tumor cells were injected i.v. into immunodeficient NSG mice and 4 days later a single therapy with CD19-specific CAR T cells with and without E81K mutation (i.v.) was administered. The different T cell populations in the blood of the mice were determined by means of flow cytometry (TN=naive T cells; TSCM: stem cell memory T cells; TCM: central memory T cells; TEM: effector memory T cells; TEFF: effector T cells).
EXAMPLES
1. Material and Methods
[0082] Primary human T cells were isolated from human blood samples and activated for two days with CD3/CD28 Dynabeads in the presence of IL-7 and IL-15. 48h after activation Dynabeads were removed and 310.sup.6 T cells were electroporated using the Neon electroporator (1400V, 3 pulses, 10 ms, Buffer T, 100 l). For each electroporation, 10 g of in vitro synthesized ARCA-capped or CleanCap-capped and PolyA-tailed mRNA of an C to T Base editor like AncBE4max and 10 g of the synthetic gRNA (IDT) were used. 24 h after electroporation, cells were transduced with the respective CAR construct using a retroviral system. Cells were cultured under presence of IL-7 and IL-15 until further experiments were performed.
[0083] The point mutation was inserted via base editing, a process making use of the CRISPR/Cas9 system. This system relies on a Cas9-nickase which is coupled to a cytidine deaminase, which converts cytosines to uracils, which are read by polymerases as thymines.
[0084] Specifically, the catalytically impaired Cas9 nuclease domain localizes a single strand (ss)DNA deaminase enzyme to a genomic target sequence of interest. Upon binding of the Cas9 protein, hybridization of the guide RNA spacer to the target DNA strand causes displacement of the PAM-containing genomic DNA strand to form a ssDNA R-loop. PAM-distal nucleotides are exposed as ssDNA and are accessible to the deaminase domain of the base editor. Therefore, gene editing on a genomic level is achieved by installing a precise point mutation in the respective locus. An additional mutation was not found so far. Nevertheless, a further mutation in the editing window of the specific sgRNA would be possible for Q79, resulting in a silent point mutation.
[0085] The AncBE4max Base editor in combination with the following sgRNA sequence gctcttgctgctccgctgtc (SEQ ID NO: 1) results in a point mutation in the PIK3CD gene, where the glutamic acid at position 81 gets mutated to a lysine. In general, all C to T Base editors with an NGG or NG PAM can induce this mutation.
[0086] The A to G Base editor ABEmax in combination with the following sgRNA sequence aagagctggaggacgagcaa (SEQ ID NO: 2) results in a point mutation in the PIK3CD gene, where the glutamic acid at position 81 gets mutated to a glycine. In general, all A to G Base editors with an NGG or NG PAM can induce this mutation.
2. Results
[0087] FIG. 1A is a schematic representation of the generation and experimental testing of base-edited CAR T cells according to the invention. T cells were isolated from buffy coats and activated using CD3/CD28 beads and the cytokines IL7/IL15 (day 0). Two days later, base editing of T cells was performed by electroporation, and the following day, viral transduction was performed to produce CAR T cells. (Re) stimulation of CAR T cells with target cells was performed at the time points shown using irradiated NALM6 cells (CD19.sup.+ B-ALL cell line) in an effector:target ratio of 2:1. At the endpoint, T cell differentiation and cytotoxic activity of CAR T cells were measured.
[0088] Editing T cells at the genomic level is particularly suitable, as this ensures that there is no exogenous overexpression of PI3K and no uncoupling of the biological receptor, so that negative regulation is also guaranteed. At the same time, however, it also ensures good efficiency and feasibility in primary immune cells.
[0089] FIG. 1B is a schematic representation of the base editing technique with the point mutation obtained. Editing at the genomic level is performed by a CRISPR/Cas9 variant, which consists of a Cas9 nickase and causes only a single-strand break. This has the great advantage that the formation of translocations as well as the DNA damage response caused by a possible double strand break is significantly reduced. The Cas9 nickase is also coupled to a cytidine deaminase and two uracil DNA glycosylase inhibitors (UGIs), which together constitute the so-called base editor. The editing results in a conversion of a cytidine (C) to a thymidine (T) in the PI3KCD gene. The initial editing occurs on the reverse strand, resulting in a guanidine (G) to adenosine (A) mutation on the coding strand. Ultimately, a mutation is present in the PI3K protein, with a lysine (Lys, K) inserted instead of a glutamic acid (Glu, E) at position 81 (E81K).
[0090] FIG. 1C shows the effect of the E81K point mutation on PI3K downstream signaling pathways shown by Western blot using wild-type or E81K-modified SupT1 (T-LBL cell line) single cell clones as examples, tubulin serves as loading control. The application of this method resulted in increased phosphorylation of the downstream protein Akt in the signaling cascade in the T cell line SupT1, indicating increased PI3K activity.
[0091] FIG. 1D demonstrates via sequencing of wild-type control CAR T cells (WT), initially edited E81K CAR T cells (E81K) and edited E81K CAR T cells re-stimulated three times with target cells (final restim. E81K), that the inserted mutation persists after multiple antigen stimulation in a CAR T-cell context. The proportion of the mutated base in the total cell population of the gene PIK3CD at nucleotide position 242 (encoding E81) determined by TIDER (http://shinyapps.datacurators.nl/tider/) is shown after final re-stimulation (D18 final restim.). The altered base is highlighted in blue.
[0092] As shown in FIG. 1E via the sequencing of T cells initially and four months after genomic editing the inserted mutation is stable in primary T cells for several weeks. The changed base is highlighted in blue.
[0093] As shown in FIG. 2A, the E81K mutation in primary 19BBz CAR T cells shows enhanced cytotoxic activity 4 h as well as 18 h after incubation with target cells compared to unmodified CAR T cells.
[0094] Even after two- and three-times antigen stimulation, the edited T cells retain the enhanced cytotoxic activity; see FIG. 2B.
[0095] Furthermore, the E81K-edited 19BBz CAR T cells showed enhanced cytotoxic activity against tumor cells with lower antigen levels; cf. FIG. 2C.
[0096] As shown in FIG. 3A and FIG. 3B, the enhanced cytotoxic activity of E81K-modified CAR T cells after 4 h and 18h was significantly different in multiple donors (n=8) and was also maintained after antigen re-stimulation three times; see FIG. 3C. In conclusion, these results demonstrate a sustained increase in the functionality of E81K-modified CAR T cells against tumor cells with different antigen levels.
[0097] Enhanced antitumor efficacy of E81K modified CAR T cells could be confirmed with CARs targeting other antigens and against solid tumors (e.g., Mesothelin-specific CARs, FIG. 4A). There is improved cytotoxicity of the E81K-modified CAR T cells compared to the unmodified CAR T cells in the context of various solid tumors (FIG. 4B). In addition, the E81K modification leads to an improved therapeutic potential of different CAR designs (e.g., 1928z1XX CAR T cells, FIG. 4C).
[0098] To exclude a transforming effect of the E81K mutation, cell growth was examined without (FIG. 5A) and with cytokines (FIG. 5B) as well as with and without antigen stimulation. Thereby, in an antigen-independent context, there was no increased cell proliferation of E81K-mutated CAR T cells compared with wild-type CAR T cells. However, under antigen-stimulation, E81K-mutated CAR T cells showed enhanced cell growth (FIG. 5C), indicating an improved specific proliferation potential of E81K-mutated CAR T cells. The E81K CAR T cells show improved antigen-dependent proliferation in different donors compared to the standard 19BBz CAR T cells after multiple tumor exposures (n=6 biological replicates) (FIG. 5D).
[0099] Re-stimulation with antigen-expressing target cells three times resulted in a slightly increased proportion of effector T cells in the E81K-mutated 19BBz CAR T cells, but without showing a significantly altered differentiation pattern; see FIG. 6A. Moreover, the E81K mutation in T cells and CAR T cells did not result in increased expression of CD57, which is known as a marker for senescent T cells; see FIG. 6B. Moreover, even under extensive chronic antigen stimulation (FIG. 6C), the E81K mutant CAR T cells did not show enhanced expression of inhibitory surface markers such as PD-1, TIM3, and LAG3, which are associated with T cell exhaustion and CAR T cell dysfunction; see FIG. 6D.
[0100] NALM6 (B-ALL) tumor cells were injected i.v. into immunodeficient NSG mice and 4 days later a single therapy with CD19-specific CAR T cells with or without E81K mutation (i.v.) was administered. The tumor growth and survival of the mice was then monitored. Mice treated with E81K 19BBz CAR T cells showed significantly improved survival compared to conventional 19BBz CAR T cells (FIG. 7).
[0101] NALM6 (B-ALL) tumor cells were injected i.v. into immunodeficient NSG mice and 4 days later a single therapy with CD19-specific CAR T cells with and without E81K mutation (i.v.) was administered. [A] The blood of CAR T cell-treated mice was analyzed for the presence and number of CAR T cells by means of flow cytometry. [B] The corresponding CAR T cells with and without E81K mutation were isolated from a group of CAR T cell-treated mice and examined ex vivo for their (preserved) cytotoxic potential against NALM6 tumor cells. There was an increased persistence of E81K CAR T cells in the blood of the mice. These also showed significantly improved and sustained cytotoxicity compared to standard CAR T cells (FIG. 8).
[0102] NALM6 (B-ALL) tumor cells were injected i.v. into immunodeficient NSG mice and 4 days later a single therapy with CD19-specific CAR T cells with and without E81K mutation (i.v.) was administered. The different T cell populations in the blood of the mice were determined by means of flow cytometry. There is an increased number of E81K CAR T cells in all T cell populations, in particular there is a dominance of the effector memory CAR T cell population (FIG. 9).
[0103] In summary, the results show that E81K mutation induced by base editing leads to a sustained increase in CAR T cell functionality. E81K mutated CAR T cells exhibit increased cytotoxicity and antigen-dependent proliferation, indicating improved therapeutic potential compared to currently approved CAR T cell products. The induced mutation is stably maintained for several weeks. A PIK3CD E81K mutation has been described in a few patients, where it resulted in an increased proportion of CD57.sup.+ T cells and an increased induction of T cell exhaustion during the course. The inventors could not observe this development in the CAR T-cell setting even after multiple antigen stimulation of the CARs, but could demonstrate a sustained increased functionality of the E81K mutated CAR T cells.
[0104] On the one hand, it is significant for the analysis of the safety of the cell product that the E81K mutated T cells do not show an increased antigen-independent proliferation potential, so that no increased transformation potential can be assumed (FIGS. 4A and C). Furthermore, the E81K mutation has been detected in tumor patients in only four cases out of a cohort of over 64,000 cases, and no T cell tumor is represented (https://www.cbioportal.org/). This suggests that the E81K mutation in T cells is not associated with malignant potential. This is also confirmed by the fact that patients with E81K mutations did not develop T-cell malignancy in the two cases described above.
[0105] A limitation of current CAR T cell products is also the reduced sensitivity to tumor cells with low antigen expressionthis can lead to recurrences of tumor cells with reduced antigen density and is also relevant in solid tumors given their heterogeneity. The functional enhancement of CAR T cells by the introduced E81K mutation is also confirmed against tumor cells with low antigen expression, so that the therapeutic potential of CAR therapy could also be improved in terms of enhanced antigen sensitivity.
[0106] Overall, the results show that an E81K mutation can be successfully generated in CAR T cells by genetic modification, leading to a sustained increase in functionality against tumor cells with different antigen expression levels. Notably, the increased activation does not lead to increased T cell exhaustion or senescence. Mutations in the PI3K pathway and in particular E81K mutations thus have great potential to further improve CAR therapies and prevent recurrences. Furthermore, this gene editing strategy can be extended to other cellular therapies, used in combination therapies and thus evaluated as a therapy for tumor diseases and other severe diseases.
