Immunotherapy with gene-modified T cells: limiting side effects provides new challenges.

Genetic tools have been developed to efficiently engineer T-cell specificity and enhance T-cell function. Chimeric antigen receptors (CAR) use the antibody variable segments to direct specificity against cell surface molecules. T-cell receptors (TCR) can redirect T cells to intracellular target proteins, fragments of which are presented in the peptide-binding groove of HLA molecules. A recent clinical trial with CAR-modified T cells redirected against the B-cell lineage antigen CD19 showed dramatic clinical benefit in chronic lymphocytic leukaemia patients. Similarly, impressive clinical responses were seen in melanoma and synovial cell carcinoma with TCR-modified T cells redirected against the melanocyte lineage antigen MART-1 and the testis-cancer antigen NY-ESO-1. However, on and off-target toxicity was associated with most of these clinical responses, and fatal complications have been observed in some patients treated with gene modified T cells. This review will discuss factors that might contribute to toxic side effects of therapy with gene modified T cells, and outline potential strategies to retain anticancer activity while reducing unwanted side effects.

Identification of two cytotoxic T lymphocyte-recognized epitopes in the Ras protein.

We have investigated the possibility of inducing cytotoxic T lymphocytes (CTL) to Ras containing a mutation at position 61 or to normal Ras, using recombinant vaccinia viruses expressing these proteins. CTL from C57Bl/10 mice immunized with vaccinia expressing mutant Ras showed specificity for the mutant Ras protein and recognition of normal Ras was inefficient. The opposite specificity was observed after immunization with vaccinia expressing normal Ras, since CTL isolated from these mice recognized normal Ras well and mutant Ras inefficiently. Levels of endogenous Ras expression were insufficient for lysis by these CTL. One CTL epitope mapped to amino acids 60-67 and residue 61 was critical for T cell recognition. CTL generated against mutant Ras protein recognized peptide 60-67 containing mutant residue 61, while anti-normal Ras CTL recognized the wild-type 60-67 sequence. A second epitope mapped to residues 152-159 of Ras and was recognized equally well by CTL raised to normal or mutant Ras. The murine data raise the possibility of exploiting Ras-specific CTL for targeted immunotherapy of certain human cancers.

Identification of a unique tumor antigen as rejection antigen by molecular cloning and gene transfer.

Tumor-specific transplantation antigens are antigens that can lead to complete immunological destruction of a transplanted cancer by the syngeneic host. When such antigens are expressed on cancers induced by chemical or physical carcinogens, then they are usually unique, i.e., antigenically different for each independently induced tumor. In this study, we show that the product of a gene encoding a novel MHC class I molecule and isolated from the murine UV light-induced regressor tumor 1591 represents one such unique tumor-specific transplantation antigen that causes tumor rejection. The major evidence comes from our finding that 1591 progressor variants regularly lost the gene encoding this antigen that is expressed in the parental tumor that regresses in normal mice; furthermore, reintroduction of this gene into a 1591 progressor variant by DNA transfection caused the progressor variant to regress in normal immunocompetent mice. Thus, the progressor tumor reverted to the parental regressor phenotype following transfection. Consistent with the conclusion that the expression of the novel MHC class I gene following transfection was responsible for the regressor phenotype is also our finding that a variant of the transfected tumor that had lost expression of the transfected gene resumed its progressive growth behavior. Finally, we show that the molecule encoded by the novel class I gene is specifically recognized by a syngeneic tumor-specific cytolytic T cell clone that we have previously shown to select in vitro for progressor variants from the parental regressor tumor cell line. It remains to be determined to what extent unique tumor-specific rejection antigens of other highly immunogenic regressor tumors are encoded by novel MHC class I genes and whether these genes represent germline mutations or somatic mutations caused by the carcinogen treatment.

Animals bearing malignant grafts reject normal grafts that express through gene transfer the same antigen.

Breaking the state of immunological unresponsiveness of tumor-bearing individuals to cancer is a prerequisite for active or passive tumor-specific immunotherapy. To study this problem the immunogenic MHC class I antigen, K216 was transfected into a progressor tumor. The transfected tumors were regularly rejected by normal mice but grew progressively in mice bearing nontransfected tumors. In addition, transgenic mice were derived to obtain normal cells and tissues expressing the same K216 gene product. Normal mice rejected K216-positive normal or malignant tissue grafts and generated K216-specific CTL in vitro and in vivo in response to these challenges. In contrast, mice bearing nontransfected tumors, though rejecting K216-positive nonmalignant tissue grafts, did not reject K216-positive tumors nor generate K216-specific CTL in response to K216-positive tumor cells. Mice bearing K216-positive tumors also rejected the nonmalignant K216-positive tissue grafts, but this in vivo response failed to lead to rejection of the simultaneously present tumor graft expressing the same antigen; in fact, immunity had no measurable effect whatsoever on tumor size or incidence and caused no selection for antigen loss variants. Taken together, the present findings suggest that transfer of expression of a target antigen into nonmalignant cells provides a way for obtaining effective stimulation of antigen-specific CTL in tumor-bearing mice, but that additional manipulations will be required to cause immunological rejection of established tumors.

Retroviral transfer of a dominant TCR prevents surface expression of a large proportion of the endogenous TCR repertoire in human T cells.

The latent membrane protein-2 (LMP2) of Epstein-Barr virus is a potential target for T-cell receptor (TCR) gene therapy of Hodgkin lymphoma and nasopharyngeal carcinoma. Here, we modified a human leukocyte antigen-A2-restricted, LMP2-specific TCR to achieve efficient expression following retroviral TCR gene transfer. The unmodified TCR was poorly expressed in primary human T cells, suggesting that it competed inefficiently with endogenous TCR chains for cell surface expression. In order to improve this TCR, we replaced the human constant region with murine sequences, linked the two TCR genes using a self-cleaving 2A sequence and finally, codon optimized the TCR-alpha-2A-beta cassette for efficient translation in human cells. Retroviral transfer of the modified TCR resulted in efficient surface expression and HLA-A2/LMP2 pentamer binding. The transduced cells showed peptide-specific interferon-gamma and interleukin-2 production and killed target cells displaying the LMP2 peptide. Importantly, the introduced LMP2-TCR suppressed the cell surface expression of a large proportion of endogenous TCR combinations present in primary human T cells. The design of dominant TCR is likely to improve TCR gene therapy by reducing the risk of potential autoreactivity of endogenous and mispaired TCR combinations.

Antigen-specific cellular immunotherapy of leukemia.

Advances in cellular and molecular immunology have led to the characterization of leukemia-specific T-cell antigens and to the development of strategies for effective augmentation of T-cell immunity in leukemia patients. While several leukemia-related antigens have been identified, this review focuses on the Wilms' tumor 1 (WT1) antigen and the proteinase 3 (Pr3) antigen that are overexpressed in leukemic cells and are already being used in the clinical setting. Moreover, WT1 is also overexpressed in a vast number of nonhematological solid tumors, thereby expanding its use as a promising target for cancer vaccines. Examples of spontaneous immune responses against WT1 and Pr3 in leukemia patients are presented and the potential of WT1 and Pr3 for adoptive T-cell immunotherapy of leukemia is discussed. We also elaborate on the use of professional antigen-presenting cells loaded with mRNA encoding WT1 exploiting the advantage of broad HLA coverage for therapeutic vaccination purposes. Finally, the summarized data underscore the potential of WT1 for the manipulation of T-cell immunity in leukemia and in cancer in general, that will likely pave the way for the development of more effective and generic cancer vaccines.

T cell recognition of a point mutation in the P21 Ras protein.

Point mutations of p21 Ras proteins correlate with many human malignancies. To determine whether the mutations of Ras proteins generate immunogenic determinants which can be recognized by T cells and possibly serve as targets for immunotherapy, we studied the murine T helper responses to synthetic Ras peptides corresponding to amino acids 1-23 of normal or mutant Ras protein. Immunization of C3H/He and B10.BR mice with Ras peptides containing a valine mutation at position 12 stimulated MHC class II-restricted T helper cells which recognised specifically the Ras mutation. Surprisingly C57BL/10 mice generated T helper responses not only against mutant but also against normal Ras peptides. Importantly, natural processing of Ras protein was found to generate the epitopes recognized by the peptide-induced T cells.

The CD68 protein as a potential target for leukaemia-reactive CTL.

CD68, a haematopoietic differentiation marker of the monocyte-macrophage lineage, is expressed in various human malignancies including chronic and acute myeloid leukaemia (AML). While the majority of normal CD34(+) cells are negative for CD68 expression, CD34(+) cells from AML patients produce elevated amounts of this protein. The purpose of this study was to identify CTL epitopes in the human CD68 protein. Mouse CD68 was also analysed to search for epitopes that could be used in murine tumor model. Peptides binding to murine H2(b) class I molecules were identified and used to stimulate CTL responses from allogeneic donor mice to avoid immunological tolerance. High avidity CTL clones specific for three different peptide epitopes did not kill CD68-expressing murine target cells, indicating that endogenous antigen processing failed to produce sufficient amounts of these peptides. In contrast, allo-restricted human CTL specific for an HLA-A2-binding peptide of CD68 recognised not only picomolar concentrations of peptide, but also displayed low levels of killing against HLA-A2-positive K562 and THP-1 leukemia cell lines and blast cells from AML patients. These data suggest that human leukaemia cells express limited amounts of CD68-derived peptides, and that high avidity CTL capable of recognising sub-picomolar concentrations of peptides are required for efficient killing of leukaemia cells.

Isolation of endothelial cells from murine tissue.

The isolation and long-term culture of murine endothelial cells (ECs) has often proven a difficult task. In this paper we describe a quick, efficient protocol for the isolation of microvascular endothelial cells from murine tissues. Murine lung or heart are mechanically minced and enzymatically digested with collagenase and trypsin. The single cell suspension obtained is then incubated with an anti-CD31 antibody, anti-CD105 antibody and with biotinylated isolectin B-4. Pure EC populations are finally obtained by magnetic bead separation using rat anti-mouse Ig- and streptavidin-conjugated microbeads. EC cultures are subsequently expanded and characterised. The surface molecule expression by the primary cultures of murine EC obtained from lung and heart tissue is analysed and compared to that of a murine endothelioma and of primary cultures of murine renal tubular epithelial cells. The phenotype and morphology of these cultures remain stable over 10-15 passages in culture, and no overgrowth of contaminating cells of non-endothelial origin is observed at any stage.

CD4-positive and B lymphocytes in transplantation immunity. I. Promotion of tumor allograft rejection through elimination of CD4-positive lymphocytes.

The elimination of CD4+ cells by anti-CD4 antibody caused regression of a malignant solid tumor allograft that does not lose a cytotoxic T lymphocyte-defined target antigen during tumor progression and requires specific CD8+ CTL for tumor rejection. Treatment with anti-CD4 antibody was effective when started 1-2 weeks after tumor challenge and was at least as effective as treating with anti-CD3 antibody or specific immunization with the antigen expressed on malignant or nonmalignant cells. None of these treatments caused rejection of tumors that were larger than 1 cm3 or had been growing for 3 weeks or longer in the host. Mice bearing large and long-established tumors treated with anti-CD4 antibody rejected a new tumor challenge but failed to reject the long-established tumor. Similarly, mice with established tumors mounted effective CTL responses to reject skin grafts but failed to reject tumors which expressed the same antigen. Treatment with anti-CD4 antibody eliminated primary T lymphocyte dependent antibody responses but failed to suppress ongoing antibody responses to continuous antigenic stimulation. Possibly, the effectiveness of early treatment and the failure of later treatment with anti-CD4 antibody results indirectly from the effect treatment has on B lymphocytes.

Use of the allogeneic TCR repertoire to enhance anti-tumor immunity.

It is well established that antigen-specific T lymphocytes can inhibit tumor growth in humans and in mice, leading to complete tumor elimination in some cases. However, in many cases T cell immunity is unable to successfully control tumor progression. Since tumors are derived from normal tissues, most antigens are shared with normal tissues, although expression levels are usually elevated in malignant cells. Nevertheless, low-level expression in normal cells can be sufficient to render autologous T cells tolerant and thus unable to mount effective immune responses against tumors. Here, we review how allogeneic T cells can be used to isolate T cells that effectively recognise and kill tumor cells, but not normal cells with low level of antigen expression. The TCR of allogeneic T cells can be introduced into patient T cells to equip them with anti-tumor specificity that may not be present in the autologous T cell repertoire.

Exploiting T cell receptor genes for cancer immunotherapy.

Adoptive antigen-specific immunotherapy is an attractive concept for the treatment of cancer because it does not require immunocompetence of patients, and the specificity of transferred lymphocytes can be targeted against tumour-associated antigens that are poorly immunogenic and thus fail to effectively trigger autologous T cell responses. As the isolation and in vitro expansion of antigen-specific lymphocytes is difficult, 'conventional' adoptive T cell therapy can only be carried out in specialized centres in small numbers of patients. However, T cell receptor (TCR) genes isolated from antigen-specific T cells can be exploited as generic therapeutic molecules for 'unconventional' antigen-specific immunotherapy. Retroviral TCR gene transfer into patient T cells can readily produce populations of antigen-specific lymphocytes after a single round of polyclonal T cell stimulation. TCR gene modified lymphocytes are functionally competent in vitro, and can have therapeutic efficacy in murine models in vivo. TCR gene expression is stable and modified lymphocytes can develop into memory T cells. Introduction of TCR genes into CD8(+) and CD4(+) lymphocytes provides an opportunity to use the same TCR specificity to produce antigen-specific killer and helper T lymphocytes. Thus, TCR gene therapy provides an attractive strategy to develop antigen-specific immunotherapy with autologous lymphocytes as a generic treatment option.

T cell epitopes in human papilloma virus proteins.

Infection by HPV is associated with several human diseases such as warts of the skin, condylomata of the genital track and carcinoma of the cervix. Although there is strong evidence for immune control of HPV types causing warts and condylomata, it is currently unclear whether patients infected with transforming HPV types can mount efficient T cell responses. Despite the apparent low immunogenicity of transforming HPV types, several Th and CTL epitopes have been identified in proteins derived from HPV16. This transforming virus is most frequently present in women with CIN and cervical carcinoma and knowledge of T cell recognisable proteins may eventually lead to the design of immune-stimulating anti-HPV16 vaccines.

Peptide-specific cytotoxic T lymphocytes restricted by nonself major histocompatibility complex class I molecules: reagents for tumor immunotherapy.

Studies in melanoma patients have revealed that self proteins can function as targets for tumor-reactive cytotoxic T lymphocytes (CTL). One group of self proteins MAGE, BAGE, and GAGE are normally only expressed in testis and placenta, whilst another group of CTL recognized proteins are melanocyte-specific differentiation antigens. In this study we have investigated whether CTL can be raised against a ubiquitously expressed self protein, mdm-2, which is frequently overexpressed in tumors. The observation that T-cell tolerance is self major histocompatibility complex-restricted was exploited to generate CTL specific for an mdm-2 derived peptide presented by nonself major histocompatibility complex class I molecules. Thus, the allo-restricted T-cell repertoire of H-2d mice was used to isolate CTL specific for the mdm100 peptide presented by allogeneic H-2Kb class I molecules. In vitro, these CTL discriminated between transformed and normal cells, killing specifically Kb-positive melanoma and lymphoma tumors but not Kb-expressing dendritic cells. In vivo, the CTL showed antitumor activity and delayed the growth of melanoma as well as lymphoma tumors in H-2b recipient mice. These experiments show that it is possible to circumvent T-cell tolerance to ubiquitously expressed self antigens, and to target CTL responses against tumors expressing elevated levels of structurally unaltered proteins.

The alpha 3 domain of major histocompatibility complex class I molecules plays a critical role in cytotoxic T lymphocyte stimulation.

Allogeneic major histocompatibility complex class I molecules induce strong cytotoxic T lymphocyte (CTL) responses whereas xenogeneic molecules do not. We have tested a series of mouse/human hybrid molecules for their ability to stimulate mouse CTL. The molecules with murine alpha 3 domains consistently stimulated stronger CTL responses than those with human alpha 3 domains, independent of the species origin of the N-terminal alpha 1 or alpha 2 domains. We have found that the ability of class I molecules to induce strong cytotoxic responses correlates positively with their ability to stimulate expansion of the CD8+CD4-T cell subset. The results indicate that mouse T cells can recognize class I molecules with human alpha 1 and/or alpha 2 domains, but for efficient stimulation of these T cells it is important that the immunizing molecule contains a murine alpha 3 domain. We suggest that T cell priming requires an efficient interaction of CD8 with the class I alpha 3 domain, and this shows some species restriction.

Prospects for immunotherapy of malignant disease.

The majority of T cell-recognized tumour antigens in humans are encoded by genes that are also present in normal tissues. Low levels of gene expression in normal cells can lead to the inactivation of high-avidity T cells by immunological tolerance mechanisms. As a consequence, low-avidity T cell responses in patients are often inadequate in providing tumour protection. Recently, several technologies have been developed to overcome tolerance, allowing the isolation of high-affinity, HLA-restricted receptors specific for tumour-associated peptide epitopes. Furthermore, transfer of HLA-restricted antigen receptors provides an opportunity to empower patient T cells with new tumour-reactive specificities that cannot be retrieved from the autologous T cell repertoire.

A synthetic peptide derived from the tumor-associated protein mdm2 can stimulate autoreactive, high avidity cytotoxic T lymphocytes that recognize naturally processed protein.

Studies in melanoma patients have shown that unaltered self proteins can function as targets for tumor-reactive CTL. Here, we have investigated in a murine model whether autoreactive CTL can be found against the widely expressed proteins cyclin D1, mdm2, and p53, which are frequently overexpressed in transformed cells. Sixteen MHC class I binding peptides were identified in these proteins, and seven of them consistently stimulated primary CTL in vitro. Avidity measurements revealed that the avidity of peptide-induced CTL differed by >1000-fold. The highest avidity CTL were induced by a peptide derived from mdm2. These CTL recognized target cells expressing mdm2 endogenously, while CTL generated against the remaining peptides were of lower avidity and did not recognize cells expressing relevant proteins endogenously. Generation of high avidity anti-mdm2 CTL required several cycles of peptide stimulation, suggesting that the CTL precursor frequency was low. The data show the normal T cell repertoire contains small numbers of potentially autoreactive CTL. Expansion of these CTL may lead to beneficial autoimmunity against tumors, but, equally, it may be the basis of detrimental autoimmune diseases.

The CD45RA molecule is expressed in naive murine CTL precursors but absent in memory and effector CTL.

We have studied the expression of the CD45RA molecule in murine cytotoxic T lymphocytes (CTL) specific for the allogeneic H-2Kb molecule at different stages of differentiation. The CD45RA phenotype of naive H-2Kb-specific CTL precursors has been determined using primary in vitro CTL responses. For the analysis of memory CTL we have immunized mice in vivo followed by restimulation in vitro. We have also determined the CD45RA expression at the CTL effector stage. Our results show that among naive CD8+ T cells both the CD45RA+ and the CD45RA- subpopulations can mount Kb-specific CTL responses. In contrast, memory CTL responses are mediated only by the CD8+ CD45RA+ T cell subpopulation. Similarly, effector CTL are CD8+ CD45RA- while the CD8+ CD45RA+ subpopulation does not exhibit specific cytolytic activity. The data indicate that CD45RA expression changes during CTL differentiation and that memory as well as effector CTL lack this marker.