Immunotherapy in Acute Leukemias: Implications and Perspectives Using Wt1 Antigen

WT1 Transcription Factor

The WT1 gene encodes a 55 kDa transcription factor (11), which has a carboxyl-terminal region containing four zinc-finger domains that mediate DNA binding. The amino terminal end, which is rich in prolines and glutamines, contains transcriptional activation domains, similar to other transcription factors (1).

This transcription factor is involved in the regulation of many cellular processes, including proliferation, differentiation, mRNA processing and apoptosis (12, 13), and may act as a repressor of growth factor genes' transcription, such as platelet-derived growth factor (PDGF), colony-stimulating factor 1 (CSF-1), insulin-like growth factor 2 (IGF-2) and transforming growth factor beta (TGF-β), as well as receptor genes for growth factors, such as IGF-1 receptor (IGF-1R), epidermal growth factor receptor (EGFR) and vitamin D receptor (14).

Furthermore, it may also act by mediating expression on other genes, such as retinoic acid receptor alpha (RAR-α), v-myb avian myeloblastosis viral oncogene homolog (c-myb), v-myc avian myelocytomatosis viral oncogene homolog (c-myc), ornithine decarboxylase (ODC) and v-myc avian myelocytomatosis viral oncogene neuroblastoma derived homolog (N-myc) (15).

The WT1 gene includes ten exons (12) and can encode more than 36 different proteins depending on RNA edition (16). Several transcriptional changes can occur; however, two alternative splicings are prevalent, by means of which the four major isoforms are encoded (5). Two of these isoforms differ from each other due to the presence or absence of 17 amino acids (17 AA (+) and 17 AA (−)) encoded by exon 5, exclusively in mammals. This set of amino acids is added in the middle of the 17 AA isoform and is essential for the WT1 interaction with prostate apoptosis response factor 4 (Par-4), a transcriptional coactivator (12).

The other two isoforms are differentiated due to the inclusion or exclusion of amino acids lysine, threonine and serine (KTS) (KTS (+) and KTS (−)), which are located between the third and fourth zinc-finger binding site (17), encoded by exons 9 and 10, respectively (18). The normal ratio of WT1 isoforms is 2 KTS (+):1KTS (−) and this can be considered stable during fetal development and throughout life (17).

It was shown that, among the four predominant isoforms, the 17 AA (+) KTS (+) and 17 AA (+) KTS (−) were the ones most expressed during primary leukemia and exerted its anti-apoptotic function by stabilizing the potential of mitochondrial membrane in leukemia cells; although, the expression of these isoforms was reduced in solid tumors (5).

In adult humans, WT1 expression is very low and limited to some tissues, including kidney (podocytes), ovaries (granulosa cells), testis (Sertoli cells), spleen, myoepithelial progenitor cells and normal hematopoietic progenitor cells (16, 19). It can be further expressed in the central nervous system and hematopoiesis-related tissues, such as BM and lymph nodes (20).

Indeed, WT1 is expressed at low levels in normal BM and restricted to CD34⁺ cell population (20). Notwithstanding, it is suggested that WT1 contributes to development of blood cells, since its expression occurs in early hematopoietic precursors and decreases after differentiation (21).

WT1 and Acute Leukemia

Even though the WT1 gene was originally defined as a tumor suppressor gene, it has been proposed that it holds oncogenic properties in leukemogenesis and tumorigenesis (6, 22). This hypothesis can be justified for several reasons: it was observed that expression is up-regulated in acute myeloid leukemia (AML) and in blast crisis of chronic myeloid leukemia (CML) (18); treatment with antisense-WT1 oligomers (3) and small-hairpin RNA (WT1-shRNA) (5) promoted growth inhibition of leukemic cell and solid tumors expressing WT1 (23); constitutive WT1 expression in myeloid progenitor cells blocked their differentiation; however, in response to granulocyte colony-stimulating factor (G-CSF), it induces proliferation (6). Moreover, inhibition of WT1 expression resulted in suppressed growth of leukemic cells in vitro (22). Otherwise, it was observed that enhanced expression of WT1 promoted motility, suppressed apoptosis and induced leukemia in transgenic mice (3).

WT1 overexpression has been described in leukemias, mostly in AML, allowing for monitoring of minimal residual disease (MRD) (24). In addition, in recent years, some results have been reported involving acute lymphoid leukemia (ALL), although they are still controversial. In pediatric ALL, Boublikova et al. (11) demonstrated that T-ALL presented increased WT1 expression compared to B-ALL patients. However, these results were not confirmed in children or in adults (16, 25).

Busse et al. (16) demonstrated that WT1 expression levels at diagnosis have no prognostic impact on the clinical outcome in adult ALL. Furthermore, a sudden increase in WT1 expression in remission patients seems to be associated with an increased risk of relapse (26). These findings were also reported in pediatric patients (27). On the other hand, Boublikova et al. (11) reported that low WT1 expression can be associated with an increased risk of relapse.

Iranparast et al. (28) showed a strong correlation between WT1 and vascular endothelial growth factor (VEGF) expressions in leukemia patients, suggesting that WT1 may regulate neoplasia development and angiogenesis in acute blood cancer.

Considering the genetic mutations presented by ALL patients, it was observed that patients with chromosomal translocation t(4;11), harboring MLL-AF4 fusion proteins, had extremely high WT1 levels compared to other chromosomal abnormalities, such as t(12;21) - TEL-AML1 and t(9;22) - BCR-ABL (11). Contrariwise, Hu et al. (27) found a positive correlation between WT1 overexpression and BCR-ABL fusion gene, in both ALL and AML pediatric patients.

Therefore, WT1 mRNA could represent a biomarker for leukemic blast cells (6) and may serve as a quantitative marker in acute leukemias (19). Taken together, these findings highlight a promising target to immunotherapy in leukemias.

WT1 and Immunotherapy

Cancer immunotherapy is considered the fourth cancer therapy after the three major ones: surgery, chemotherapy and radiotherapy. For maximal efficacy, immunotherapy should be started as soon as possible after the diagnosis of cancer and continued as long as possible, so that surgery, chemotherapy and radiotherapy can be performed under conditions of enhanced cancer immunity (3).

Several tumoral antigens, recognized by T cells, have already been identified, based on protein expression pattern, and classified in a few groups (29): antigens resulting from mutations in proto-oncogenes and tumor suppressor genes, such as RAS and p53 proteins (30); tumor-specific shared antigens, such as MAGE-1 (31); overexpressed tumor antigens, such as HER-2/Neu (32); and antigens derived from oncogenic viruses, such as E7 oncoprotein of human papilloma virus (HPV) 16 (33).

Research in cancer immunology has advanced significantly after the characterization of these antigens that may represent potential candidates for cancer immunotherapy. In this regard, 75 representative antigens in cancer have been selected and studied according to the following criteria: therapeutic function; immunogenicity; function of oncogenic antigen; specificity; antigen-positive cells and expression levels; expression in stem cells; number of cancer patients who present the antigen; number of antigenic epitopes and cellular localization of expression of antigen. Cheever et al. (34) evaluated these antigens and, although none of them presents all ideal antigen characteristics, WT1 was ranked at the top. This result has initiated the development of immunotherapy using WT1 as the target antigen in cancer.

In general, the identification of WT1 as a tumor antigen is most often associated with acute leukemias, since several studies have shown increased expression in leukemic cells compared to normal hematopoietic cells. Indeed, the viability of leukemic cells can be partly maintained by WT1 expression (35).

Among the cancer immunotherapy strategies, vaccines represent a specific active therapy with massive development prospects. The therapy using WT1 as a vaccine aims to increase the number of specific B and T lymphocytes against the antigen present on tumor cells. The administration of a synthetic cytotoxic T lymphocyte (CTL) WT1 epitope, in combination with an adjuvant, was able to induce a WT1-specific CTL response. The adjuvant activates dendritic cells (DCs) in skin and the WT1 peptide fits into the groove of human leukocyte antigen (HLA) class I molecules on the cell surface. Then, this complex moves to lymph nodes, where they activate CD8⁺ T lymphocytes that recognize the complex (36).

Hence, the development of vaccines targeting WT1 antigen depends on the characterization of peptides associated with HLA molecules on the cell surface. In this context, different epitopes have been identified with this potential (23, 36).

In an experimental model, animals were sensitized with WT1 9-mer peptide, containing binding domains to HLA class I. It was verified that CTLs were able to destroy tumor cells expressing the antigen. In addition, animals immunized with WT1 peptide were immunogenic, responding successfully against WT1-expressing tumor cells (6).

Moreover, vaccinated animals that had been stimulated with WT1-positive tumor cells were significantly longer survivors than the animals that had been vaccinated with control plasmid or saline (35). Furthermore, HLA-A *0201-restricted CTLs, a type of HLA class I frequent in Caucasians, were produced in vitro to recognize WT1 9-mer peptide. It was found that isolated cells from CML patients, HLA-A *0201-negative in chronic phase, were not recognized by specific CTLs for this peptide.

Oka et al. (37) have also demonstrated that, through immunization of experimental animals, CTLs against the WT1 protein could discriminate differences in WT1 expression levels between abnormally WT1-overexpressing tumor cells and physiologically WT1-expressing normal cells, thus resulting in tumor cell death without damage to normal tissues.

Several studies in clinical trials have evaluated the safe and effective vaccination of various WT1 peptides in AML (38-40). This approach increased the immune system activation and response against tumor cells (22, 41). Sawada et al. (42) have used WT1 peptide vaccination in children with various types of malignancies and observed, in some patients, an increase in CD8⁺ T cells that recognized WT1 antigen and, consequently, an increase of interferon gamma (IFN-γ) in peripheral blood.

Although CTLs are the most effective immune elements against tumor cells, CD4⁺ T lymphocyte response is extremely important in the generation and maintenance of antitumor cytotoxic responses. In this context, several groups have investigated epitopes capable of inducing immune responses via class II major histocompatibility antigens (MHC) (43, 44).

Guo et al. (44) reported the possibility of inducing specific CD4⁺ T cells for WT1 20-mer peptide, with restricted cytotoxicity to HLA class II. In this regard, the CD4⁺ T cells, specific for WT1, played an important role against leukemic cells expressing WT1 and induced CD8⁺ CTL activity.

On the other hand, Brayer et al. (43) have used a combined vaccine capable of inducing responses via class I and II MHC, checking their safety, tolerability and immunogenicity. In a protocol involving patients with AML and myelodysplastic syndrome (MDS), the vaccine was well-tolerated and several patients showed clinical improvement; however, it was not possible to detect consistent WT1-specific immune responses.

Sera from patients with AML were tested against the WT1 antigen and 3 out of 18 patients had antibodies directed against the N-terminal region of the protein, suggesting that the immune system recognizes the WT1 antigen (45). These results were also observed in patients with ALL, CML and MDS. The authors detected IgM antibodies against WT1 in 54.8% of patients, compared to 16.2% of healthy controls; IgG antibodies were detected in a similar proportion (46).

Although more studies are needed to confirm the role of WT1 expression in clinical practice, Qi et al. (47) suggested that WT1 can also be a potential marker for predicting disease-free survival, relapse or relapse-free survival and progression-free survival in patients with solid tumors. In particular, the administration of natural or modified 9-mer peptide was able to induce regression of breast cancer with colon metastasis, as well as maintained free survival or disease recurrence. In addition, patients with lung cancer showed tumor mass reduction after the second dose of WT1 vaccine (48).

In order to improve the specific immune response to WT1 soluble antigen, adoptive immunotherapy has been proposed in acute leukemias.

WT1 in Adoptive Immunotherapy

Adoptive immunotherapy consists of isolation and infusion of cells with antitumor activity in patients. This property is acquired in ex vivo cultured cells, promoting expansion and modification, depending on the clinical design (49).

Immune response targeting of adoptively transferred cells is important to ensure effective and therapeutic action. Particularly, it is initially mandatory to search for tumoral or tumor-associated antigens that can be the therapeutic targets. Moreover, it is necessary to evaluate antigen immunogenicity and its physiological involvement, aiming treatment effectiveness and avoiding tissue toxicity. For this reason, the increased expression of WT1 in acute leukemias suggests a potential target for the adoptive therapy design.

It has been described that presentation of WT1 peptides via HLA-A*2402 and HLA-A*0201 is able to generate immune responses of CTLs (50, 51). Various WT1 peptides were reported to induce specific CTLs in a murine experimental model (45). Among tested peptides, the most explored is the WT1 9-mer, presented via HLA-A*2402, and the p126 (peptide sequence: RMFPNAPYL) WT1, presented via HLA-A*0201 (51, 52).

It is known that peptides binding with high affinity to HLA-A exhibit enhanced immunogenicity (53). Thus, modifications on WT1 peptides were proposed to increase its specificity and immunogenicity. Tsuboi et al. (54) found that the substitution of an amino acid at position 2 of WT1 9-mer peptide increased the specificity for HLA-A*2402 and, also, increased the efficiency of inducing a CTL response when compared with the unmodified peptide. In addition, amino-acid substitutions at HLA-A*0201-binding anchor positions in native 9-mer WT1 peptide have more efficiently induced WT1-specific HLA-A*0201-restricted CTL than the natural 9-mer WT1 peptides (55). Essentially, CTLs are very important cells in the immune response against tumor cells, with the activation of their effector mechanisms occurring by recognition of tumor antigen peptides presented via MHC class I (56).

In a clinical study (57), transplanted patients with relapsed acute leukemias, CML or MDS received 5 infusions of WT1-specific CTLs derived from matched donors. The results suggested absence of significant treatment-related toxicities and that CTLs could be mobilized to bone marrow and have anti-leukemic activity, due to reduction of leukemic blasts.

Another clinical trial (58) tested WT1-specific CTLs derived from matched donors, generated by DCs infected with adenovirus vector, which encodes a truncated WT1 protein by deletion of the N-terminal region (1-147aa), in a transplanted patient with AML. After four infusions, the presence of WT1-specific CTLs was observed and maintained until the last examination over 2 years. Unfortunately, this patient died due to graft-versus-host disease (GVHD).

Ochi et al. (59) produced a retroviral vector (WT1-silencer siTCR) with a specific alpha and beta T cell receptor (TCR) chains for WT1 that promoted the silencing of non-specific TCR chains for WT1 with small hairpin RNA (shRNA) that binds to the constant regions of endogenous TCR of CTLs. The modified human CTLs from leukemia patients successfully lysed autologous leukemia cells but not normal hematopoietic progenitor cells. In a mouse xenograft model, adoptively transferred modified CTLs exerted distinct anti-leukemia efficacy and did not inhibit human hematopoiesis, which have cells with low WT1 expression. Likewise, using the same methodology, no renal injury was observed in vivo mediated by modified CTLs (60). These findings were important for establishing the immunotherapy using this methodology, because of safety and efficiency.

In addition to these findings, Fujiwara et al. (61) have modified CD4⁺ T cells and CTLs using the same experimental design. They found that modified CD4⁺ T cells migrated to leukemia sites and, subsequently, mediated an attraction of WT1-specific CTLs via chemotaxis, resulting in leukemia suppression. Importantly, the presence of modified CD4⁺ T cells was correlated with longer survival and enhanced formation of memory T cells by modified CD8⁺ T cells. In this context, these authors suggested that adoptive immunotherapy, using modified CD4⁺ T cell and modified CTLs together, would be clinically advantageous for the treatment of leukemia.

Furthermore, Isao et al. (62) administered modified lymphocytes with a retroviral vector MS3-WT1-siTCR, which encodes a specific TCR for WT1 235-243 peptide in patients with AML and MDS. The preliminary results showed transient decline of peripheral abnormal cells that were MDS-related. They also observed a decrease of bone marrow blasts and no toxicity induced by therapy.

In addition, DCs have an important role in the tumor environment. DCs are known as the most potent professional antigen presenting cells (APCs), inducing tumor-specific effector T cells, which can reduce the tumor mass specifically, maintaining innate and adoptive immune responses and promoting immunological memory. In this regard, the use of DCs for the WT1-specific CTLs generation has been assured as an efficient methodology in adoptive immunotherapy (63, 64).

In a clinical phase II study, autologous WT1 mRNA-electroporated-DCs were administered in 10 patients with AML at high risk of full relapse, treated with polychemotherapy, in order to observe its anti-leukemic effect. Some patients reached complete and some partial remission. The activation of both innate and adaptive immune responses was also noted (65). However, the production of such cells is time consuming and strenuous to be carried out on a large scale.

Sundarasetty et al. (66) evaluated the effect of smart dendritic cells (SmartDCs, which means “Self differentiated myeloid-derived antigen presenting cells reactive against tumor DCs”), produced using a vector expressing truncated WT1, stimulated with granulocyte-macrophage colony-stimulating factor (GM-CSF) and infused in mice. Noteworthy, the production of SmartDCs is faster and such cells have higher viability, biodistribution and specific immune response than conventional DCs (67, 68). In this experiment, they observed that the WT1-specific SmartDCs are able to attract WT1 specific human CTLs in vivo. In another setting, ex vivo combination of human CTLs and SmartDCs inhibited tumor growth by activating the specific response of CTLs to WT1, suggesting a potential use of these cells in leukemia adoptive immunotherapies (66).

Another in vitro study, using electroporated DCs with IFN-α and WT1 mRNA, successfully demonstrated a stimulation of antigen-specific CTLs and enhancement of natural killer (NK) cells effector functions, thereby activating the innate and adaptive immunity (69).

Thus, for maintenance of the CTLs population, administration of modified DCs that present WT1 epitopes represents a promising and exciting approach in modern anticancer therapy as it can provide a glimpse into the future of acute leukemia treatment involving adoptive immunotherapy.