Receptor tyrosine kinases in carcinogenesis.

Receptor tyrosine kinases (RTKs) are cell surface glycoproteins with enzymatic activity involved in the regulation of various important functions. In all-important physiological functions including differentiation, cell-cell interactions, survival, proliferation, metabolism, migration and signaling these receptors are the key players of regulation. Additionally, mutations of RTKs or their overexpression have been described in many human cancers and are being explored as a novel avenue for a new therapeutic approach. Some of the deregulated RTKs observed to be significantly affected in cancers included vascular endothelial growth factor receptor, epidermal growth factor receptor, fibroblast growth factor receptor, RTK-like orphan receptor 1 (ROR1) and the platelet-derived growth factor receptor. These deregulated RTKs offer attractive possibilities for the new anticancer therapeutic approach involving specific targeting by monoclonal antibodies as well as kinase. The present review aimed to highlight recent perspectives of RTK ROR1 in cancer.

Introduction

The receptor tyrosine kinase (RTK)-like orphan receptor 1 (ROR1) expression has been observed to be significantly elevated in various blood and solid malignancies (1–6). These high expressions of ROR1 during carcinogenesis have encouraged investigation of the functional advantage conferred by ROR1 and ROR1 for targeted therapy in cancer. ROR1 expression was initially identified in chronic lymphocytic leukemia and subsequently in other malignancies (7–9). In 2001, two independent gene-profiling studies identified high ROR1 expressions (45-fold increase) in CLL as compared to normal mature B-lymphocytes (10,11). ROR1 protein expression has been shown in CLL (12) and in other malignancies such as acute lymphocytic leukemia (ALL), breast cancer, renal cell carcinoma, melanoma, lung adenocarcinoma and other lymphoid and myeloid malignancies (13–20).

The number of ROR1 receptors on the surface of CLL cells was estimated to be in the range of 10,000/cells, which are sufficient to be targeted by monoclonal antibodies (21). ROR1 expression during CLL becomes further increased during disease progression. It has been observed that STAT3 is constitutive phosphorylated in CLL and has been shown to bind to the promoter region of ROR1 in CLL (22). Thus, STAT3 may be involved as a promoting factor in the expression of ROR1. Previous findings showed that ROR1 expression was also influenced by IL-6 in a STAT3 dose-dependent manner (23). Wnt5a is suggested to be one of the ligands proving favorable for ROR1 expression. Additionally, the above observation was confirmed by co-culturing of Wnt5a that resulted in increased survival of CLL cells in comparison to those without Wnt5a (24). On the other hand, in a similar study the Wnt5a-dependent survival of CLL cells was inhibited by ROR1 antisera (25). Previously, it was suggested that Wnt5a maintained phosphorylation of ROR1, thus favors its expression (15). ROR1 phosphorylation varied from patient to patient in CLL. High ROR1 phosphorylation intensity was found in progressive compared to non-progressive CLL. The same pattern was observed in other malignancies including breast, lung and ovarian cancer cells with an aggressive course (21). Furthermore, ROR1 expression at the protein level was significantly higher in aggressive tumors (15). Collectively these data suggested that the expression pattern of ROR1 is related to aggressiveness. ROR1 also showed an elevated expression during marginal zone lymphoma (MZL), multiple myeloma, follicular lymphoma (FL) and mantle cell lymphoma (MCL) at the gene and protein level. A high ROR1 expression was also observed in the cases of ALL (25). Consequently, ROR1 expression has been observed to be altered in various types of cancer. Thus, further studies are required to add to the present results and clarify the future perspectives of this factor.

Expression of ROR1 and associated factors in solid tumors

ROR1 expression was observed by staining in a recent study in various types of cancer including prostate, testicular, uterine, ovarian, lymphoma, adrenal and melanoma cancers (15). Strong ROR1 staining was found in 30% or greater of primary samples in the lung, colon and pancreatic cancers (22). In lung adenocarcinoma, TITF1 has been shown to regulate ROR1 expression. ROR1 upregulation was linked with the potentiation of epidermal growth factor (EGF) ligand-induced EGF receptor (EGFR) signaling, phosphorylation and activation of c-SRC (26).

ROR1 expression was also detected in all melanoma cell lines at mRNA level as well as protein level as assessed by western blotting and RT-PCR in addition to surface ROR1 staining by flow cytometry. Furthermore, knockdown of ROR1 in melanoma cell lines resulted in induction of apoptosis (21). ROR1 was shown to be phosphorylated at tyrosine and serine residues. Furthermore, monoclonal antibodies raised against ROR1 induced apoptosis. Another study on similar grounds observed ROR1 and ROR2 expression to be inversely expressed in melanoma cells and concluded that both ROR1 and ROR2 negatively regulate each other (17). Hypoxia also transfers ROR1 expression from moderate to aggressive in melanoma cells (17). Additionally, breast cancer cell lines with a strong ROR1 expression were more aggressive and invasive, but decreased in non-migrating cells. siRNA gene silencing therapy using specific ROR1 siRNA resulted in the downregulation of ROR1 expression in human breast cancer cell lines, where ROR1 was shown to activate PI3K-mediated AKT and CREB signaling pathways by interacting with casein kinase 1 (CK1). Wnt5a increased the survival of ROR1 expressing breast cancer cells, confirming the hypothesis of Wnt5a as a ligand for ROR1 (27).

ROR1 and EMT transition

High levels of ROR1 expression in patients and cell lines were linked to genes contributing to epithelial-mesenchymal transition (EMT) such as ZEB1 and vimentin and inversely related to adherent junction proteins. ROR1 expression was high in breast adenocarcinomas with a high level of EMT-associated genes and with a high capacity to metastasize. Silencing of ROR1 in the triple-negative breast cancer cell line MDA-MB321 by small hairpin (sh) RNA reduced in vitro cell migration as well as bone and lung foci xenografts (17). Knockdown of ROR1 in triple-negative breast cancer cells also reduced the EMT genes such as SNAI1, SNAI2, ZEB1 and vimentin (17). Similarly, ROR1-transfected MCF-7 cell lines showed a low level expression of adherent junction proteins E-cadherin and CK-19, which contribute to homing of cells at proliferation sites. However, ROR1 transfection did not change the expression levels of SNAI1, SNAI2 or vimentin.

ROR1 as therapeutic targets in cancer

ROR1, similar to other oncogenic RTKs, may be targeted in cancer. There are two main strategies; targeting the extracellular part of the receptor by monoclonal antibodies or by tyrosine kinase inhibitors directed against the intracellular kinase domain. Targeting the extracellular region of RTK by mAbs may disrupt the cytoplasmic kinase signaling by neutralization of the ligand, hampering the ligand binding, internalization of the receptor or by interacting with immune effectors targeting the tumor cells. Different RTKs HER-2, EGFR, vascular endothelial growth factor receptor (VEGFR) and VEGF and their ligands have been targeted in various types of cancer by mAbs. Trastumuzab was the first approved antibody against HER-2 for the treatment of breast cancer patients (28). Pertuzumab is another antibody approved against HER-2 to prevent dimerization of HER-2 with members of EGFR family. Similarly, tyrosine kinase inhibitors against the intracellular kinase domain in various type of cancer have been designed and approved for clinical use; gefitinib (29) and erlotinib (30) against EGFR and lapatinib (31) against HER2.

In a novel combination approach, the extracellular part of RTK is usually targeted by mAbs and the intracellular kinase domain by tyrosine kinase inhibitors. This combination has been shown to be synergistic in preventing the tumor growth and proliferation of cancer cells. Treatment of xenograft mice expressing HER-2 with trastuzumab and lapatinib resulted in significant tumor growth inhibition (21). Similarly, treatment of colon cancer cell lines expressing EGFR with cetuximab and gefitinib prevented proliferation and induced apoptosis (33). Targeting of HER-2 by trastuzumab and lapatinib has shown better clinical activity than either alone in HER-2 positive breast cancer patients (34).

Monoclonal antibodies against ROR1

The extracellular region of ROR1 contains Ig-like, CRD and KNG domains, which could be targeted by monoclonal antibodies. Monoclonal antibodies directed against these extracellular domains have been developed as potential therapeutic agents (35). The anti-ROR1 mAbs may kill cells by direct apoptosis or by activation of complement or immune effector cells. Most effective anti-ROR1 mAbs to induce significant apoptosis in CLL cells were those against the CRD and KNG domains. Anti-ROR1 CRD mAb also induced apoptosis in pancreatic cancer cell lines. De-phosphorylation of ROR1, the PI3Kδ, AKT and mTOR was also observed prior to apoptosis by the treatment of anti-ROR1 CRD mAb suggesting inhibition of downstream signaling but there was no effect on ERK and PKC proteins. These findings suggested that ROR1 signaling occurs via PI3K/AKT/mTOR axis in pancreatic cancer cell lines.

The anti-ROR1 mAbs also resulted in killing of melanoma cell lines by direct apoptosis as well as complement-dependent cytotoxicity (CDC) and antibody-dependent cellular cytotoxicity (ADCC) (18). ESTDAB081 and ESTDAB094 melanoma cell lines were resistant to direct apoptosis by anti-ROR1 mAbs alone but sensitive to apoptosis by anti-ROR mAbs mediated through CDC and ADCC. Knockdown of ROR1 through specific siRNA resulted in apoptosis of melanoma cell lines. The effects of anti-ROR mAbs in TCL1 transgenic mice expressing ROR1+/CD5+/B220low leukemic B cells as a model for in vivo studies were analyzed (36). Two anti-ROR1 antibodies, D10 and 4A5 against different epitopes produced different effects in vivo. D10 anti-ROR1 mAb reduced the phosphorylation of AKT, but 4A5 mAb did not. Leukemic cells were cleared from the blood and spleen of transgenic mice following intravenous injections of D10 mAb, but not by 4A5 (37). Cirmtuzumab, a humanized anti-ROR1 mAb killed tumor cells and was internalized by malignant cells. Cirmtuzumab with antibody-drug conjugates (ADCs) cleared ROR1 expressing CLL cells in xenografted mice as well as induced apoptosis of breast and pancreas cell lines in vitro.

Anti-ROR1 antibodies have also been utilized to deliver toxins. Immunotoxin (BT-1) from Pseudomonas exotoxin (PE38) conjugated with the variable fragments of an anti-ROR1 mAb showed a dose-dependent and selective binding to leukemic cells from CLL and MCL patients, followed by internalization of the immunotoxin and subsequent apoptosis in vitro (38). ROR1 expression has been shown to be associated with EMT of tumor cells as indicated above (27). Treatment with anti-ROR1 mAbs prevented metastasis by downregulation of proteins involved in cell motility and metastasis of breast cancer cell lines.

Immunotherapeutic strategies targeting ROR1

ROR1 is highly expressed in many malignancies, but not in healthy adult tissues. ROR1 as an oncofetal antigen may thus be recognized by the immune system. A humoral immune response against ROR1 in CLL patients was observed after vaccination with Ad-CD154 transduced CLL cells (24). Antibodies induced against ROR1 blocked the interaction of Wnt5a and ROR1 and reduced proliferation of CLL cells. Lenalidomide, an immune modulating drug treatment induced ROR1 antibodies in CLL patients indicating that ROR1 may be an immunodominant epitope (39). CLL patients may also mount a type 1 T cell response against ROR1 and a humoral response against ROR1 (40). Transgenic leukemic ROR1+ mice immunized with a ROR1 peptide produced high titers of anti-ROR1 antibodies that inhibited the engraftment of ROR1+ CLL cells (41). ROR1 may therefore act as a tumor antigen for vaccination in a similar way as HER2 derived vaccine in breast cancer patients (42).

T cells from healthy donors or CLL patients have also been genetically modified to express ROR1-CART targeting ROR1+ tumor cells. CD8+ T cells engineered to express specific ROR1-CART lysed CLL and MCL cells but not mature normal cells. ROR1-CARTs were shown be as potent as CD19-CARTs in an immune deficient mouse model of human MCL (43). Clinical trials on ROR1-CART are expected to start.

Tyrosine kinase inhibitors against ROR1

Most oncogenic RTKs are highly upregulated and activated in malignant cells but have no or low expression and activity in normal tissues (44). RTKs in tumors may also be targeted by TKIs against the intracellular kinase domain (45). Axl is an RTK constitutively expressed in CLL and by targeting the intracellular kinase domain by a specific inhibitor, R428, a robust apoptosis of CLL cells was induced in a dose- and time-dependent manner (46). The VEGFR inhibitors vatalanib and pazopanib decreased phosphorylation of the VEGF receptor and induced apoptosis of CLL cells in clinically achievable concentrations with a mild cytotoxic effect on healthy B cells (47). Tyrosine kinase inhibitors may also target the intracellular kinase domain of ROR1. Treatment of CLL cells with ROR1 TKI dephosphorylated ROR1. ROR1 TKI induced apoptosis in CLL cells via caspase activation, PARP cleavage and downregulation of Mcl-1 and Bcl-2. Oral administration of ROR1 TKI in xenografted transplanted NOD SCID mice reduced leukemic cells.

Conclusions

It can be concluded from the above that RTK ROR1 may be a promising therapeutic target for ROR1 TKI and mAbs in CLL and other malignancies.