Oncolytic therapy has reportedly excellent efficacy in the treatment of many tumors, such as lymphoma, ovarian cancer, mesothelioma, breast cancer and hepatocellular carcinoma [9
]. Specifically, there are many advantages in using oncolytic therapy to treat liver malignancies; at some medical centers, oncolytic therapy has even been evaluated in clinical trials [37
]. As a special subtype of hepatocellular carcinoma, HB originates from the liver embryonic tissues and has the potential for diverse differentiation. Many components, such as the epithelium, bone and cartilage, can be included within the tumor; as a result, HB differs from general hepatocellular carcinoma in its histological and the biological characteristics. Histologically, the tumors are divided into epithelial and mixed epithelial/mesenchymal subtypes. Tumor cells may appear with a wide variety of characteristics ranging from almost liver-cell-like to undifferentiated blastomal cells. The majority of HB cells are epithelial, consisting of embryonal and fetal cells. About 5% of the tumors belong to the small-cell undifferentiated subtype, which is associated with a worse prognosis [5
]. Despite many advances in the use of oncolytic therapies for other liver malignancies, little is known about the use of oncolytic therapy in human HB. In this study, we report the potent therapeutic efficacy of oncolytic virus against human HB cells.
Currently, a large variety of oncolytic viruses are under evaluation in clinical trials [16
]; the most common viruses tested are derived from the attenuated Edmonston vaccine strain of the measles virus [9
], adenovirus [16
], herpes simplex virus (HSV) [18
], Newcastle disease virus (NDV) [20
], parvovirus [21
], and poliovirus [22
]. The present results demonstrate that virotherapeutics can be used safely and efficiently as cancer therapies. For safety reasons, vaccine virus-derived virotherapeutics are of special interest (e.g., measles vaccine virotherapeutics) because they are approved for human use, have been applied millions of times with a longstanding excellent safety record, and exhibit a potent natural oncolytic activity [41
]. As one of the most tested vaccines, the attenuated Edmonston vaccine strain of the measles virus has been well reported in hepatocellular carcinoma but not in HB treatment. MV is a replicating virus; therefore, the MV-Edm vaccine strain derivatives of MV can offer the potential advantage of increased dissemination in the tumor and of potentially enhanced therapeutic benefit compared to non-replicating viral or non-viral vector systems.
The MV enters the cells through the interaction of the H-glycoprotein with the MV receptors, CD150 (signaling lymphocyte-activation molecule, SLAM) and CD46 [41
]. Of note, the wild-type measles virus enters more efficiently through the SLAM receptor, whereas the Edmonston vaccine strain of measles virus enters the cells predominantly through the CD46 receptor. The MV receptor CD46 (membrane cofactor protein) belongs to the family of membrane-associated complement regulatory proteins that serve as an important mechanism of self-protection against complement-mediated lysis. Tumor cells frequently overexpress CD46. These mechanisms contribute to the tumor selectivity of MV-Edm. The effectiveness of MV-Edm-mediated oncolysis is highly dependent upon the expression of the cellular attachment receptor CD46, which is expressed more frequently in human cancer cells than in normal cells. In this study, we have demonstrated that the measles virus vaccine strain derivative MV-CEA, which has been genetically engineered to produce CEA, has significant antitumor activity against HB as indicated by the CPE of MV-CEA on HB cell lines in vitro and the efficacy of MV-CEA in an HB xenograft model. Our data show that CD46 is overexpressed in HB cell lines compared to normal liver cells, and MV-CEA successfully infected human HB cells, resulting in transgene expression, syncytium formation, and tumor cell killing; therefore, we conclude that HB fulfills the requirements for viral uptake and selective cell fusion and killing.
It is likely that additional factors contribute to the MV-Edm tumor selectivity. In 2011, two independent groups reported the identification of a novel MV-Edm receptor, nectin 4 [44
]. It is a tumor cell marker found in breast, lung and ovarian carcinomas and rendered cells susceptible to the MV-Edm. The transient knockdown of nectin 4 with siRNA abolished the wild-type MV infection in these cell lines. Similar results were also confirmed in the MV-Edm. The binding of the V domain of nectin 4 to MV-H has been considered a potential mechanism for the MV pathogenicity [44
]. Also, a few studies have indicated that MV infection can occur via CD147 and virion-associated CypB, independent of MV-H [47
]. Watanabe et al [47
]. identified CypB as a binding partner of MV-N and further showed that the wild-type MV recognizes CD147 as a receptor on epithelial cells via the CypB that is incorporated into the virus particles. It is still not clear whether other MV strains, such as the MV-Edm, share a similar pathway. Although a variety of other receptors including H protein dependent or independent ones can be used for MV-Edm infection, it has been reported that their efficiency is far lower than CD46 [46
]. The CD46 receptor is still the main player for MV-Edm spreading between cells. Therefore, it is not necessary to detect all of the other receptors one by one when a high level of CD46 is sufficient for demonstrating the oncolytic mechanism.
MV-CEA resulted in a strong CPE in vitro and in vivo. In this study, two human HB cell lines, Hep2G and HUH6, were used. These two lines have features that are the most characteristic of human HB and have been widely used in HB-related investigations [40
]. Although the HB subtype and the risk level (either standard-risk or high-risk) induced by these two lines are still not verified, it is not critical for the oncolytic virus to be used in the HB biotherapy. Only cells that express high levels of the CD46 receptor are infected by MV-Edm and lytically killed. This statement was verified in the present study; both of the tested cell lines were susceptible to the cytotoxic effect of MV-CEA but differed in their cell death kinetics. The Hep2G cells were eliminated very efficiently and quickly, whereas the cytotoxic effect of MV-CEA on the HUH6 cells was observed later. The HB cell lines used in our study showed variable susceptibility to the cytotoxic effect of MV-CEA, most likely resembling the situation in primary human tumors, which are composed of heterogeneous tumor cell populations [5
]. Both cell lines express comparable levels of CD46 but differ in other aspects, such as their histological subtypes. Differences in the components of these two lines or other not-yet-identified factors in the process of measles-induced cytotoxicity could explain the differences in susceptibility.
As one of the most common causes of cell death, apoptosis has been well described in other malignancies. It has been reported that MV-Edm can induce apoptosis in both the tumor cells and the syncytia by a series of signal pathways, such as the Fas-associated death domain (FADD), protein kinase C (PKC), and the janus kinase-signal transducer and activator of transcription (JAK-STAT) signaling pathways [49
]. Similarly, we have used a variety of techniques that demonstrate extensive apoptosis after infection of MV-Edm in human HB cell lines [26
]. These results are in agreement with prior work that has indicated that apoptosis is the main mechanism of death for MV-induced syncytia [51
]. The exact mechanism of cell death after MV-Edm-induced syncytium formation is still unknown and should be further investigated.
One of the main advantages of MV-Edm therapy in human HB is that the tumor offers the possibility of localized treatments compared with other malignancies such as ovarian cancers, glioblastoma and prostate cancers. In most cases of HB, the tumor is located within the liver and does not have distant metastases after regular chemotherapy; also, the tumor is easily accessible using ultrasound or computed tomography (CT) guidance. HB tumors can easily be localized and injected with a therapeutic agent. In any case, the presence of anti-MV antibodies is not expected to significantly decrease efficacy. In a mouse model in which mice received passive transfer of anti-MV antibodies, Grote et al. found that intratumoral MV-Edm therapy of human lymphoma xenografts resulted in effective tumor regression without compromise through the presence of anti-MV antibodies [52
]. These findings were in accordance with the results of a study of intratumoral therapy with a retrovirus in immune-competent C3H mice and of a clinical phase II study of a genetically modified adenovirus in patients with advanced head and neck cancer [53
A major drawback of many cancer agents is the lack of convenient methods for monitoring the agent after administration to the patient. MV-Edm derivatives engineered to express CEA allow noninvasive tracking of the viral gene expression as well as localization of the infected tumor tissue. The advantage of this method has been well described in previous investigations [15
]. In this study, the same result was verified in HB both in vitro and in vivo. Given the fact that Hep2G is known to intrinsically produce CEA, these cells themselves may be the source of the CEA, which would impact the serum CEA. However, the impact on the final result would be minor because all of the comparisons were carried out within the Hep2G groups instead of between the Hep2G and other groups. In addition, in the clinical practice, fewer than 10% of patients with HB can produce CEA and the amount is very little; it is likely unnecessary to worry about the confusion of intrinsic CEA.