DCE-MRI assessment of the effect of Epstein-Barr virus-encoded latent membrane protein-1 targeted DNAzyme on tumor vasculature in patients with nasopharyngeal carcinomas

Nasopharyngeal carcinoma (NPC) is a common malignancy in southern China. Currently, radiation therapy is the preferred treatment for this disease. With the emergence of intensity-modulated radiation therapy (IMRT), local control rates for tumors have significantly increased [1]. However, increased radiosensitivity, reduced radiation doses, and improved quality of life for NPC patients remain active areas of research. In particular, receptor-targeted molecular therapy, as well as the identification of key genes, regulatory elements, and kinases in NPC cells, are being studied.

Epstein-Barr virus (EBV) has been implicated in the pathogenesis of NPC, and EBV-encoded latent membrane protein 1 (EBV-LMP1)) has been detected in 90% of NPC clinical samples [2]. EBV-LMP1 is an important oncogenic protein that has been shown to engage a plethora of signaling pathways that involve nuclear factor kappa B (NFκB), c-Jun N-terminal kinase (JNK), c-Jun activator protein 1 (AP1), mitogen A activated protein kinase (p38a MAPK), activating transcriptional factor (ATF), Janus kinase (JAK) signal transducers, and activators of transcription protein (STAT). Moreover, constitutive activation of these pathways appears to be central to the ability of LMP1 to induce multiple morphological and phenotypic alterations [3]. For example, EBV-LMP1 is involved in the transformation, proliferation, apoptosis, and differentiation of NPC cells, and has also exhibited a positive effect on tumor angiogenesis by directly up-regulating expression of vascular endothelial growth factor (VEGF) via the Stat 3 transcription factor [4]. Therefore, it has been hypothesized that blocking of EBV-LMP1 represents a novel targeted molecular therapy for NPC [5].

Deoxyribozymes (DNAzymes) are single-stranded DNA fragments that have been selected to have catalytic functions and ribozyme activity using in vitro molecular evolution techniques. DNAzymes have the ability to cleave RNA with high efficiency [6], and therefore, are able to inhibit gene expression at the mRNA level and regulate expression of target proteins [7, 8]. Correspondingly, DNAzymes are potentially applicable to gene inactivation strategies [9, 10]. Previously, a 33-mer oligonucleotide LMP1-targeted DNAzyme containing three phosphorothioate linkages at its 5’ and 3’ ends was developed to specifically target LMP1 mRNA [11]. Down-regulation of LMP1 expression using this LMP1-targeted DNAzyme was found to inhibit the growth of NPC cells both in vivo and in vitro by suppressing cell proliferation and inducing apoptosis [2, 12]. Moreover, in a recent study, LMP1-targeted DNAzyme was found to enhance the radiosensitivity of LMP1-positive NPC cells by inhibiting telomerase activity [13].

Conventionally, prospective, random, and well-controlled double-blind trials have been the “gold standard” for evaluating drug safety and effectiveness. Unfortunately, however, this method has also resulted in the soaring costs associated with medical innovations. Consequently, it has been proposed that biomarkers could be used to provide initial evidence regarding drug efficacy and safety. In recent years, with advances in imaging technology, imaging biomarkers have been increasingly utilized to assess drug efficacy. Furthermore, newer imaging technologies have been able to provide functional information in addition to structural information for several diseases [14].

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) uses fast T1-weighted sequences to assess changes in signal intensity before, during, and after the intravenous administration of contrast agent (CA). The dynamic contrast images obtained are then used to quantitate parameters which characterize tumor microcirculation. For example, the volume constant for the transfer of CA from the plasma to the extravascular extracellular space (e.g., K ^trans minute^-1) represents an important parameter, and is primarily related to blood perfusion and microvessel permeability [15]. Over the past 10 years, DCE-MRI and extracted kinetic parameters have been applied to phase I and phase II clinical trials of anti-angiogenic drugs and vascular disrupting agents, with K ^trans generally recognized as a marker of tumor blood flow and permeability [15‐17]. In addition, K ^trans has been recommended as a primary endpoint for anti-cancer treatment trials by the U.S. National Cancer Institute [18].

Although EBV LMP1 has been shown to promote tumor angiogenesis [4], it remains unknown whether an LMP1-targeted DNAzyme affects NPC vasculature. In the present study, an LMP1-targeted DNAzyme was injected locally into NPC tissues in conjunction with radical radiation therapy and DCE-MRI was used to evaluate the efficacy of this treatment for NPC patients. To the best of our knowledge, this study is the first to examine the effects of an LMP1-targeted DNAzyme using DCE-MRI in vivo.

To validate whether DCE-MRI can monitor the therapeutic efficacy of this treatment approach for NPC patients, parameters of vascular permeability and intra- and extravascular volumes were subsequently obtained (including K ^trans, k _ep, v _e). Prior to radiotherapy, there were no statistically significant differences in the K ^trans for the tumor tissues of the combined treatment group versus the tumor tissues of the radiotherapy alone group (t-test, P =0.175). However, for both groups, the 95% confidence interval (CI) of the K ^trans values declined gradually from the Pre-RT timepoint to the 3 months post-RT timepoint (Figure 1A). Moreover, for the combined treatment group, statistically significant differences in the K ^trans values were observed between the Pre-RT timepoint and the RT (50 Gy), RT (70 Gy), and 3 months post-RT timepoints. In particular, the latter values were lower than the K ^trans value for the Pre-RT timepoint. In contrast, the radiotherapy alone group exhibited no significant differences in the K ^trans values between the Pre-RT and the RT (50 Gy) and RT (70 Gy) timepoints. However, there was a statistically significant difference between the K ^trans values for the Pre-RT and 3 months post-RT timepoints for the radiotherapy alone group (Table 2). Representative K ^trans maps are shown in Figures 2 and 3.

There were no statistically significant differences in the k _ep values for tumor tissues between the combined treatment group and the radiotherapy alone group before radiotherapy (t-test, P =0.770). However, for both groups, the k _ep values gradually declined following therapy. The 95% CI for the k_ep values are shown in Figure 1B. When k _ep values were compared for the four timepoints, statistically significant differences were observed between the Pre-RT timepoint and the RT (50 Gy), RT (70 Gy), and 3 months post-RT timepoints for the combined treatment group, and the radiotherapy alone group. In particular, the latter values were lower than the k _ep value for the Pre-RT timepoint (Table 3). Representative k _ep maps are shown in Figures 2 and 3.

Statistically significant differences were observed between the v _e values for the Pre-RT timepoint and the RT (50 Gy), RT (70 Gy), and 3 months post-RT timepoints for all tumor tissues. Furthermore, the latter values were higher than the v _e value for the Pre-RT timepoint (Table 4).

There were no adverse events that could be attributed to LMP1-targeted DNAzyme injections. Furthermore, analyses of white blood cell number, hemoglobin concentration, platelet number, and lymphocyte cell number showed no significant differences between the combined treatment group and the radiotherapy alone group. There were also no significant differences in liver or renal function, or in impairment of skin, mucous membranes, or salivary gland, between the two treatment groups.

Studies have shown that DCE-MRI can be used to monitor the efficacy of various treatments and to predict response to treatment. In particular, this has been demonstrated for neoadjuvant chemotherapy and radiation therapy for bladder cancer, breast cancer, and osteosarcomas [21, 22]. Over the past 10 years, studies have also shown that DCE-MRI and extracted kinetic parameters can be used as an in vivo cancer imaging tool for the diagnosis, monitoring of treatment effect, and evaluation of anti-cancer drugs. Correspondingly, this method has been applied to phase I and phase II clinical trials of anti-angiogenic drugs and vascular disrupting agents [23, 24]. In particular, K ^trans is currently recognized as a general marker of tumor blood flow [15‐17], and has been recommended as a primary endpoint for an anti-cancer treatment trial conducted by the U.S. National Cancer Institute [18]. DCE-MRI has been widely used in the development of anti-angiogenic drugs, and has detected the efficacy of drugs earlier than conventionally observed changes in tumor volume. This is particularly beneficial for the selection of individualized patient treatment plans for patients that are diagnosed in the early stages of disease [25]. In a summary by O'Connor et al. describing their experience with DCE-MRI for the early clinical development of vascular-directed anticancer therapies over the past decade, they demonstrated that perfusion imaging provides unique information regarding the vascular properties of tumors, and for tumor responses to antiangiogenic agents and VDAs in pre-clinical and early clinical studies [15]. DCE-MRI has also been used as a biomarker for certain chemotherapy drugs in order to separate the biological effects of chemotherapy from those related to angiogenesis inhibitors [26].

To investigate the effects of an EBV LMP1-targeted DNAzyme on NPC vasculature in the present study, two methods were compared (e.g., radiation therapy in combination with a LMP1-targeted DNAzyme and radiation therapy alone) using DCE-MRI. It was observed that the K ^trans values for NPC tumor tissues decreased for both treatment groups over the course of treatment. K ^trans is a factor that accounts for the complex functions of blood flow, endothelial surface area, and endothelial permeability. However, it can also have multiple physiologic interpretations depending on the balance between capillary permeability and blood flow in a tissue. In high-permeability situations (where flux across the endothelium is flow limited), K ^trans is equal to the blood plasma flow per unit volume of tissue. Conversely, for low permeability conditions where tracer flux is permeability limited, the K ^trans value is equal to the permeability surface area product per unit volume of tissue [19]. NPC tumor tissues are characterized by extensive angiogenesis, an incomplete vascular basement membrane, and high permeability. Thus, NPC tumor tissues exhibit high permeability conditions, with leakage of CA to the EES being rapid and of great quantity. Accordingly, the K ^trans value for NPC tissue is very high. In contrast, muscle tissue, fibrous tissue, normal brain tissue, and tumor tissues in the late stages of chemotherapy or radiation therapy are characterized by low permeability. The K ^trans value for these tissue types is dependent on cross-vessel wall transport, and it is significantly lower than the K ^trans value for tumor tissues [27, 28]. Swelling, degeneration, and necrosis of tumor capillary walls, as well as narrowing of vessel lumen and thrombosis due to radiotherapy, also contribute to decreased permeability and reduced perfusion, thereby resulting in a lower K ^trans value. Many studies have reported on the prognostic value of K ^trans and other DCE-MRI parameters in assessing the response to radiotherapy, which, however, demonstrated contrary findings. Longitudinal study and repeated DCE-MRI may reduce these discrepancies, and be recommended [29].

A decrease in the K ^trans values for the NPC tumor tissues was observed significantly earlier (e.g., at the RT 50 Gy timepoint) in the combined treatment group than in the radiotherapy alone group (3 months post-RT). These results suggest that LMP1-targeted DNAzymes accelerate the decline of the K ^trans value for NPC. It has previously been shown that EBV LMP1 promotes tumor angiogenesis by up-regulating VEGF expression via activation of Stat 3 transcriptional factor [3]. Although the relationship between DCE-MRI and VEGF remains unclear, a positive correlation between levels of soluble vascular endothelial growth factor recptor-1 (sVEGFR-1) and sVEGFR-2 in plasma and K ^trans for patients with advanced ovarian cancer were observed in a previous study [30]. Taken together, these observations suggest that LMP1-targeted DNAzyme treatment affects the angiogenesis and microvascular permeability of NPC, and this can be proved with further pathology and molecular biology studies of NPC.

In cancer treatment trials, k _ep and v _e values have been recommended as secondary endpoints [31]. In the present study, tumor tissues of both the combined treatment group and the radiotherapy alone group had k _ep values that declined following therapy due to the slower transfer of CA from the EES to plasma as a result of decreased permeability. This similar trend in k _ep values for both groups suggests that the decrease is mainly related to the radiation treatments. In contrast, v _e values for all of the tumor tissues increased following radiotherapy, and this is consistent with an increase in EES secondary to the loss of tumor cells with radiation exposure. Moreover, continuous accumulation of CA in the EES is consistent with a slower transfer of CA from EES to plasma post-therapy [31, 32].

There were limitations associated with the present study. First, since surgery is not the preferred treatment for NPC, a comparison of pathology and DCE-MRI was not conducted. Secondly, the majority of the tumor lesions examined exhibited reduced tumor volume and disappeared quickly following treatment. Consequently, there was a lack of DCE-MRI data after the first year following treatment. Lastly, the parameters examined were based on the use of a tracer-kinetic model, which assumes that transcytolemmal water exchange contributes negligibly to changes in signal intensity [33]. Therefore, the amount of CA in tissues could be overestimated.

The use of an EBV LMP1-targeted DNAzyme was found to accelerate the decline of K ^trans values for NPC tissues, thereby suggesting that this DNAzyme affects the angiogenesis and microvascular permeability of NPC in vivo.