Background
Epithelial ovarian cancer (EOC) is the fifth leading cause of cancer-related deaths in women and is the most lethal of the gynecologic malignancies [
1]. The standard of care for newly diagnosed EOC patients is surgical debulking and administration of a platinum and taxane -based chemotherapy regimen, usually carboplatin and paclitaxel, given either as neo-adjuvant or adjuvant therapy. With this regimen, 80–90% will initially respond but less than 10–15% will remain in complete remission [
2,
3]. The percentage of non-responders increases significantly to 65–75% for recurrent cancers[
3]. Additionally, some patients progress during or shortly after completion of chemotherapy.
Recurrent ovarian cancer is characterized by chemoresistance to prior treatments, most commonly to Paclitaxel. Previously, we described the identification of a sub-population of EOC cells that are resistant to this agent. This sub-group of cells (Type I EOC cells) has a functional Toll Like Receptor-4-Myeloid Differentiation Protein 88- Nuclear factor κB (TLR-4/MyD88/NF-κB) pathway, and the ligation of TLR-4 by Paclitaxel (a known TLR-4 ligand) is able to induce NF-κB activation and secretion of pro-inflammatory and pro-tumor cytokines IL-6, IL-8, MCP-1, and GRO-α [
4,
5]. This response confers resistance to apoptosis, and more importantly, enhances tumor growth [
4]. In contrast, these events were not observed in the group of EOC cells that did not have a functional TLR4-MyD88 pathway (Type II EOC cells) and are sensitive to Paclitaxel.
The treatment of Type I EOC cells with Paclitaxel is not only ineffective in killing these cells, but more importantly, can be detrimental since it may enhance tumor growth. Therefore, the identification of potential new therapies for this specific cell population would be beneficial for the treatment of ovarian cancer patients.
ARRY-520 is an inhibitor of the mitotic kinesin, KSP. KSP inhibition prevents bipolar spindle formation leading to mitotic arrest and cell death [
6]. In studies comparing ARRY-520 with some of the more clinically advanced compounds and standard of care agents, ARRY-520 was shown to have superior efficacy in multiple xenograft models [
7] and is currently in a Phase I trial [
8]. More importantly, since KSP is expressed predominantly in proliferating cells and is absent from post-mitotic neurons, KSP inhibitors do not induce peripheral neuropathy usually observed with traditional microtubule disrupting agents such as Paclitaxel [
9]. The objective of this study is two-fold. First, to determine and characterize the anti-tumor activity of the KSP-inhibitor, ARRY-520, in EOC cells; and second, to determine whether it is effective against Type I EOC cells and therefore could be used as a substitute for Paclitaxel.
We demonstrate that ARRY-520 is able to promote cell death in EOC cells through an apoptosis mediated mechanism, involving caspase-2 activation. More importantly, we showed that contrary to Paclitaxel, ARRY-520 has no effect on the TLR4 pathway and does not induce the secretion of pro-inflammatory and pro-tumor cytokines in Type I EOC cells.
Discussion
We demonstrate in this study that the KSP inhibitor, ARRY-520, has similar anti-tumor activity in EOC cells compared to Paclitaxel. More importantly though, unlike Paclitaxel, ARRY-520 does not activate NF-κB and does not induce secretion of pro-tumor cytokines in Type I EOC cells. Therefore, ARRY-520 may represent an alternative to Paclitaxel in this subgroup of EOC cells.
KSP is a microtubule-associated motor protein, which is essential for centrosome separation, formation of a bipolar mitotic spindle, and proper segregation of sister chromatids during mitosis [
24]. Inhibition of KSP forms monopolar mitotic spindles and arrests cells at mitosis, which leads to cell death [
25,
26]. KSP inhibitors have been shown to exhibit antitumor activity and are currently in clinical trials [
7,
9]. Because KSP localizes to mitotic microtubules, KSP inhibitors function exclusively during mitosis and are therefore selective to mitotic cells. Indeed, KSP inhibitors are shown to spare post mitotic neurons and thus do not cause peripheral neuropathy, which is a major side effect observed in Paclitaxel treatment [
9]. In the present study, we showed an additional advantage for the use of the KSP inhibitor ARRY-520 over Paclitaxel, specifically in Type I EOC cells.
In the subgroup of EOC cells with a functional TLR-4/MyD88/NF-κB pathway, Paclitaxel treatment leads to proliferation and NF-κB activation [
4,
14]. The activation of NF-κB is a major component in cancer initiation and progression [
27] and plays a central role in the control of apoptosis, cell proliferation, and survival [
28,
29]. Animal models have further supported the link between NF-κB activation and cancer progression [
30]. The demonstration that Paclitaxel can bind to TLR4 [
31] and therefore activate NFκB could explain why we observe tumor growth during Paclitaxel treatment [
4]. The absence of NFκB activation after ARRY-520 treatment suggests that ARRY-520 may be a better treatment option in patient with Type I EOC cells.
Another important aspect associated with NF-κB activation is the potential effect on the immune system. We showed previously that in Type I EOC cells, Paclitaxel treatment is able to induce the secretion of the pro-inflammatory cytokines IL-6, IL-8, MCP-1, and GROα [
5,
14]. All of these cytokines have been shown to directly affect cancer cell survival and growth [
32,
33] and also have implications in the resulting immune response. Indeed, our group has shown that the secretion of these cytokines by the Type I EOC cells is able to modulate the type of cytokines produced by the monocyte-like THP-1 cell line [
34]
It was noted that the mice with xenografts obtained from either the Type I or Type II cell lines responded equally to both compounds. These results did not reflect those seen
in vitro where Type I EOC cells are more resistant to treatment. Our group recently reported the identification and characterization of the ovarian cancer stem cells using the cell surface marker, CD44 [
14]. In this report, we showed that CD44+ cells represent the specific cell population that has a functional TLR-4/MyD88/NF-κB pathway. Indeed injection of R182 cells in mice (which is > 90% CD44+ by flow cytometry pre-injection) resulted in s.c. tumors containing < 10% CD44+ positive cells [
14]. The differentiation of the R182 cells from Type I to Type II
in vivo may explain the equivalent chemoresponse observed from the two xenograft models.
It is important to emphasize that this response induced by Paclitaxel is not observed in all EOC cells, but is limited to a specific sub-group, the Type I EOC cells.
In summary, ARRY-520 may represent an alternative to Paclitaxel in Type I EOC cells. This suggests the importance of identifying the molecular phenotype of the tumor prior to the initiation of therapy.
Conclusion
Administration of Paclitaxel to patients with high percentage Type I cancer cells could have detrimental effects due to Paclitaxel-induced enhancement of NF-κB and ERK activities and cytokine production (e.g. IL-6), which promote chemoresistance and tumor progression. ARRY-520 has similar anti-tumor activity in EOC cells as that of Paclitaxel. However, unlike Paclitaxel, it does not induce these pro-tumor effects in Type I cells. Therefore, the KSP inhibitor ARRY-520 may represent an alternative to Paclitaxel in this subgroup of EOC patients.
Acknowledgements
This work was supported in part by NCI RO1CA118678. The KSP inhibitor ARRY-520 was provided by Array Biopharma, Boulder, CO. The authors would like to thank Ms. Paulomi Aldo and Ms. Irene Visintin for assistance in the experiments involving the xMAP technology, Ms. Jamie Green for editing and proofreading the manuscript, and the UAB Arthritis and Musculoskeletal Center flow cytometry core facility for providing the instrumentation for FACS analysis.
Competing interests
KK, YX, ET, GM, and AA do not have competing interests. RW is an employee of Array Biopharma.
Authors' contributions
KK and YX performed cell viability assays, western blots, and luciferase assays. ET performed the mitochondrial depolarization assay. RW performed the in vivo experiments. GM participated in the design of the study and helped to draft the manuscript. AA participated in the design, analysis, and coordination of the study and the final drafting of the manuscript. All authors have read and approved the final manuscript.