A combination of sorafenib and SC-43 is a synergistic SHP-1 agonist duo to advance hepatocellular carcinoma therapy
Tzu-I Chao a,1, Wei-Tien Tai a,b,1, Man-Hsin Hung c,d,1, Ming-Hsien Tsai a, Min-Hsuan Chen a,
Mao-Ju Chang a, Chung-Wai Shiau e, Kuen-Feng Chen a,b,*
a Department of Medical Research, National Taiwan University Hospital, Taipei, Taiwan
b National Center of Excellence for Clinical Trial and Research, National Taiwan University Hospital, Taipei, Taiwan
c Division of Hematology and Oncology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan
d Program in Molecular Medicine, School of Life Sciences, School of Medicine, National Yang-Ming University, Taipei, Taiwan
e Institute of Biopharmaceutical Sciences, National Yang-Ming University, Taipei, Taiwan

A R T I C L E I N F O A B S T R A C T

Article history:
Received 21 July 2015
Received in revised form 21 October 2015 Accepted 30 November 2015

Keywords: Sorafenib SC-43 HCC
SHP-1 STAT3
Sorafenib is the first and currently the only standard treatment for advanced hepatocellular carcinoma (HCC). We previously developed a sorafenib derivative SC-43, which exhibits much more enhanced anti- HCC activity than sorafenib and also promotes apoptosis in sorafenib-resistant HCC cells. Herein, a novel “sorafenib plus” combination therapy was developed by coupling sorafenib treatment with SC-43. Both sorafenib and SC-43 are proven Src homology region 2 domain containing phosphatase 1 (SHP-1) ago- nists. The combined actions of sorafenib and SC-43 enhanced SHP-1 activity, which was associated with diminished STAT3-related signals and stronger expression of apoptotic genes above that of either drug alone, culminating in increased cell death. Decreased p-STAT3 signaling and tumor size, as well as in- creased SHP-1 activity were observed in mice receiving the combination therapy in a subcutaneous HCC model. More reduced orthotopic HCC tumor size and prolonged survival were also observed in mice in the combination treatment arm compared to mice in either of the monotherapy arms. These results in the preclinical setting pave the way for further clinical studies to treat unresectable HCC.
© 2015 Elsevier Ireland Ltd. All rights reserved.

Introduction

HCC represents about 90% of primary liver cancers and ranks as the third cause of cancer-associated death globally [1]. Despite in- creased understanding of hepatocarcinogenesis, limited treatment options have been established to improve survival for patients suf- fering from unresectable, advanced HCC [2–4]. Sorafenib (Nexavar), which is an antiangiogenic multikinase inhibitor, is the first and cur- rently only approved medication for the treatment of advanced HCC [5–7]. Results from phase III clinical trials of sorafenib in ad- vanced HCC patients showed that sorafenib treatment significantly,

Abbreviations: HCC, hepatocellular carcinoma; SHP-1, Src homology region 2 domain containing phosphatase 1; STAT3, signal transducer and activator of tran- scription 3; p-STAT3, phosphorylated STAT3; Raf, rapidly accelerated fibrosarcoma; VEGFR, vascular endothelial growth factor receptor; PDGFR, platelet-derived growth factor receptor; PTP, protein tyrosine phosphatase; SH2, Src homology region 2; DMSO, dimethyl sulfoxide; DMEM, Dulbecco’s modified Eagles’ medium; PARP, Poly (ADP- ribose) polymerase; Mcl-1, myeloid cell leukemia 1; PI, propidium iodide; IACUC, Institutional Animal Care and Use Committee; CI, combination index; TKI, tyrosine kinase inhibitor.
* Corresponding author. Tel.: +886 2 23123456, ext: 63548; fax: +886 2 23225329.
E-mail address: [email protected] (K.-F. Chen).
1 These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.canlet.2015.11.039

0304-3835/© 2015 Elsevier Ireland Ltd. All rights reserved.
albeit marginally, improved overall survival and delayed time to progression [4,8]. In clinical practice, however, unsatisfactorily low tumor regression (around 2 to 3 percent) and median overall sur- vival (usually less than one year) are observed. In addition, sorafenib treatment at full dose (800 mg/day) induces a number of adverse drug reactions which prevent patients from continuing therapy. Thus, a reduced dose is often required [9–11]. These factors underscore the urgent need to optimize or develop an “add-on” strategy for use with sorafenib to further expand the early successes of employing sorafenib.
Attempts have been made to improve the efficacy of sorafenib- based HCC therapy by combining sorafenib, which may have low anti-HCC activity as a single agent, with other therapy regimens [12–20]. These coupled treatments are based on the assumption that the combination therapy, by targeting different signaling path- ways, may have an additive or synergistic effect against tumors. Results from both in vitro and in vivo experiments revealed that no improvement was observed using the combination of sorafenib and erlotinib, a direct inhibitor of epidermal growth factor receptor (EGFR) tyrosine kinase which has strong expression in most HCCs [16]. On the other hand, combining sorafenib with tegafur/uracil resulted in modest improvement of sorafenib efficacy [20]. Another report published by Abou-Alfa et al. claimed that the sorafenib plus doxorubicin combination therapy showed prolonged median time

to progression and improved survival [14]. It is noteworthy that the comparison is based on sorafenib plus doxorubicin versus doxoru- bicin plus placebo, instead of having sorafenib in both arms. However, regarding improving overall survival, the current established sorafenib-based regimen, either sorafenib monotherapy or a “sorafenib plus” combination therapy, leaves much room for improvement.
Sorafenib acts by inhibiting a wide range of molecular targets, including rapidly accelerated fibrosarcoma (Raf) kinases, vascular endothelial growth factor receptor (VEGFR), and platelet-derived growth factor receptor (PDGFR), among others. Previously, we dis- covered that SHP-1, which is a nonreceptor protein tyrosine phosphatase (PTP) that negatively regulates p-STAT3, is also a direct target of sorafenib [21]. Sorafenib interferes with the association between the catalytic PTP domain and the N-SH2 domain of SHP-1 leading to a conformational change and enhanced activity of SHP-1. Based on the finding that sorafenib relieves the autoinhibition of SHP-1, we synthesized and selected a sorafenib analog, SC-43, func- tioning as a stronger SHP-1 agonist than sorafenib [21]. SC-43 treatment resulted in weaker p-STAT3 signaling, enhanced apop- tosis and displayed better survival benefits, compared to sorafenib treatment in the orthotopic HCC model.
Here, we developed a novel “sorafenib plus” approach, by ex- erting two SHP-1 agonists, sorafenib and SC-43, to further boost SHP-1 phosphatase activity. This new “sorafenib plus” concept was applied to inhibit one signaling pathway, the STAT3 signaling pathway, aiming to completely eliminate STAT3-related pro- survival signals in HCC. Sorafenib and SC-43 act in a synergistic fashion, contributing to the increased apoptosis in HCC cells and better survival benefits. Moreover, the sorafenib plus SC-43 design potentially minimizes the dose of sorafenib. Hence, limited side effects can be expected. To the best of our knowledge, this is the first rational design to enhance phosphatase activity by employ- ing sorafenib and its derivative to treat advanced HCC.

Materials and methods

Cell viability

The human HCC cells lines were seeded at a density of 5 × 103 per cm2 in 96 well plates. Cells were treated with Sorafenib, SC-43, or Sorafenib+SC-43 for 72 hours at different pre-determined concentration. Then MTT assays were performed [22] to evaluate the cell viability after drug treatment.

Apoptosis analysis

The percentage of apoptotic sub-G1 fraction was evaluated by flow cytometry as described [23]. Sorafenib+SC-43, SC-43, or sorafenib-treated cells were collect- ed and fixed at −20 °C overnight in 70% ethanol solution before being stained with 20 g/ml propidium iodide (PI).

Animal models and experiments

All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the National Taiwan University. Nude mice (4–6 weeks of age) were purchased from the National Laboratory Animal Center (Taipei, Taiwan). For orthotopic transplantation of HCC cells (n = 5–6), mice were directly inoculated with luc2-expressing PLC5 cells (1 × 105 cells) into the left lobe of the liver. Mice were assigned to 4 groups, including vehicle, sorafenib only (10 mg/kg/day), SC-43 (1 mg/kg/day) and sorafenib+SC-43 (10:1 mg/kg/day). Drug administration was performed when the luciferase activity is detected. Tumor weights and tumor sizes were measured in a blinded manner. For the subcutaneous implantation (n = 8–10), 2 × 106 of HCC cells were suspended in 100 μl of serum-free medium supple- mented with 50% Matrigel (BD Biosciences, Bedofrd, MA) and subcutaneously injected in the dorsal flank of each mouse. Tumors were allowed to grow to 100–200 mm3. After that, mice were subdivided into 6 different treatment groups, receiving vehicle, sorafenib (10 mg/kg/day), SC-43 (1 or 5 mg/kg/day), sorafenib (10 mg/kg/day) + SC-
43 (5 mg/kg/day) or sorafenib (10 mg/kg/day) + SC-43 (1 mg/kg/day). Tumor dimensions were measured two times per week by utilizing calipers. Tumor sizes were determined using the following equation: 0.523 × (L) × (W) × (H), where “L”, “W”, “H” indicate length, width, and height, respectively.
Synergism evaluation

SC-43, sorafenib, or sorafenib+SC-43-treated cells were collected and the per- centage of apoptotic cells was evaluated by flow cytometry. The data acquired were used to determine drug synergism. The drug synergism was examined by using the method developed by Chou and Talalay [24,25]. The software, CompuSyn, (ComboSyn, Inc., Paramus, NJ) was used to calculate the combination index (CI). A CI value of less than one was defined as synergistic.

Statistical analysis

Values are shown as mean ± SD. Statistical analyses were performed by em- ploying independent samples t test in the SPSS for Windows software, version 12.0 (SPSS, Chicago, IL, USA). P values less than 0.05 were considered to be statistically significant.
Other methods, including chemicals, reagents, cell culture, and plasmid con- structs, are described in the supplementary materials section.

Results

To identify the most effective combination formula for the sorafenib+SC-43 regime, a variety of sorafenib concentrations (0.5, 0.75, 1, 2.5, 5, and 7.5 μM) were tested in combination with 0.5 μM of SC-43. The sorafenib plus SC-43 treatment lead to significantly reduced cell viability, in a dose-dependent manner, in all of the HCC cell lines investigated (Fig. 1A). Interestingly, the effect of the com- bination of sorafenib and SC-43 on HCC cytotoxicity was observed using a sorafenib concentration as low as 0.5 μM. Employing this relatively low dose of sorafenib, the sorafenib plus SC-43 combi- nation arm already exhibited an approximate twenty percent decrease in cell viability, compared to the sorafenib-only arm. Maximal tumor cell cytotoxicity can be reached, with a significant difference between sorafenib monotherapy and the sorafenib plus SC-43 combination, when sorafenib is at a dose of 5 μM in the com- bination group (a molar ratio of 10:1 between sorafenib and SC-43). Then, we evaluated the potency of this developed combination therapy on cell apoptosis (Fig. 1B). A marked increase in apoptotic cells was detected in the sorafenib+SC-43 groups, as evidenced by sub-G1 analysis using flow cytometry. It is noteworthy that PLC5 and HepG2 cells were relatively more resistant to the drugs used, at the level of apoptosis induction, compared to HA59T and SK-Hep1 cells. Compared to groups that received sorafenib alone, a combi- nation of 5 μM of sorafenib and 0.5 μΜ of SC-43 (sorafenib: SC-43 at 10:1) was the lowest dosage capable of significantly boosting cell apoptosis in all four tested HCC lines. To elucidate the enhanced apoptosis resulting from the interaction between sorafenib and SC- 43, we performed median effect analysis. As indicated in Fig. 2, the combination index (CI) values for synergism evaluation were less than 1 in both SK-Hep1 cells (Fig. 2A) and PLC5 cells (Fig. 2B). The CI values were smaller than 1 in HCC lines tested, indicating that sorafenib and SC-43 functioned in a synergistic manner to induce HCC cell apoptosis. Curve shift analyses [26] were also performed to further confirm the synergism (Supplementary Fig. S1). In both SK-Hep1 and PLC5 cells, the combined actions of sorafenib and SC- 43 led to synergistically increased apoptotic cells. A leftward shift was observed in the combination treatment curve (green), com- pared to the curves for both of the single drugs, an indication of synergism. Together, our data suggested that the combination of the two SHP-1 agonists, sorafenib and SC-43, contributed to the in- creased HCC cell death and synergistically promoted apoptosis.
Previously, we demonstrated that both sorafenib and SC-43 con-
tributed to anti-HCC activity by elevating SHP-1 activity and de- phosphorylating p-STAT3 [21]. Therefore, next we investigated the effect of combining sorafenib treatment with SC-43 on SHP-1 ac- tivity. As shown in Fig. 3A, the activity of SHP-1 increased in PLC5 cells treated with the sorafenib plus SC-43 combo, compared to cells treated with either sorafenib (10 μM) or SC-43 (1 μM) alone. P-STAT3 is a downstream target of SHP-1 phosphatase and can be inacti- vated by SHP-1 mediated de-phosphorylation. We then determined

Fig. 1. Sorafenib plus SC-43 treatment exhibits improved anti-HCC activity. (A) Cell viability evaluation of different types of HCC cells that received sorafenib-only treat- ment or sorafenib plus SC-43 combination treatment. Values are shown as mean ± SD (n = 8). (*, P < 0.05; **, P < 0.01) (B) Histogram of sub-G1 analysis. HCC cells were treated with sorafenib alone or sorafenib plus SC-43. The percentage of apoptotic cells was determined by sub-G1 analysis. Data are shown as mean ± SD (n ≥ 5). (*, P < 0.05; **, P < 0.01).

Fig. 2. Sorafenib plus SC-43 synergistically contributes to the increased HCC apoptosis. (A) The percentage of apoptotic SK-Hep1 cells was determined by sub-G1 analysis and the CI values were calculated to evaluate synergism. A CI value less than one was considered to be synergistic. (n = 5) (B) Synergism evaluation of PLC5 cells. (n = 6).

Fig. 3. The sorafenib plus SC-43 combination boosts SHP-1 activity and abolishes p-STAT3 signaling. (A) PLC5 cells were subjected to different treatments and SHP-1 ac- tivity was analyzed (n = 3). (*, P < 0.05; **, P < 0.01) (B) HCC cells that received different treatments were collected and western blots were performed to detect p-STAT3 and its downstream signals.

Fig. 4. The “SHP-1-STAT3 axis” plays a pivotal role in the sorafenib plus SC-43 treatment-mediated HCC apoptosis. (A) Overexpression of STAT3 rescued PLC5 cells from the combination treatment-induced apoptosis (n = 3). (***, P < 0.001) (B) Inhibition of SHP-1 by PTP inhibitor III restored STAT3 phosphorylation and alleviated the pro- grammed cell death. (n = 3). (***, P < 0.001) (C) Knockdown of SHP-1 expression by SHP-1 siRNA treatment reduced apoptosis. (n = 3). (*, P < 0.05).

the expression of P-STAT3 and its downstream signaling following sorafenib, SC-43, or sorafenib plus SC-43 combination treatments. As revealed in Fig. 3B we demonstrated that, compared to sorafenib treatment alone, coupling sorafenib with SC-43 further abolished STAT3 phosphorylation in all four HCC cell lines evaluated. While sorafenib alone indeed knocked down the expression of cyclin D1, Mcl-1, and survivin, the sorafenib plus SC-43 combination further reduced the expression of these molecules that are involved in cell cycle and survival. On the other hand, the substantial upregulation of cleaved caspase-3 and PARP, which are both involved in pro- grammed cell death, was detected in HCC cells receiving the combination treatment (Fig. 3B). Collectively, these data support the concept that coupling the activity of sorafenib and SC-43, the two SHP-1 agonists, amplified SHP-1 activity. Consequently, the expres- sion of P-STAT3, cyclin D1, Mcl-1 and survivin was diminished while the quantity of cleaved caspase 3 and PARP was considerably in- creased in HCC cells.
Next, we sought to confirm the role of the “SHP-1-STAT3 axis” in this unique SHP-1 agonist dyad–mediated HCC apoptosis. Sub- G1 population analysis demonstrated that overexpression of STAT3 restored cell viability of PLC5 cells receiving the combination treat- ment (Fig. 4A). Furthermore, adding PTP inhibitor III (a SHP-1 inhibitor) to PLC5 cells restored STAT3 phosphorylation and alle- viated the programmed cell death induced by sorafenib plus SC-43 treatment (Fig. 4B). In parallel, the expression of endogenous SHP-1 was knocked down by SHP-1 siRNA treatment, as presented in Fig. 4C. Compared to HCC cells treated with the sorafenib plus SC-43 combo, silencing SHP-1 expression before receiving the combination therapy increased P-STAT3 level and reduced cell apoptosis.
Our data from these in vitro experiments prompted us to monitor the efficacy of our combination approach in vivo. We thus treated subcutaneous PLC5-bearing HCC mice with sorafenib and SC-43. The combination of sorafenib (10 μM) with SC-43 (both 1 μM and 5 μM) contributed to significantly smaller tumor size compared to sorafenib monotherapy (10 μM) (Fig. 5A). It is particularly noteworthy that the efficacy of this newly developed combination therapy in- creased proportionally to the concentration of SC-43. SC-43 (5 μM)

supplemented with sorafenib (10 μM) had better anti-HCC activi- ty than SC-43 (1 μM) and sorafenib (10 μM). In addition, using a dosage of 5 μM SC-43 already showed more potent tumor- cytotoxicity than 10 μM of sorafenib, which is currently the standard treatment for advanced HCC (Fig. 5A). We measured the tumor weight to further confirm the efficacy of sorafenib plus the SC-43 regime (Fig. 5B). As illustrated in Fig. 5B, the tumor weights of mice receiving combination therapy are significantly less than the tumor weights of mice from the sorafenib-only group or mice from the vehicle group. Elevating the concentration of SC-43 from 1 μM to 5 μM in the combination therapy groups also increased therapeu- tic efficacy, in terms of tumor weight. Our novel “add-on” HCC therapy strategy is based on the concept of supplementing sorafenib with another SHP-1 agonist, SC-43. In this regard, we evaluated the activity of SHP-1 in PLC5 tumors from different groups of mice (Fig. 5C). A surge of SHP-1 activity was observed in the tumor samples of mice receiving combination therapy (sorafenib 10 μM plus SC-43 5 μM), compared to the SHP-1 activity of samples from the vehicle group and sorafenib-only group. Tumor samples were also collect- ed for immunoblot analysis. STAT3 phosphorylation at Tyr 705 diminished in tumor samples from the combination group, com- pared to that from samples in sorafenib monotherapy group (Fig. 5D). To further shed light on the efficacy of the sorafenib plus SC-43 combo, an orthotopic HCC model was also established by inject- ing Luc2-expressing PLC5 cells into the left hepatic lobe of anesthetized nude mice. The mice in the sorafenib (10 mg/kg/day) plus SC-43 (1 mg/kg/day) arm acquired significant survival benefit, in terms of tumor size (Fig. 6A) and percentage of survival (Fig. 6B). All of the mice that received sorafenib plus SC-43 treatment sur- vived at the end of the study, while dead mice were observed in the sorafenib monotherapy group starting from week 9. To quan- titatively evaluate the efficacy of the combination treatment, the luminescence associated with Luc2-expressing PLC5 cells was also measured (Fig. 6C). Significantly less tumor-associated lumines- cence was detected in mice that received the combination therapy, compared to mice in the sorafenib group. Lastly, no reduction in body
weight occurred in the combo group (Fig. 6D).

Fig. 5. Sorafenib plus SC-43 combination therapy exhibits increased SHP-1 activity and enhanced anti-HCC activity in a subcutaneous HCC model. (A) Effects of the com- bination therapy on tumor volume (n ≥ 6) (*, P < 0.05; **, P < 0.01; ***, P < 0.001). (B) Effects of the sorafenib plus SC-43 therapy on tumor weight (n ≥ 6) (*, P < 0.05; **, P < 0.01;
***, P < 0.001). (C) Sorafenib plus SC-43 raised SHP-1 activity in vivo (n = 3) (*, P < 0.05; **, P < 0.01). (D) The combination approach abolished P-STAT3 signals in vivo. SC43-1: SC-43 1 μM; SC43-5: SC-43 5 μM.

Discussion

HCC patients typically have poor prognosis and the majority of patients are ineligible for tumor resection since they usually present at the intermediate/advanced stage. Local treatments are viable [27], but the odds of recurrence are usually high. Hence, the develop- ment of an effective and systematic treatment is an urgent need.
The development of sorafenib, currently the only approved treat- ment for unresectable HCC, provides hope for patients with advanced HCC. Sorafenib is a multiple kinase inhibitor [28], and inhibits a
number of angiogenic factors including RAF-1 and VEGFR [29], among others. However, several clinical trials employing other anti- angiogenesis agents, such as brivanib and linifanib, to treat HCC have failed [30,31]. In addition, the potent antiangiogenic effect of sorafenib does not translate well to clinical treatment of HCC, as only a 2–3% overall response rate was observed in the SHARP and Asian-Pacific trials [4,8]. This phenomenon suggests that other un- recognized molecules and signaling pathways may play a more pivotal role in the sorafenib-induced anti-HCC effect. We previ- ously unveiled that SHP-1, a nonreceptor protein tyrosine

Fig. 6. The sorafenib and SC-43 combination treatment contributes to decreased tumor size and improved survival in an orthotopic HCC model. (A) The combination therapy led to tumor growth inhibition and provided survival benefits. PLC5/luc-2-bearing mice (n ≥ 5) were treated with vehicle, sorafenib only, SC-43 only, or sorafenib plus SC-43. Tumor size was monitored by using the in vivo imaging system (IVIS). (B) Effects of sorafenib plus SC-43 on survival (n ≥ 5). (C) Quantitative evaluation of the luminescence associated with Luc2-expressing PLC5 cells (n ≥ 4). (*, P < 0.05) (D) Body weight measurement of mice that received different treatments. (n ≥ 4). (*, P < 0.05).

phosphatase, is a direct target of sorafenib [21]. Sorafenib en- hances SHP-1 activity by hindering the association between the catalytic PTP domain and the N-SH2 domain of SHP-1 relieving its auto-inhibition. We therefore developed a sorafenib derivative, SC- 43, in which the amide group and the pyridine ring of sorafenib were substituted with phenyl cyanide. SC-43 exhibited more potent anti- HCC activity than sorafenib as indicated by lower IC50, noticeable apoptosis in sorafenib-resistant HCC cells, improved survival in both a subcutaneous model and orthotopic model, as well as enhanced SHP-1 activity [21].
In the current study we examined the preclinical efficacy of a rationally-designed combination treatment, by employing two SHP-1 agonists, sorafenib and SC-43, for HCC therapy. The combined actions of the SHP-1 agonist duo leaded to a marked increase in SHP-1 ac- tivity. Significantly enhanced HCC cell apoptosis observed in the combo group, compared to sorafenib-only treatment, suggested that sorafenib and SC-43 functioned in a synergistic manner to combat HCC. P-STAT3-related signals in the combination group were also noticeably lower than that in sorafenib-only group. Above all, the developed add-on strategy not only resulted in smaller tumor size than sorafenib-only treatment in the subcutaneous model, but also contributed to 100% survival in the orthotopic HCC model (Fig. 6). An interesting observation is that Fig. 3A shows no clear evidence indicating synergism regarding SHP-1 activity enhancement; however, Fig. 2 shows that the combined actions of sorafenib and SC-43 synergistically increased the percentage of apoptotic HCC cells, which contributed to a marked decrease in tumor volume. Hence, another target, other than SHP-1, may be involved in the combi- nation drug regime’s ability to promote cell apoptosis and reduce tumor size.
Attempts have been made to target HCC by combination of
sorafenib, a multikinase inhibitor, with chemotherapy agents or other agents for targeted therapy, aiming to circumvent the activation of compensatory pro-survival signals and drug resistance [29]. However, most of the sorafenib combination therapies developed so far was designed to additionally target another signaling pathway, in par- allel with the benefits acquired by using sorafenib. Our approach, on the other hand, employed two SHP-1 agonists, sorafenib and its derivative SC-43, therefore, focusing on maximizing one entity, SHP-1 activity. This approach minimizes not only the possibility of antag- onism between the drugs used, but also the side effects that may have been caused. Consequently, no reduction in body weight (Fig. 6), an indication of a better safety profile, has been observed in the com- bination therapy group. HCC is a complex tumor with aberrant phosphorylation and activation of different pathways [29]. Al- though efforts have been made to inhibit phosphorylation of molecules involved in HCC progression by employing kinase in- hibitors in parallel with sorafenib, removal of phosphate groups represents another strategy for HCC treatment. Our proof of concept trial by using two SHP-1 agonists to maximize SHP-1 phosphatase activity, while minimizing the possibility of antagonism between the drugs used or the side effects, leads to sustained p-STAT3 sig- naling inhibition, decreased HCC tumor size, and superior survival benefits, including 100% survival in the orthotopic HCC model. STAT3 signaling, which has been known to have an oncogenic role in breast cancer, colorectal cancer, non-small-cell lung cancer and prostate cancer [32,33], can be expected to be abolished in further sorafenib plus SC-43 experiments.
Sorafenib, a multikinase inhibitor, is the current standard treat-
ment for HCC. However, from the clinical point of view, there is still much need for improvement in this therapy. The overall survival of HCC patients receiving sorafenib is merely 3 months longer than that of patients receiving placebo control, raising legitimate con- cerns about sorafenib resistance. The early success gained by treating HCC with multikinase inhibitors may hence be further expanded by using phosphatase agonists to combat tyrosine kinase inhibitor
(TKI)-associated resistance. Our combination approach, which couples sorafenib with SC-43, a SHP-1 agonist which has been shown to promote marked apoptosis in sorafenib-resistant HCC cells [21], can therefore be expected to supplement sorafenib as a potential first-line HCC regime. In addition, adverse drug reactions, such as hand-foot skin reaction and nausea, sometimes prevent patients from taking the full dose of sorafenib. Hence, the efficacy of the treat- ment may be compromised. Our sorafenib plus SC-43 combination effectively enhanced SHP-1 activity and abolished p-STAT3 signal- ing, compared to using sorafenib alone. As a result, a lower dose sorafenib can be expected to be the optimal dose for such a com- binational therapy. Indeed, a significant difference in cell viability was observed by combination of 0.5 μM sorafenib, a relatively low dose, with SC-43, compared to sorafenib alone (Fig. 1). Supple- menting sorafenib with SC-43 can thus be expected to lower the amount of sorafenib needed while maintaining its anti-HCC activity. In conclusion, our novel “add-on” strategy which supplements the treatment of sorafenib with another SHP-1 agonist, SC-43, dis- played significantly elevated anti-HCC activity. The combined effects between the two SHP-1 agonists enhanced SHP-1 activity in vitro. Compared to HCC cells that received sorafenib, diminished STAT3 phosphorylation and increased amount of cleaved caspase-3 were observed in cells treated with sorafenib plus SC-43. Treating HCC cells with SHP-1 inhibitor or siRNA against SHP-1 alleviated the apop- tosis induced by the combination treatment further confirming the pivotal role of SHP-1 in this “sorafenib plus” setting. The in vivo data corroborated the in vitro findings, since enhanced SHP-1 activity and abolished P-STAT3 signals were detected in the subcutaneous tumor samples from mice that received combination therapy. The afore- mentioned sorafenib plus SC-43 regime contributed to smaller tumor size and provided superior survival benefit in both a subcutane- ous and orthotopic HCC model. Our innovative approach which employed two SHP-1 agonists, sorafenib and its analog SC-43, aiming to reduce STAT3 signaling, provides a solution for advanced HCC therapy. These promising results acquired in the preclinical stage warrant further development of SC-43 for potential clinical
applications.

Acknowledgements

This research was supported by grants MOST 103-2622-B-002- 006 and MOST 103-2325-B-002-016 from the Ministry of Science and Technology, Taiwan.

Conflict of interest

None.

Appendix: Supplementary material

Supplementary data to this article can be found online at doi:10.1016/j.canlet.2015.11.039.

References

A⦁ . Jemal, ⦁ F. ⦁ Bray, M.M. Center, J. Ferlay, E. Ward, D. Forman, Global cancer ⦁ s⦁ tatistics, CA Cancer J. Clin. 61 (2011)⦁ ⦁ 69–90.
J.M. Llovet, J. Bruix, Systematic review of randomized trials for unresectable ⦁ hepatocellular⦁ ⦁ carcinoma:⦁ ⦁ chemoembolization⦁ ⦁ improves⦁ ⦁ survival,⦁ ⦁ Hepatology ⦁ 37 (2003) 429–442.
P.M. Lopez, ⦁ A. ⦁ Villanueva, J.M. Llovet, Systematic review: evidence-based ⦁ management⦁ ⦁ of⦁ ⦁ hepatocellular⦁ ⦁ carcinoma–an⦁ ⦁ updated⦁ ⦁ analysis⦁ ⦁ of⦁ ⦁ randomized ⦁ cont⦁ rolled trials, Aliment. Pharmacol. Ther. 23 (2006)⦁ ⦁ 1535–1547.
J.M.⦁ ⦁ Llovet,⦁ ⦁ S.⦁ ⦁ Ricci,⦁ ⦁ V.⦁ ⦁ Mazzaferro,⦁ ⦁ P.⦁ ⦁ Hilgard,⦁ ⦁ E.⦁ ⦁ Gane,⦁ ⦁ J.F.⦁ ⦁ Blanc,⦁ ⦁ et⦁ ⦁ al.,⦁ ⦁ Sorafenib ⦁ in⦁ ⦁ advanced⦁ ⦁ hepatocellular⦁ ⦁ carcinoma,⦁ ⦁ N.⦁ ⦁ Engl.⦁ ⦁ J.⦁ ⦁ Med.⦁ ⦁ 359⦁ ⦁ (2008)⦁ ⦁ 378–390.
M.A.⦁ ⦁ Worns,⦁ ⦁ P.R.⦁ ⦁ Galle,⦁ ⦁ HCC⦁ ⦁ therapies–lessons⦁ ⦁ learned,⦁ ⦁ Nat.⦁ ⦁ Rev.⦁ ⦁ Gastroenterol. ⦁ Hepatol. 11 (2014) 447–452.
J. Bruix, M. Sherman, Management of hepatocellular carcinoma: an update, ⦁ Hepatology 53 (2011) 1020–1022.

European Association for the Study of the Liver, European Organisation for ⦁ Research and Treatment of Cancer, EASL-EORTC clinical practice guidelines: ⦁ management of hepatocellular carcinoma, J. Hepatol. 56 (2012)⦁ ⦁ 908–943.
A.L.⦁ ⦁ Cheng,⦁ ⦁ Y.K.⦁ ⦁ Kang,⦁ ⦁ Z.⦁ ⦁ Chen,⦁ ⦁ C.J.⦁ ⦁ Tsao,⦁ ⦁ S.⦁ ⦁ Qin,⦁ ⦁ J.S.⦁ ⦁ Kim,⦁ ⦁ et⦁ ⦁ al.,⦁ ⦁ Efficacy⦁ ⦁ and⦁ ⦁ safety ⦁ of⦁ ⦁ sorafenib⦁ ⦁ in⦁ ⦁ patients⦁ ⦁ in⦁ ⦁ the⦁ ⦁ Asia-Pacific⦁ ⦁ region⦁ ⦁ with⦁ ⦁ advanced⦁ ⦁ hepatocellular ⦁ carcinoma:⦁ ⦁ a⦁ ⦁ phase⦁ ⦁ III⦁ ⦁ randomised,⦁ ⦁ double-blind,⦁ ⦁ placebo-controlled⦁ ⦁ trial,⦁ ⦁ Lancet ⦁ Oncol.⦁ 10 (2009)⦁ ⦁ 25–34.
M. Iavarone, G. Cabibbo, ⦁ F. ⦁ Piscaglia, C. Zavaglia, ⦁ A. ⦁ Grieco, E. Villa, et al., ⦁ Field-practice study of sorafenib therapy for hepatocellular carcinoma: a ⦁ prospective⦁ ⦁ multicenter⦁ ⦁ study⦁ ⦁ in⦁ ⦁ Italy,⦁ ⦁ Hepatology⦁ ⦁ 54⦁ ⦁ (2011)⦁ ⦁ 2055–2063.
M. Morimoto, K. Numata, M. Kondo, H. Hidaka, J. Takada, A. Shibuya, et al., ⦁ Higher⦁ ⦁ discontinuation⦁ ⦁ and⦁ ⦁ lower⦁ ⦁ survival⦁ ⦁ rates⦁ ⦁ are⦁ ⦁ likely⦁ ⦁ in⦁ ⦁ elderly⦁ ⦁ Japanese ⦁ patients⦁ ⦁ with⦁ ⦁ advanced⦁ ⦁ hepatocellular⦁ ⦁ carcinoma⦁ ⦁ receiving⦁ ⦁ sorafenib,⦁ ⦁ Hepatol. ⦁ Res. 41 (2011) 296–302.
M. Morimoto, K. Numata, M. Kondo, S. Kobayashi, S. Ohkawa, H. Hidaka, et al., ⦁ Field⦁ ⦁ practice⦁ ⦁ study⦁ ⦁ of⦁ ⦁ half-dose⦁ ⦁ sorafenib⦁ ⦁ treatment⦁ ⦁ on⦁ ⦁ safety⦁ ⦁ and⦁ ⦁ efficacy⦁ ⦁ for ⦁ hepatocellular⦁ ⦁ carcinoma:⦁ ⦁ a⦁ ⦁ propensity⦁ ⦁ score⦁ ⦁ analysis,⦁ ⦁ Hepatol.⦁ ⦁ Res.⦁ ⦁ 45⦁ ⦁ (2015) ⦁ 279–287.
Y. ⦁ Chao, ⦁ Y.H. ⦁ Chung, G. Han, J.H. Yoon, J. Yang, J. Wang, et al., The combination ⦁ of⦁ ⦁ transcatheter⦁ ⦁ arterial⦁ ⦁ chemoembolization⦁ ⦁ and⦁ ⦁ sorafenib⦁ ⦁ is⦁ ⦁ well⦁ ⦁ tolerated⦁ ⦁ and ⦁ effective ⦁ in Asian patients with hepatocellular carcinoma: final results of the ⦁ START trial, Int. J. Cancer 136 (2015)⦁ ⦁ 1458–1467.
L. Liu, H. Chen, M. Wang, ⦁ Y. ⦁ Zhao, G. Cai, X. Qi, et al., Combination therapy of ⦁ sorafenib and TACE for unresectable HCC: a systematic review and meta- ⦁ ana⦁ lysis, PLoS ONE 9 (2014)⦁ ⦁ e91124.
G.K. Abou-Alfa, ⦁ P. ⦁ Johnson, J.J. Knox, M. Capanu, I. Davidenko, J. Lacava, et al., ⦁ Doxorubicin plus sorafenib vs doxorubicin alone in patients with advanced ⦁ hepatocellular⦁ ⦁ carcinoma:⦁ ⦁ a⦁ ⦁ randomized⦁ ⦁ trial,⦁ ⦁ JAMA⦁ ⦁ 304⦁ ⦁ (2010)⦁ ⦁ 2154–2160.
O. ⦁ Abdel-Rahman, ⦁ M. ⦁ Fouad, Sorafenib-based combination ⦁ as a first ⦁ line ⦁ treatment for advanced hepatocellular carcinoma: a systematic review of the ⦁ li⦁ terature, Crit. ⦁ Rev. ⦁ Oncol. Hematol 91 (2014)⦁ ⦁ 1–8.
W.⦁ ⦁ Sieghart,⦁ ⦁ M.⦁ ⦁ Pinter,⦁ ⦁ B.⦁ ⦁ Dauser,⦁ ⦁ N.⦁ ⦁ Rohr-Udilova,⦁ ⦁ A.C.⦁ ⦁ Piguet,⦁ ⦁ G.⦁ ⦁ Prager,⦁ ⦁ et⦁ ⦁ al., ⦁ Erlotinib⦁ ⦁ and⦁ ⦁ sorafenib⦁ ⦁ in⦁ ⦁ an⦁ ⦁ orthotopic⦁ ⦁ rat⦁ ⦁ model⦁ ⦁ of⦁ ⦁ hepatocellular⦁ ⦁ carcinoma, ⦁ J. Hepatol. 57 (2012)⦁ ⦁ 592–599.
A.X. Zhu, O. Rosmorduc, T.R. Evans, ⦁ P.J. ⦁ Ross, A. Santoro, F.J. Carrilho, et al., ⦁ SEARCH: a phase III, randomized, double-blind, placebo-controlled trial of ⦁ sorafenib plus erlotinib in patients with advanced hepatocellular carcinoma, ⦁ J. Clin. Oncol. 33 (2015)⦁ ⦁ 559–566.
C. Gomez-Martin, J. Bustamante, J.F. Castroagudin, M. Salcedo, E. Garralda, M. ⦁ Testillano, ⦁ et ⦁ al., Efficacy ⦁ and ⦁ safety ⦁ of ⦁ sorafenib ⦁ in ⦁ combination ⦁ with ⦁ mammalian ⦁ target ⦁ of ⦁ rapamycin ⦁ inhibitors ⦁ for ⦁ recurrent hepatocellular ⦁ carcinoma⦁ ⦁ after⦁ ⦁ liver⦁ ⦁ transplantation,⦁ ⦁ Liver⦁ ⦁ Transpl.⦁ ⦁ 18⦁ ⦁ (2012)⦁ ⦁ 45–52.
S.D. Prete, L. Montella, M. Caraglia, L. Maiorino, G. Cennamo, V. Montesarchio, ⦁ et al., Sorafenib plus octreotide is an effective and safe treatment in⦁ ⦁ advanced
hepatocellular carcinoma: multicenter phase II So.LAR. study, Cancer Chemother. Pharmacol. 66 (2010) 837–844.
C.H. Hsu, ⦁ Y.C. ⦁ Shen, Z.Z. Lin, ⦁ P.J. ⦁ Chen, ⦁ Y.Y. ⦁ Shao, ⦁ Y.H. ⦁ Ding, et al., Phase II study ⦁ of ⦁ combining sorafenib with metronomic tegafur/uracil ⦁ for ⦁ advanced ⦁ hepa⦁ tocellular carcinoma, J. Hepatol. 53 (2010)⦁ ⦁ 126–131.
W.T. ⦁ Tai, C.W. Shiau, ⦁ P.J. ⦁ Chen, ⦁ P.Y. ⦁ Chu, ⦁ H.P. ⦁ Huang, ⦁ C.Y. ⦁ Liu, et al., Discovery of ⦁ novel⦁ ⦁ Src⦁ ⦁ homology⦁ ⦁ region⦁ ⦁ 2⦁ ⦁ domain-containing⦁ ⦁ phosphatase⦁ ⦁ 1⦁ ⦁ agonists⦁ ⦁ from ⦁ sorafenib⦁ ⦁ for⦁ ⦁ the⦁ ⦁ treatment⦁ ⦁ of⦁ ⦁ hepatocellular⦁ ⦁ carcinoma,⦁ ⦁ Hepatology⦁ ⦁ 59⦁ ⦁ (2014) ⦁ 190–201.
W.T. ⦁ Tai, A.L. Cheng, C.W. Shiau, ⦁ C.Y. ⦁ Liu, C.H. Ko, M.W. Lin, et al., Dovitinib ⦁ induces apoptosis ⦁ and ⦁ overcomes sorafenib resistance ⦁ in ⦁ hepatocellular ⦁ carcinoma⦁ ⦁ through⦁ ⦁ SHP-1-mediated⦁ ⦁ inhibition⦁ ⦁ of⦁ ⦁ STAT3,⦁ ⦁ Mol.⦁ ⦁ Cancer⦁ ⦁ Ther.⦁ ⦁ 11 ⦁ (2012)⦁ 452–463.
W.T. ⦁ Tai, A.L. Cheng, C.W. Shiau, ⦁ H.P. ⦁ Huang, J.W. Huang, ⦁ P.J. ⦁ Chen, et al., Signal ⦁ transducer⦁ ⦁ and⦁ ⦁ activator⦁ ⦁ of⦁ ⦁ transcription⦁ ⦁ 3⦁ ⦁ is⦁ ⦁ a⦁ ⦁ major⦁ ⦁ kinase-independent⦁ ⦁ target ⦁ of⦁ ⦁ sorafenib⦁ ⦁ in⦁ ⦁ hepatocellular⦁ ⦁ carcinoma,⦁ ⦁ J.⦁ ⦁ Hepatol.⦁ ⦁ 55⦁ ⦁ (2011)⦁ ⦁ 1041–1048.
T.C. Chou, ⦁ P. ⦁ Talalay, Quantitative analysis of dose-effect relationships: the ⦁ combined ⦁ effects ⦁ of multiple drugs or enzyme inhibitors, ⦁ Adv. ⦁ Enzyme Regul. ⦁ 22 (1984) 27–55.
T.C.⦁ ⦁ Chou,⦁ ⦁ Drug⦁ ⦁ combination⦁ ⦁ studies⦁ ⦁ and⦁ ⦁ their⦁ ⦁ synergy⦁ ⦁ quantification⦁ ⦁ using⦁ ⦁ the ⦁ Chou⦁ -Talalay method, Cancer Res. 70 (2010)⦁ ⦁ 440–446.
L.⦁ ⦁ Zhao,⦁ ⦁ J.L.⦁ ⦁ Au,⦁ ⦁ M.G.⦁ ⦁ Wientjes,⦁ ⦁ Comparison⦁ ⦁ of⦁ ⦁ methods⦁ ⦁ for⦁ ⦁ evaluating⦁ ⦁ drug-drug ⦁ in⦁ teraction, Front. Biosci. (Elite Ed.) 2 (2010)⦁ ⦁ 241–249.
R.T.⦁ ⦁ Poon,⦁ ⦁ S.T.⦁ ⦁ Fan,⦁ ⦁ F.H.⦁ ⦁ Tsang,⦁ ⦁ J.⦁ ⦁ Wong,⦁ ⦁ Locoregional⦁ ⦁ therapies⦁ ⦁ for⦁ ⦁ hepatocellular ⦁ carcinoma: a critical review from the surgeon’s perspective, Ann. Surg. 235 ⦁ (2002)⦁ 466–486.
G. Ranieri, G. Gadaleta-Caldarola, V. Goffredo, R. Patruno, A. Mangia, A. Rizzo, ⦁ et al., Sorafenib (BAY 43-9006) in hepatocellular carcinoma patients: from ⦁ disc⦁ overy to clinical development, Curr. Med. Chem. 19 (2012)⦁ ⦁ 938–944.
J.M.⦁ ⦁ Llovet,⦁ ⦁ J.⦁ ⦁ Bruix,⦁ ⦁ Molecular⦁ ⦁ targeted⦁ ⦁ therapies⦁ ⦁ in⦁ ⦁ hepatocellular⦁ ⦁ carcinoma, ⦁ Hepa⦁ tology 48 (2008) 1312–1327.
J.M.⦁ ⦁ Llovet,⦁ ⦁ V.⦁ ⦁ Hernandez-Gea,⦁ ⦁ Hepatocellular⦁ ⦁ carcinoma:⦁ ⦁ reasons⦁ ⦁ for⦁ ⦁ phase⦁ ⦁ III ⦁ failure and novel perspectives on trial design, Clin. Cancer Res. 20 (2014) ⦁ 2072–2079.
J.M.⦁ ⦁ Llovet,⦁ ⦁ T.⦁ ⦁ Decaens,⦁ ⦁ J.L.⦁ ⦁ Raoul,⦁ ⦁ E.⦁ ⦁ Boucher,⦁ ⦁ M.⦁ ⦁ Kudo,⦁ ⦁ C.⦁ ⦁ Chang,⦁ ⦁ et⦁ ⦁ al.,⦁ ⦁ Brivanib ⦁ in patients with advanced hepatocellular carcinoma who were intolerant to ⦁ sorafenib⦁ ⦁ or⦁ ⦁ for⦁ ⦁ whom⦁ ⦁ sorafenib⦁ ⦁ failed:⦁ ⦁ results⦁ ⦁ from⦁ ⦁ the⦁ ⦁ randomized⦁ ⦁ phase⦁ ⦁ III ⦁ BRIS⦁ K-PS study, J. Clin. Oncol. 31 (2013)⦁ ⦁ 3509–3516.
K.S.⦁ ⦁ Siveen,⦁ ⦁ S.⦁ ⦁ Sikka,⦁ ⦁ R.⦁ ⦁ Surana,⦁ ⦁ X.⦁ ⦁ Dai,⦁ ⦁ J.⦁ ⦁ Zhang,⦁ ⦁ A.P.⦁ ⦁ Kumar,⦁ ⦁ et⦁ ⦁ al.,⦁ ⦁ Targeting⦁ ⦁ the ⦁ STAT3 ⦁ signaling pathway in cancer: role of synthetic and natural inhibitors, ⦁ Biochim. Biophys. Acta 2014 (1845)⦁ ⦁ 136–154.
H.⦁ ⦁ Yu,⦁ ⦁ H.⦁ ⦁ Lee,⦁ ⦁ A.⦁ ⦁ Herrmann,⦁ ⦁ R.⦁ ⦁ Buettner,⦁ ⦁ R.⦁ ⦁ Jove,⦁ ⦁ Revisiting⦁ STAT3⦁ ⦁ signalling⦁ ⦁ in ⦁ cancer: new and unexpected biological functions, Nat. ⦁ Rev. ⦁ Cancer 14 (2014) ⦁ 736–746.