Strategic combination therapy overcomes tyrosine kinase coactivation in adrenocortical carcinoma
Chi-Iou Lin, PhD,a Edward E. Whang, MD,a Jacob Moalem, MD,b and Daniel T. Ruan, MD,a Boston, MA, and Rochester, NY
Background. Coactivation of tyrosine kinase limits the efficacy of tyrosine kinase inhibitors. We hypothesized that a strategic combination therapy could overcome tyrosine kinase coactivation and compensatory oncogenic signaling in patients with adrenocortical carcinoma (ACC).
Methods. We profiled 88 tyrosine kinases before and after treatment with sunitinib in H295R and SW13 ACC cells. The effects of monotherapy and strategic combination regimens were determined by the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (ie, MTS) assay. Results. The minimum inhibitory concentrations (ICmin) of sunitinib quenched its primary targets: FLT-3, VEGFR-2, and RET. In contrast, ERK, HCK, Chk2, YES, CREB, MEK, MSK, p38, FGR, and AXL were hyperactivated. Monotherapy with sunitinib or PD98059 at their ICmin reduced proliferation by 23% and 19%, respectively, in H295R cells and by 25% and 24%, respectively, in SW13 cells. Sunitinib and PD98059 in combination decreased proliferation by 68% and 64% in H295R and in SW13 cells, re- spectively (P < . 05 versus monotherapy). The effects of combination treatment exceeded the sum of the effects observed with each individual agent alone.
Conclusion. We describe the first preclinical model to develop strategic combination therapy to overcome tyrosine kinase coactivation in ACC. Because many tyrosine kinase inhibitors are readily available, this model can be immediately tested in clinical trials for patients with advanced ACC. (Surgery 2012;152:1045-50.)
From the Department of Surgery,” Brigham and Women’s Hospital, Boston, MA; and Department of Surgery,b University of Rochester Medical Center, Rochester, NY
ADRENOCORTICAL CARCINOMA (ACC) is a rare but ag- gressive malignancy that accounts for 0.02% of cancers reported in the United States.1 Although long-term survival can be achieved after complete operative resection, approximately 70% of patients present in advanced stages that are not amenable to curative surgery. For patients with metastatic dis- ease or in whom complete resection is not feasible, operation is not indicated. Although systemic ther- apy is the principal therapy for advanced ACC, it is generally ineffective and does not improve overall survival.2
The adrenolytic agent mitotane is the first-line therapy for advanced ACC. Although mitotane
Supported by a Fish Family Junior Faculty Research Grant from the Department of Surgery at Brigham and Women’s Hospital. Accepted for publication August 20, 2012.
Reprint requests: Daniel T. Ruan, MD, Department of Surgery, Brigham and Women’s Hospital, 75 Francis St., Boston, MA 02115. E-mail: druan@partners.org.
0039-6060/$ - see front matter
@ 2012 Mosby, Inc. All rights reserved.
http://dx.doi.org/10.1016/j.surg.2012.08.035
treatment has not improved overall survival in clinical trials, it can decrease cortisol levels and provide some symptomatic relief for patients. However, mitotane therapy often is limited be- cause of drug toxicity or disease progression.3 Cy- totoxic drugs, such as cisplatin4 and irinotecan, have minimal efficacy in ACC. The highly resistant nature of ACC to conventional chemotherapeutics has prompted efforts to target the molecular de- fects that drive oncogenesis.
Sunitinib is tyrosine kinase inhibitor (TKI) that targets multiple pro-oncogenic-signaling path- ways. In a recent case report, treatment with sunitinib resulted in a partial response on radiog- raphy in a patient with ACC.6 Currently, a phase 2 study evaluating the efficacy of sunitinib mono- therapy as second-line therapy in advanced ACC is underway.
However, monotherapy with sunitinib in other cancer contexts generally is ineffective. In multiple studies investigators have found that monotherapy with TKIs results in the coactivation of compensa- tory oncogenic signaling pathways.7 We hypothe- sized that the characterization of compensatory
signaling activated by sunitinib treatment could enable development of rational combination regi- mens to overcome tyrosine kinase coactivation in ACC.
METHODS
Materials. Mouse anti-actin antibody, pan Ab-5, was purchased from Neomarker (Fremont, CA). Rabbit antihuman phospho-ERK1/2 Thr202/ Tyr204, extracellular signal-regulated kinase (ERK), phospho-FLT-3 Tyr591, FLT-3, phospho- RET Tyr905, RET antibodies were purchased from Cell Signaling (Beverly, MA). Secondary horseradish peroxidase-conjugated mouse and rabbit antibodies were obtained from Vector Lab- oratories (Burlingame, CA). PD98059, an ERK inhibitor, was obtained from Calbiochem (La Jolla, CA). Sunitinib was obtained from Selleck Chemi- cals LLC (Houston, TX).
Cell culture. The human ACC cell lines NCIH295R (H295R) and SW13, originally ob- tained from ATCC, were generous gifts from Dr Gary D. Hammer (University of Michigan Medical School, Ann Arbor, MI). H295R cells were cultured in complete media containing Dulbecco’s modi- fied Eagle’s medium/F12 (Invitrogen, Carlsbad, CA) supplemented with 100 U/mL penicillin, 100 mg/mL streptomycin, 250 µg/mL fungizone, 10% fetal bovine serum (Invitrogen), 2.5% ITS+ supplement (Sigma-Aldrich, St. Louis, MO). SW13 cells were cultured in complete media containing Dulbecco’s modified Eagle’s medium (Invitrogen) supplemented with 100 U/mL penicillin, 100 mg/ mL streptomycin, 250 µg/mL fungizone, and 10% fetal bovine serum. SW13 and H295R cells were used throughout the study between passages 3 and 11 and 7 to 18, respectively.
Western Blotting and receptor tyrosine kinase arrays. Lysates from ACC cells treated with suni- tinib were subjected to Western Blotting.8 Human Phospho-RTK Array (ARY001) and Human Phospho-Kinase Array (ARY003) kits (R&D Sys- tems, Minneapolis, MN) were used to assay the relative level of tyrosine phosphorylation. Two membranes (Membrane #1 and #2) in array ARY001 contained spotted antibodies correspond- ing to 42 distinct receptor tyrosine kinases (RTKs). One membrane (Membrane #3) in array ARY003 contained spotted antibodies correspond- ing to 46 distinct RTKs other than the RTKs in ARY001. Both positive and negative controls were plotted in both arrays ARY001 and ARY003. ARY001 and ARY003 profile 202 unique tyrosine phosphorylation residues of 88 RTKs. Cells were lysed by NP-40 lysis buffer according to the
manufacturer’s protocol. The arrays were exposed to blocking buffer and incubated with 450 µg of cell lysate overnight at 4℃. The arrays were washed, incubated with a horseradish peroxida- se-conjugated phospho-tyrosine detection anti- body, treated with ECL solution, and exposed to film.
MTS cell proliferation assay. Cell proliferation was colorimetrically determined at 490 nm using a MTS cell proliferation assay kit (CellTiter 96 AQueous nonradioactive cell proliferation assay; Promega, Madison, WI) as previously described.8 After incuba- tion with sunitinib or PD98059 for 48 hours, cells in a 96-well plate were incubated with 333 mg/L 3-(4,5- dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2- (4-sulfophenyl)-2H-tetrazolium (MTS) and 25 um of phenazine methosulfate solution for 2 hours at 37℃ in a humidified, 50 mL/liter CO2 atmosphere. The absorbance of soluble formazan produced by cellular reduction of MTS was measured at 490 nm with an ELISA reader (SpectraMax M5 Multi-Mode Micro- plate Reader; Molecular Devices, Synnyvale, CA). Per- cent proliferation relative to the controls was calculated on the basis of the MTS readout. Experi- ments were repeated 4 times, and each had quadru- plicate samples.
Statistical analysis. To determine the minimum concentration of each agent necessary to achieve a significant effect (ICmin), we performed the MTS assay as described previously with various drug con- centrations. Each assay was done in triplicate. The ICmin was determined as the lowest concentration of drug with a statistically significant 2-tailed Stu- dent t test when compared with the negative control.
RESULTS
Treatment with sunitinib activates multiple tyrosine kinases. We first sought to determine whether treatment with sunitinib resulted in the compensatory hyperactivation of any tyrosine ki- nase. We performed sunitinib dose- and time- response experiments in H295R and SW13 cells (Fig 1). We could not achieve greater than a 50% decrease in proliferation at concentrations lower than 20 nM, even after 120 hours of treatment (data not shown). A 50% decrease in proliferation was achieved by 48 hours at 20 nM, and we deter- mined that this would be an ideal dose and time point for analysis.
The principal targets of sunitinib, including platelet-derived growth factor receptors, vascular endothelial growth factor receptors, and RET, showed no phosphorylation after 48 hours of exposure to sunitinib at 10 nM (Fig 2). Treatment
120
OD490 nm
Absorbance relative to control %
H295R cells
SW13 cells
100
80
*
*
60
*
*
40
20
0
Sunitinib
0 nM
2 nM
5 nM
10 nM
20 nM
0 nM
2 nM
5 nM
10 nM
20 nM
OD490 nm
Absorbance Relative to Control %
120
H295R cells
SW13 cells
100
80
*
*
60
*
*
40
20
0
PD98059
0 nM
10 nM
20 nM
50 nM
100 nM
0 nM
10 nM
20 nM
50 nM
100 nM
with sunitinib resulted in increased phosphoryla- tion of ERK, HCK, Chk2, YES, CREB, MEK, MSK, p38, FGR, and AXL (Fig 2). Although there were many differences in the phosphorylation status of individual kinases in H295R compared with SW13 cells before treatment, the pattern of tyrosine ki- nase activation after sunitinib exposure were strik- ingly similar in both cell lines (Fig 2).
ERK is the most activated tyrosine kinase after sunitinib treatment. We compared the degree of phosphorylation before and after 48 hours of sunitinib and found that ERK1/2 was the most hyperactivated by sunitinib in both cell lines (Fig 2). We validated this finding with immunoblot- ting and found that ERK phosphorylation was first observed at 4 hours after treatment then increased thereafter up to 48 hours after treatment (Fig 3). Immunoblottting confirmed that sunitinib consis- tently quashed FLT-3 and RET, 2 of its principal tar- gets, in a time-dependent manner (Fig 3). This technical validation confirmed our findings from the kinome array.
Sunitinb plus ERK inhibition has additive anti- proliferative effects. Because monotherapy with sunitinib resulted in the compensatory activation of ERK, we next determined the effect of combin- ing sunitinib with PD98059, a TKI that specifically inhibits ERK. Monotherapy with sunitinib or PD98059 at their ICmin concentrations (10 nM
for sunitinib and 50 nM for PD98059) for 48 hours reduced proliferation by 23% and 19%, respec- tively, in H295R cells, and by 25% and 24%, respec- tively, in SW13 cells (Fig 4). However, strategic combination treatment with these agents de- creased proliferation by 68% and 64% at 48 hours in H295R and SW13 cells, respectively (P <. 05 ver- sus monotherapy; Fig 4).
DISCUSSION
A trial to evaluate the efficacy of second-line sunitinib in advanced ACC is now underway. There is reason to believe that sunitinib monotherapy will fail to improve overall survival in patients with advanced ACC. In other cancer contexts, tyrosine kinase coactivation is an important mechanism of resistance to TKI monotherapy. Our premise is that combination regimens that inhibit multiple targets simultaneously are needed to overcome coactivation. We sought to develop a preclinical model to select strategic combinatorial regimens to overcome compensatory oncogenic signaling observed after monotherapy with sunitinib. The results from such modeling could be used to predict which combinatorial regimens are most likely to be successful in clinical trials for this rare malignancy.
We performed an unbiased global assessment of the kinome and found that different ACC cells lines have substantial differences in the pattern of kinase activation under baseline conditions. How- ever, treatment with sunitinib resulted in a specific pattern of compensatory signaling in which ERK was the most hyperactivated tyrosine kinase. The antiproliferative effects of combination therapy with sunitinib and the ERK inhibitor PD98059 exceeded the sum of the effects observed with each individual agent alone.
In other contexts, a number of distinct mecha- nisms can result in resistance to sunitinib. In gastrointestinal stromal tumors, mutations in the critical c-KIT gene result in sunitinib resistance.9 In renal cell carcinoma, the autocrine secretion of interleukin-8 can drive cell proliferation and di- minish the antiproliferative effects of sunitinib.10 In a recent report, investigators demonstrated that lysosomal drug sequestration can diminish sensitivity to sunitinib.11 We are the first to charac- terize compensatory tyrosine kinase coactivation in ACC as a principal mechanism of TKI resistance.
Interestingly, Gotink et al11 reported that treatment with sunitinib increased ERK phospho- rylation levels in sunitinib-resistant colorectal and renal carcinoma cell lines. A recent study showed that combination treatment of sunitinib
A
H295R
H295R
SW13
SW13
11
13
DMSO
..
DMSO
2
Sunitinib
10
..
9
B
Part #3
Sunitinib
7
1
HCK¡
10 ± 1.4
8
2
Chk21
19.3 ± 3.9
6
Part #1
3
YESİ
9.5# 1.3
4
CREBỘ
25 + 5.2
H295R
SW13
5
MEK1/21
5.3 ± 2.2
DMSO
6
MSK1/21
9 + 2.9
7
P3Sof
17,5 # 4.7
8
ERK1/21
57.3 + 18.2
9
Sunitinib
Fgrf
4.6 ± 0.9
10
AxIt
7.8 ±5.7
11
PDGFR (including Flt-3)!
-4.3 ± 0.8
Part #2
Targets of sunitinib
12
VEGFR Į
-6.7 ±2.6
13
RET Į
-6.2 ± 2.0
c-Kit is not in the arrays
0 h
4 h
8 h
16 h
32 h
48 h
0 h
4 h
8 h
16 h
32 h
48 h
Sunitinib
p-FLT-3 (120 kD)
+ FLT-3 (120 kD)
+ p-RET (120 kD)
+ RET (120 kD)
+ p-ERK1/2 (42, 44 kD)
+ ERK (42, 44 kD)
H295R
SW13
with mitotane results in the rapid metabolism of sunitinib. Specifically, treatment with mitotane up-regulates CYP3A4, which results in diminished sunitinib levels and chemoresistance.12 This study of combination therapy with mitotane and
120
OD490 nm
absorbance relative to control %
H295R cells
SW13 cells
100
*
*
*
80
*
I
Y.
T
60
40
*
*
20
0
0
Sun
PD
Sun + PD
C
Sun
PD
Sun + PD
sunitinib highlights the fact that not all combina- tion drug regimens offer additive or synergistic ef- fects and supports the use of preclinical models to predict which drugs should be used together.
There are several issues that remain to be addressed. First, administration of multiple TKIs may be associated with additive toxicity risks.
However, if combination therapy has synergistic anticancer effects, a lower dose of each agent could be used with equal efficacy and a more favorable side effect profile. Although low-dose TKIs in combination could be well-tolerated, the use of a second class of targeted agents could also be used. For instance, monoclonal antibodies, rather than TKIs, could be used to target specific signaling pathways. Our work did not investigate the efficacy and toxicity of the proposed combina- tion therapy in animals. Future studies should test the efficacy of sunitinib in combination with and ERK-targeting agent in animals to confirm the safety and synergistic anticancer activity in vivo.
In summary, preclinical modeling of tyrosine kinase coactivation is a useful tool to develop rational combinatorial therapies. Combination treatment with sunitinib and PD98059 is a promising therapeutic strategy for patients with advanced ACC and future studies should test this strategic combination regimen in clinical trials.
We thank Dr Gary D. Hammer (University of Michi- gan Medical School, Ann Arbor, MI) providing H295R and SW13 cells.
REFERENCES
1. Phan AT. Adrenal cortical carcinoma-review of current knowledge and treatment practices. Hematol Oncol Clin North Am 2007;21:489-507; viii-ix.
2. Schteingart DE, Doherty GM, Gauger PG, et al. Manage- ment of patients with adrenal cancer: recommendations of an international consensus conference. Endocr Relat Cancer 2005;12:667-80.
3. Costa R, Wesolowski R, Raghavan D. Chemotherapy for ad- vanced adrenal cancer: improvement from a molecular ap- proach? BJU Int 2011;108:1546-54.
4. Chun HG, Yagoda A, Kemeny N, Watson RC. Cisplatin for adrenal cortical carcinoma. Cancer Treat Rep 1983;67: 513-514.
5. Baudin E, Docao C, Gicquel C, et al. Use of a topoisomerase I inhibitor (irinotecan, CPT-11) in metastatic adrenocorti- cal carcinoma. Ann Oncol 2002;13:1806-9.
6. Lee JO, Lee KW, Kim CJ, et al. Metastatic adrenocortical carcinoma treated with sunitinib: a case report. Jpn J Clin Oncol 2009;39:183-5.
7. Stommel JM, Kimmelman AC, Ying H, et al. Coactivation of receptor tyrosine kinases affects the response of tumor cells to targeted therapies. Science 2007;318:287-90.
8. Lin CI, Whang EE, Abramson MA, et al. Galectin-3 regu- lates apoptosis and doxorubicin chemoresistance in papil- lary thyroid cancer cells. Biochem Biophys Res Commun 2009;379:626-31.
9. Guo T, Hajdu M, Agaram NP, et al. Mechanisms of sunitinib resistance in gastrointestinal stromal tumors harboring KITAY502-3ins mutation: an in vitro mutagenesis screen for drug resistance. Clin Cancer Res 2009;15:6862-70.
10. Huang D, Ding Y, Zhou M, et al. Interleukin-8 mediates re- sistance to antiangiogenic agent sunitinib in renal cell car- cinoma. Cancer Res 2010;70:1063-71.
11. Gotink KJ, Broxterman HJ, Labots M, et al. Lysosomal se- questration of sunitinib: a novel mechanism of drug resis- tance. Clin Cancer Res 2011;17:7337-46.
12. Kroiss M, Quinkler M, Lutz WK, Allolio B, Fassnacht M. Drug interactions with mitotane by induction of CYP3A4 metabolism in the clinical management of adrenocortical carcinoma. Clin Endocrinol (Oxf) 2011;75:585-91.
DISCUSSION
Dr Electron Kebebew (Bethesda, MD): I have 2 ques- tions for you. First, have you examined the kinome in adrenocortical cancers themselves? We have not found many of the tyrosine kinases to be activated, at least in the samples that we have looked at.
Second, in your in vitro studies, have you looked at the temporal relationship with the synergistic effect of a second TKI, whether it’s used at the same time or later?
Dr Daniel Ruan: I have not studied the kinome in hu- man adrenocortical carcinoma specimens. At this meet- ing, about 7 years ago, someone presented their experience studying the transcriptome, using a cDNA- based array, and they found considerable heterogeneity when comparing adrenocortical carcinomas but not be- tween adenomas. I am not aware of anyone doing a com- prehensive kinome profile of adrenocortical carcinoma in humans. However, surgeon-scientists are the best suited to perform kinome studies in human tumors. If you check genotyping in a specimen in paraffin, it is usu- ally not a problem. If you harvest a tumor specimen within minutes, for most RNAs that have reasonable half lives, it’s going to be fine if handled properly. But with kinome studies, you really need to handle the spec- imen immediately because the kinome changes in sec- onds, not in minutes or days. Surgeon-scientists are the best suited to handle tumors in the operating room to do kinome work.
Second, your question was, have I looked at the temporal relationship of efficacy? I have. And, actually, tomorrow, Lutske Lodewijk from Netherlands is going to show a similar experiment performed in medullary thyroid cancer cells. Along the same line of reasoning, in patients with medullary thyroid cancer, if you treat them with vandetanib, there is coactivation. If you target the most hyperactivated tyrosine kinase, in this case, epidermal growth factor receptor, with a second drug, there is really no additive effect until the time point, when you see compensatory hyperactivation, which is about 2 to 3 days in that model.
We haven’t observed any synergistic effect using this kind of modeling until you see compensatory hyperactiva- tion. And if you just randomly combine TKIs together, they typically don’t have synergistic, or even additive, effects. Furthermore, in the negative control that we used in this model, a MEK inhibitor, it has minimal antiproliferative effects. And when you combine it with other drugs, even those that target TKIs that are activated, they have no synergistic effect together in combination.
So the answer to your question is, until you see compensatory oncogene switching, there are no additive anticancer effects.
Dr Michael Demeure (Scottsdale, AZ): I have a ques- tion methodologically, primarily. You used the term “syn- ergy.” But I don’t know that you’ve established synergy or that the effects are just additive. Can you clarify why you think they are synergistic?
Dr Daniel Ruan: Yes. The way we defined synergy is-
Dr Michael Demeure: It’s clearly defined in the liter- ature, so I don’t think you have a license to redefine synergy as you wish, so please tell me-
Dr Daniel Ruan: The definition that we adapted from the literature is that if you take the minimum inhibitory concentration of different agents and add them to- gether, their added net effect is greater than the sum of their individual components.
The reason why we used this definition is because it’s very simple and straightforward, and you don’t have to do multiple time- and dose-response curves, which is prohibitively expensive.
Surgery is abstracted and/or indexed in Index Medicus, Science Citation Index, Current Contents/ Clinical Medicine, Current Contents/Life Sciences, and MEDLINE.
This Journal has been registered with Copyright Clearance Center, Inc, 222 Rosewood Dr, Danvers, MA 01923. Consent is given for the copying of articles for personal or internal use of specific clients. This consent is given on the condition that the copier pay directly to the Center the per-copy fee stated on the first page of each article for copying beyond that permitted by US Copyright Law. This consent does not extend to other kinds of copying, such as for general distribution, resale, advertising and promotional purposes, or for creating new collective works. All inquiries regarding copyrighted material from this publication other than those that can be handled through Copyright Clearance Center should be directed to Journals Permission Department, Elsevier Inc, 3521 Riverport Lane, Maryland Heights, MO 63043; (314) 447-8871