Dual inhibition of BRAF and MEK increases expression of sodium iodide symporter in patient-derived papillary thyroid cancer cells in vitro
Timothy M. Ullmann, MD, Heng Liang, MS, Maureen D. Moore, MD, Isra Al-Jamed, MD, Katherine D. Gray, MD, Jessica Limberg, MD, Dessislava Stefanova, MD, Jessica L. Buicko, MD, Brendan Finnerty, MD, Toni Beninato, MD, Rasa Zarnegar, MD, Irene M. Min, PhD, Thomas J. Fahey III, MD*
Abstract
Background: The majority of papillary thyroid cancers are driven by acquired mutations typically in the BRAF or RAS genes that aberrantly activate the mitogen-activated protein kinase pathway. This process leads to malignant transformation, dedifferentiation, and a decrease in the expression of the sodiumiodide symporter (NIS; SLC5A5), which results in resistance to radioactive iodine therapy. We sought to determine whether inhibition of aberrant mitogen-activated protein kinase-signaling can restore NIS expression.
Methods: We prospectively developed cultures of papillary thyroid cancers derived from operative specimens and applied drug treatments for 24 hours. Samples were genotyped to identify BRAF and RAS mutations. We performed quantitative PCR to measure NIS expression after treatment.
Results: We evaluated 24 patient papillary thyroid cancer specimens; BRAFV600E mutations were identified in 18 out of 24 (75.0%); 1 patient tumor had an HRAS mutation, and the remaining 5 were BRAF and RAS wildtype. Dual treatment with dabrafenib and trametinib increased NIS expression (mean fold change 4.01 ± 2.04, P < .001), and single treatment with dabrafenib had no effect (mean fold change 0.98 ± 0.42, P ¼ .84). Tumor samples that had above-median NIS expression increases came from younger patients (39 vs 63 years, P < .05).
Conclusion: Dual treatment with BRAF and MEK inhibitors upregulated NIS expression, suggesting that this treatment regimen may increase tumor iodine uptake. The effect was greatest in tumor cells from younger patients.
Introduction
The majority of papillary thyroid cancers (PTCs) are driven by acquired somatic mutations in either the B-raf serine-threonine kinase (BRAF) or Ras GTPase (RAS) family of genes.1,2 These mutations lead to constitutive activation of the translated proteins and downstream signaling through MEK and ERK in the mitogenactivated protein kinase (MAPK) pathway, resulting in dedifferentiation, malignant transformation, and decreased expression of thyroid-specific genes, such as the sodium-iodide symporter (NIS; SLC5A5).2 NIS encodes a transmembrane protein responsible for uptake of iodide ions from the blood into thyrocytes for the production of thyroid hormone. Radioactive iodine (RAI) therapy exploits this NIS-mediated iodide uptake to treat thyroid cancers such as PTC that originate from these cells.3 Aggressive or advanced thyroid cancers, however, often have dramatically decreased NIS expression and are RAI-refractory.2,4 Therefore, strategies for redifferentiating PTC cells to increase NIS levels may be necessary to improve the efficacy of RAI for many patients.
Expression of thyroid-specific genes including NIS seem to be decreased by constitutive BRAF and MAPK activity,5e7 and data suggest that inhibition of this aberrant MAPK signaling may increase NIS expression and therefore RAI sensitivity in patients with advanced PTC. Nagarajah et al used a mouse model of inducible BRAFV600E mutant thyroid cancer to show that MEK inhibition leads to increased RAI uptake and sensitivity.8 Similar results have been shown in human thyroid cancer cell lines.3,7,9,10 This approach has also had some efficacy for patients with RAIrefractory advanced PTC in clinical trials. Ho et al demonstrated in a small cohort of patients that RAI uptake and sensitivity could be improved after treatment with selumetinib, a MEK inhibitor (MEKi).11 Dabrafenib, a BRAF inhibitor (BRAFi), was also shown to increase RAI uptake in patients with RAI-refractory, BRAFV600E mutant cancers.12 More recently, another group found that genotype-based MAPK inhibition could restore RAI uptake in a small cohort of RAS and BRAF mutant PTC patients.13 But these trials included only patients who had RAI-refractory cancers. At this point in the course of the disease, prognosis is poor, and complete cure may not be possible.2 Therefore, patients with aggressive cancers may benefit from earlier kinase inhibitor therapy to presensitize any remnant cancer cells to the initial postthyroidectomy RAI treatment with the goal of increasing cure rates and preventing the development of RAI-refractory disease.
It is unknown whether MAPK inhibition is effective in increasing NIS expression in RAI-naïve human PTCs. Furthermore, the small sample sizes of these limited clinical trials have made it difficult to predict which patients may have the best response to treatment. To address these questions, we designed a system of patient-derived, primary PTC culture that allowed us to measure NIS expression in individual patients treated versus untreated cells and to identify clinicopathologic features associated with greater NIS increases in response to treatment. Methods
Human subjects
All human subject-based experiments and data collection were approved by the Institutional Review Board of Weill Cornell Medicine. Written informed consent was obtained from all subjects, all of whom were approached by a member of the research team before operation and enrolled prospectively in the study. Confidentiality was strictly maintained in compliance with Health Insurance Portability and Accountability Act laws and hospital policy.
Patients were enrolled from November 2017 to January 2019. All patients undergoing initial operation for suspicious or malignant (Bethesda V or VI on fine-needle aspirate cytology) thyroid nodules 1.0 cm or larger or those with indeterminate (Bethesda IIIeIV) cytology and molecular testing suggesting greater risk of malignancy were approached for consent to participate. We excluded patients whose tumors were found to be nonpapillary cancers or noninvasive, follicular thyroid neoplasms with papillary-like nuclear features, or whose tumors were too small (at the discretion of the pathologist) to obtain a sample without compromising the integrity of the final diagnosis. All patients that met inclusion criteria during the study period were approached for consent.
Cell culture
BCPAP cells, a human BRAFV600E-mutant PTC cell line (Leibniz Institute DSMZ, Germany), and patient-derived cell cultures were grown and treated in standard Roswell Park Memorial Institute media with 10% fetal bovine serum and penicillin and streptomycin to prevent bacterial contamination. Primary cultures were developed from individual patient tumors immediately after operative excision of the specimen. The specimen was taken directly from the operating room to the pathology department by a member of the research team. The pathologists then performed a gross examination of each specimen and provided the research team with a representative piece of tissue from each primary tumor. The tissue was placed into ice-cold, sterile, phosphate-buffered saline and then minced with a sterile blade before being plated in 6-well, standard cell culture dishes with Roswell Park Memorial Institute media. Cultures were plated 24 hours before drug treatments. Treatments for patient-derived cell cultures were assigned sequentially over the course of the study; the earliest patient samples were treated with dabrafenib alone, the next group with the addition of supratherapeutic thyroid stimulating hormone, and finally, the last in combination with trametinib.
Drug treatment was applied for 24 hours before cells were harvested for ribonucleic acid (RNA) or protein extraction. Dabrafenib and trametinib were obtained from Selleckchem (Houston, TX). These drugs were chosen because of their recent approval for use in anaplastic thyroid cancer and their efficacy in treating BRAFV600E-mutant melanoma.14 Drugs were dissolved in dimethyl sulfoxide, which was used as the vehicle control in all experiments. Drugs and dimethyl sulfoxide were added at 0.1% (v/v) final concentration in culture medium.
Quantitative PCR
Bulk RNA was extracted from BCPAP and primary tumor cells using the RNEasy extraction kit (Qiagen, Hilden, Germany) and cDNA synthesized using the High Capacity Reverse Transcription Kit (ThermoFisher Scientific, Waltham, MA) according to the respective manufacturer’s instructions. Real-time quantitative PCR (qPCR) was performed using the TaqMan system (ThermoFisher Scientific), with separate kits ordered for NIS/SLC5A5 (ID#: Hs00166567) and for GAPDH (ID#: Hs99999905) as the housekeeping control. PCR was performed in triplicate for each gene and tumor sample and normalized to GAPDH using the DDCt technique for calculation of fold change.15
Immunoblotting
Cells were lysed in ice-cold, radioimmunoprecipitation assay buffer (ThermoFisher Scientific) with phosphatase and protease inhibitor added, and protein concentrations were measured using the Pierce BCA Protein Assay Kit (ThermoFisher Scientific). Lysates were run on 4% to 20% gradient, polyacrylamide gels and transferred using the methanol-based, wet-transfer technique onto 0.45 mm nitrocellulose membranes (Bio-Rad Laboratories, Hercules, CA). The membranes were blocked with 5% bovine serum albumin for both primary and secondary antibody application and developed using SuperSignal West Pico PLUS Chemiluminescent Substrate (ThermoFisher Scientific). All antibodies were obtained from Cell Signaling Technology (Danvers, MA). Primary antibodies against phosphorylated ERK (Thr202/Tyr204; #9101S) and AKT (Ser473; #4060S), and total ERK (#4695S), AKT (antibody #4691P), BRAF (#14814), and a-tubulin (#2125S) were used at 1:1000 dilutions. Secondary antirabbit antibody (#7074S) was used at 1:20,000 dilution for each primary antibody. Quantification was performed using ImageJ software (National Institute of Health, Bethesda, MD).
Cell viability
BCPAP cells were seeded in triplicate in a 96-well plate and drug treated after 24 hours. Cell viability was measured after drug treatment at various time points using the Invitrogen Molecular Probes Vybrant MTT Cell Proliferation Assay Kit (ThermoFisher Scientific) according to manufacturer’s specifications.
Genotyping
Tumor sequencing was done independently of the research team; all tumors diagnosed as PTC on final pathology undergo next-generation sequencing as part of standard of care at New York Presbyterian Hospital-Weill Cornell Medical Center. Deoxyribonucleic acid (DNA) was extracted from macrodissected, paraffin-embedded tumor using the Maxwell 16 FFPE Plus LEV DNA Kit (Promega, Madison, WI). The extracted DNA was amplified and prepared using the Ion Chef Instrument and subjected to next-generation sequencing using the Ion Ampliseq Cancer Panel hotspot v2 on the Ion Torrent Personal Genome Machine (Life Technologies, Carlsbad, CA). The panel contains the BRAF, HRAS, KRAS, and NRAS genes commonly mutated in PTC and detects single-nucleotide variants and indels at mutational hotspots in these genes; this technique, however, does not sequence them in their entirety and may not detect genetic rearrangements or large deletions. The complete list of partially sequenced genes is provided in Supplemental Table I.
Statistical analyses
Categorical variables were compared using Fisher exact test; continuous, normally distributed variables using either paired or unpaired t tests or analysis of variance with post-hoc Dunnett’s tests where appropriate; and skewed variables were compared using the Mann-Whitney or Kruskal-Wallis tests. To determine treatment effect on primary culture cells from human tumors, DCt values (NISeGAPDH) were calculated for control and treatment conditions and compared as paired data using t tests.16 All statistical analyses were performed using Stata software, version 13.1 (Stata Corp. College Station, TX).
Results
BCPAP cell treatment
The human BCPAP thyroid cancer cell line containing the BRAFV600E mutation was used to determine an effective dosing regimen as a basis for the patient-derived, primary culture experiments. NIS expression was measured by qPCR after drug treatment with dabrafenib, a BRAF inhibitor; trametinib, a MEK inhibitor; or the combination of both. Although minimal increase in NIS expression was seen after treatment with either drug alone, the combination increased NIS expression approximately 6-fold at doses of 0.5 mM dabrafenib and 0.25 mM trametinib (Fig 1, A). The majority of treated cells were still alive after 24 hours of treatment, but at 48 hours, only 50% of cells remained viable (Fig 1, B). Immunoblotting demonstrated that dabrafenib decreased MAPK activity as measured by ERK phosphorylation within 24 hours. Trametinib, whether in combination with dabrafenib or used as a single agent, completely blocked MAPK signaling (Fig 2, A and B). These drugs had no effect on PI3K-AKT-mTOR signaling as measured by AKT phosphorylation (Fig 2, A). Trametinib at 0.25 mM with dabrafenib 0.5 mM and 1.0 mM concentrations for 24 hours of treatment yielded the maximal increases in NIS expression with minimal cell death; these doses were chosen for the patientderived primary culture experiments.
Patient-derived, primary PTC cultures
Patients were prospectively enrolled in the study to determine treatment effects in patient-derived tumor cultures. Sixty-one patients met inclusion criteria and consented, 5 patients’ cells were used for protein extraction, and 52 patients’ cells were used for RNA extraction. Four patients had neoplasms with papillary-like nuclear features on final pathology and were excluded. RNA extraction yielded insufficient or poor-quality RNA in 24 patients (46.2%), and these were not processed for qPCR. Included patient demographics and clinicopathologic data are summarized in Table I. BRAFV600E mutations were identified in 18 out of 24 (75.0%) PTCs; 1 patient had an HRASQ61R mutation. The remainder were BRAF and RAS wildtype.
Dabrafenib and trametinib decreased MAPK signaling in patient-derived cells in a similar manner to their effects in BCPAP cells (Fig 3, A and B). Dabrafenib alone produced a modest decrease in ERK phosphorylation, whereas trametinib abolished it. Similar to the BCPAP cells, treatment with dabrafenib alone produced little effect on NIS expression (Fig 4, A). The addition of 500 mIU/mL of thyroid stimulating hormone (roughly 100 times the upper limit of normal levels in human serum) did not increase the efficacy of dabrafenib (Fig 4, A). In contrast to BCPAP cells, trametinib alone did increased NIS expression in patient-derived PTC cells (mean fold change 3.10, P < .05), but the combination of trametinib and dabrafenib induced a greater fold change in these same patients (mean fold change 4.61, P < .001; Fig 4, B). Unlike BCPAP cells, patientderived cells demonstrated greater NIS expression after treatment with 0.25 mM trametinib and 1.0 mM dabrafenib (mean fold change 4.01, P < .001) rather than 0.5 mM dabrafenib (mean fold change 2.53, P < .05; Fig 4, C). This effect was cancer specific; there was no increase in NIS expression in adenomas treated with this regimen (mean fold change 1.14, P ¼ .62; Fig 4, C). Cells from 4 patients were treated with both single agent trametinib and a dual treatment combination of dabrafenib and trametinib. The addition of dabrafenib increased NIS expression even in cells from a patient with an HRASQ61R mutation and wildtype BRAF, but had minimal effect in cells from 2 patients’ tumors with the BRAFV600E mutation (Fig 4, D).
Patients (n ¼ 13) whose tumor cells were treated with the most effective drug regimen (dabrafenib 1.0 mM and trametinib 0.25 mM) were then divided into 2 groups: those whose tumor cells had above-median increase in NIS expression were included in the high increase group, and those below the median were included in the low increase group. Although clinicopathologic characteristics, including mutation status, were not different between the 2 groups, the high increase group included younger patients (39 ± 14 years vs 63 ± 14 year, P < .05; Table II).
Discussion
This is the first study to demonstrate that the combination of BRAFi and MEKi can increase NIS expression in patient-derived, primary PTC tumor cultures. All of the tumor cells treated in these experiments were from RAI-naïve patients. Previous clinical trials have focused primarily on treating patients with RAIrefractory tumors. In contrast, the data reported here suggest that MAPK inhibitors may increase NIS levels and possibly RAI sensitivity even in RAI-naïve tumor cells.
Furthermore, although we were able to demonstrate an ontarget effect and complete blockade of MAPK signaling with trametinib alone, it was most effective at increasing NIS expression when used in combination with dabrafenib. This observation suggests that MAPK inhibition alone is insufficient to increase NIS levels and that there are other regulatory mechanisms affecting NIS expression in PTC. This conclusion is in agreement with clinical trials, which have shown inconsistent patient responses to treatment despite the use of targeted therapies and preclinical data that suggest that epigenetic regulation of the NIS locus and NIS membrane trafficking play important roles in determining RAI-uptake in PTC.11e13,17e20 Another possibility is that the cancer cells may have developed adaptive resistance to MAPK inhibition, thus limiting their response to therapy; however, although up to 20% of BRAFi or MEKi resistant melanomas show upregulation of the PI3K-AKTmTOR signaling pathway, we did not find the same results in BCPAP cells (Fig 2, A).21 Indeed, the combination of dabrafenib and trametinib effectively killed BCPAP cells after 48 hours (Fig 1, B), further suggesting that the cells were sensitive to the drugs. Taken together, the data presented here add to the growing body of evidence that MAPK inhibition alone is not sufficient to increase NIS in all PTCs.
We have shown that tumor cells with the greatest posttreatment increases in NIS expression come from younger patients. This finding may explain in part some of the variability in patient responses to MAPK inhibitors seen in clinical trials. Furthermore, RAI treatment, particularly in high doses, has been associated with life-long adverse events, such as infertility and development of secondary malignancies including leukemias.22e24 Younger patients, therefore, may benefit the most from treatments that can decrease the necessary RAI dose to decrease the lifetime risk of RAI-related, adverse events.
There are some limitations to the present study. We measured NIS expression using qPCR, but this is only a surrogate marker for NIS protein levels and may not correlate directly with RAI uptake. Future studies directly measuring cellular RAI absorption are necessary to demonstrate that these results translate to increased intracellular I131 after drug treatment. Furthermore, as with all in vitro experiments, these results may not translate to human physiology and clinical practice. In assessing predictors of a high increase in NIS expression, we are limited by sample size and cannot perform a multivariable analysis; therefore, there may be confounders we cannot eliminate. Finally, PTCs contain heterogeneous populations of cells, and tumor samples were selected by gross examination only. Therefore, the abundance of malignant cells in each tumor sample could differ, and results could be skewed by samples containing more or fewer malignant cells, or a greater or lesser relative abundance of mutant BRAF, RAS, or other proteins.
Despite these limitations, we have provided preclinical evidence that RAI-naïve human PTCs demonstrate increased NIS expression after dual treatment with dabrafenib and trametinib. We have further shown that this effect cannot be explained entirely by MAPK blockade and that it is genotype independent. Using patientderived primary tumor cells to investigate the effects of MAPK inhibition in PTC allowed us to identify patient clinicopathologic factors of patients that correlate with cellular responses to drug treatment. As a result, we found that tumor cells with the greatest increases in NIS expression in response to treatment were from younger patients. Additional studies are necessary to validate these results in vivo and potentially to demonstrate clinical efficacy of this drug regimen.
References
1. Cancer Genome Atlas Research N. Integrated genomic characterization of papillary thyroid carcinoma. Cell. 2014;159:676e690.
2. Xing M. Molecular pathogenesis and mechanisms of thyroid cancer. Nat Rev Cancer. 2013;13:184e199.
3. De la Vieja A, Santisteban P. Role of iodide metabolism in physiology and cancer. Endocr Relat Cancer. 2018;25:R225eR245.
4. Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: The American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid. 2016;26:1e133.
5. Chakravarty D, Santos E, Ryder M, et al. Small-molecule MAPK inhibitors restore radioiodine incorporation in mouse thyroid cancers with conditional BRAF activation. J Clin Invest. 2011;121:4700e4711.
6. Durante C, Puxeddu E, Ferretti E, et al. BRAF mutations in papillary thyroid carcinomas inhibit genes involved in iodine metabolism. J Clin Endocrinol Metab. 2007;92:2840e2843.
7. Liu D, Hu S, Hou P, Jiang D, Condouris S, Xing M. Suppression of BRAF/MEK/ MAP kinase pathway restores expression of iodide-metabolizing genes in thyroid cells expressing the V600E BRAF mutant. Clin Cancer Res. 2007;13: 1341e1349.
8. Nagarajah J, Le M, Knauf JA, et al. Sustained ERK inhibition maximizes responses of BrafV600E thyroid cancers to radioiodine. J Clin Invest. 2016;126: 4119e4124.
9. Cheng L, Jin Y, Liu M, Ruan M, Chen L. HER inhibitor promotes BRAF/MEK inhibitor-induced redifferentiation in papillary thyroid cancer harboring BRAFV600E. Oncotarget. 2017;8:19843e19854.
10. Zhang H, Chen D. Synergistic inhibition of MEK/ERK and BRAF V600E with PD98059 and PLX4032 induces sodium/iodide symporter (NIS) expression and radioiodine uptake in BRAF mutated papillary thyroid cancer cells. Thyroid Res. 2018;11:13.
11. Ho AL, Grewal RK, Leboeuf R, et al. Selumetinib-enhanced radioiodine uptake in advanced thyroid cancer. N Engl J Med. 2013;368:623e632.
12. Rothenberg SM, McFadden DG, Palmer EL, Daniels GH, Wirth LJ. Redifferentiation of iodine-refractory BRAF V600E-mutant metastatic papillary thyroid cancer with dabrafenib. Clin Cancer Res. 2015;21:1028e1035.
13. Jaber T, Waguespack SG, Cabanillas ME, et al. Targeted therapy in advanced thyroid cancer to resensitize tumors to radioactive iodine. J Clin Endocrinol Metab. 2018;103:3698e3705.
14. Subbiah V, Kreitman RJ, Wainberg ZA, et al. Dabrafenib and trametinib treatment in patients with locally advanced or metastatic BRAF V600-mutant anaplastic thyroid cancer. J Clin Oncol. 2018;36:7e13.
15. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using realtime quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25:402e408.
16. Yuan JS, Reed A, Chen F, Stewart Jr CN. Statistical analysis of real-time PCR data. BMC Bioinformatics. 2006;7:85.
17. Choi YW, Kim HJ, Kim YH, et al. B-RafV600E inhibits sodium iodide symporter expression via regulation of DNA methyltransferase 1. Exp Mol Med. 2014;46, e120.
18. Riesco-Eizaguirre G, Gutierrez-Martinez P, Garcia-Cabezas MA, Nistal M, Santisteban P. The oncogene BRAF V600E is associated with a high risk of recurrence and less differentiated papillary thyroid carcinoma due to the impairment of Naþ/I- targeting to the membrane. Endocr Relat Cancer. 2006;13:257e269.
19. Sodre AK, Rubio IG, Galrao AL, et al. Association of low sodium-iodide symporter messenger ribonucleic acid expression in malignant thyroid nodules with increased intracellular protein staining. J Clin Endocrinol Metab. 2008;93: 4141e4145.
20. Zhang Z, Liu D, Murugan AK, Liu Z, Xing M. Histone deacetylation of NIS promoter underlies BRAF V600E-promoted NIS silencing in thyroid cancer. Endocr Relat Cancer. 2014;21:161e173.
21. Lim SY, Menzies AM, Rizos H. Mechanisms and strategies to GSK1120212 overcome resistance to molecularly targeted therapy for melanoma. Cancer. 2017;123: 2118e2129.
22. Lee SL. Complications of radioactive iodine treatment of thyroid carcinoma. J Natl Compr Canc Netw. 2010;8:1277e1286; quiz 87.
23. Teng CJ, Hu YW, Chen SC, et al. Use of radioactive iodine for thyroid cancer and risk of second primary malignancy: A nationwide population-based study. J Natl Cancer Inst. 2016;108.
24. Wu JX, Young S, Ro K, et al. Reproductive outcomes and nononcologic complications after radioactive iodine ablation for well-differentiated thyroid cancer. Thyroid. 2015;25:133e138.