Abstract
Breast cancer is the leading cause of cancer-related death in women worldwide. Human epidermal growth factor receptor 2 (HER2)-positive subtype comprises 20% of sporadic breast cancers and is an aggressive disease. While targeted therapies have greatly improved its management, primary and acquired resistance remain a major roadblock to making it a curable malignancy. Ganetespib, an Hsp90 (Heat shock protein 90) small molecule inhibitor, shows preferential efficacy in HER2-positive breast cancer, including therapy-refractory cases, and has an excellent safety profile in ongoing clinical trials (38 in total, six on breast cancer). However, Ganetespib itself evokes acquired resistance, which is a significant obstacle to its clinical advancement. Here, we show that Ganetespib potently, albeit temporarily, suppresses HER2-positive breast cancer in genetic mouse models, but the animals eventually succumb via acquired resistance. We found that Ganetespib-resistant tumors upregulate several compensatory HSPs, as well as a wide network of phospho-activated receptor tyrosine kinases (RTKs), many of which are HSP clients. Downstream of p-RTKs, the MAPK pathway remains suppressed in the resistant tumors, as is HER2 itself. In contrast, the p-RTK effector Akt is stabilized and phospho-activated. Notably, pharmacological inhibition of Akt significantly delays acquired Ganetespib resistance, by 50%. These data establish Akt as a unifying actionable node downstream of the broadly upregulated HSP/p-RTK resistance program and suggests that Akt co-targeting with Ganetespib may be a superior therapeutic strategy in the clinic.
Similar content being viewed by others
Introduction
Breast cancer is the most common female malignancy and the major cause of cancer-related mortality in women worldwide1. Human epidermal growth factor receptor 2 (HER2)-positive breast cancer comprises about 20% of all breast cancers and is defined by HER2 gene amplification/protein upregulation2,3,4,5. HER2 overexpression is a significant negative predictor of overall survival and time to relapse, due to enhanced cell proliferation, angiogenesis and metastasis, and reduced apoptosis6,7,8. Moreover, this molecular subtype is more frequently diagnosed in younger patients (median 57 vs. 64 y.o., METABRIC9) and at advanced metastatic stage III/IV, in 54% of cases compared with 36% of hormone receptor-positive/HER2-negative breast cancer and 41% of triple-negative breast cancer, another aggressive subtype5.
HER2-targeted therapies (Trastuzumab, Lapatinib) have greatly improved the treatment of HER2-positive breast cancer4. However, 36–76% patients show primary resistance to Trastuzumab, and the majority of initial responders progress within one year10,11. The search for alternative vulnerabilities revealed that the HER2 oncogene heavily relies on the heat shock protein 90 (Hsp90) molecular chaperone and is very sensitive to Hsp90 inhibitors12,13,14. Hsp90, one of the most abundant proteins in the cell, regulates normal cellular homeostasis by maintaining proper folding, stability and activity of its “client” proteins15,16. During tumorigenesis, Hsp90 is further upregulated in response to the proteotoxic stress and thereby confers superior proliferative, survival, angiogenic and metastatic properties to cancer cells17,18. Notably, of over 700 identified Hsp90 clients (www.picard.ch/downloads/Hsp90interactors.pdf), many are oncoproteins, including oncogenic drivers of breast cancer (e.g., estrogen and progesterone receptors, BRCA1/2), components of the HER2 signaling pathway (HER2, its co-receptor EGFR, downstream effectors Raf1, Erk, Akt, mTOR), and common mediators of therapeutic resistance in breast and other cancers (e.g., mutant HER2 and EGFR, HER3, IGF-1R)12,13,14,15,19,20,21,22,23.
Although Hsp90 inhibitors intercept multiple oncogenic and resistance pathways at once18,24, their clinical activity to date has been restricted to the cancers driven by the most sensitive Hsp90 clients, e.g., HER2-positive breast cancer25,26,27 and EML4-ALK-positive non-small-cell lung cancer28,29. Indeed, multiple first-generation Hsp90 inhibitors showed promising results in HER2-positive breast cancer clinical trials, but their development was stalled due to significant adverse effects, in particular liver and ocular toxicities13. In turn, a next-generation Hsp90 inhibitor Ganetespib (STA-9090, Synta Pharmaceuticals) emerged as a significantly safer alternative25,26 that has so far been evaluated in 38 clinical trials (two of them Phase III), including six on breast cancer. Of the latter, three completed studies strongly point to Ganetespib’s preferential efficacy in the HER2 subtype25,26,27. First, in the patients with metastatic breast cancer of all molecular subtypes, Ganetespib produced positive responses primarily in the HER2 group25. Second, Ganetespib—in combination with Trastuzumab and Paclitaxel—tested specifically in the HER2-positive breast cancer patients again showed a positive response and excellent safety profile26. Finally, a retrospective search for biomarkers of sensitivity to Hsp90 inhibitors found that HER2 is the most important—“and perhaps the only”—such biomarker27.
Despite Ganetespib’s great promise for HER2-positive breast cancer, patients in the clinical trials eventually progress on it, i.e., develop therapeutic resistance25,26. Therefore, the understanding and pharmacological interception of acquired Ganetespib resistance is critical for its clinical advancement and will also advance our knowledge on the entire class of Hsp90 inhibitors (none is FDA approved so far). We recently reported that Ganetespib is effective in a genetic mouse model of HER2-positive breast cancer after acquiring Lapatinib resistance, thus mimicking its activity in human therapy-refractory disease30. Therefore, genetic models are both appropriate and the most pre-clinically advanced to study this drug. Here, we tested the long-term effectiveness of Ganetespib in mouse models and found that, similar to patients, it evokes acquired resistance. Therefore, we set out to gain an in-depth understanding of the molecular mechanisms of acquired Ganetespib resistance in vivo, with the ultimate goal of advancing treatment options and improving outcomes for patients with HER2-positive breast cancer.
Materials and methods
Mouse strains, mammary fat pad transplantation, drug treatments
MMTV-Neu mice (strain FVB/N-Tg(MMTVneu)202Mul/J) and MMTV-ErbB2 mice (strain FVB/N-Tg(MMTV-ERBB2)NK1Mul/J) were from the Jackson Laboratory and were previously described30,31,32. Only females were used, since breast cancer is 100 times more prevalent in females than in males. Mice were treated with Ganetespib when autochthonous tumors appeared, i.e., at the age of 6−18 months. Immunodeficient athymic nude mice were from the Jackson Laboratory (strain Foxn1nu//Foxn1nu, Bar Harbor, ME). For mammary fat pad transplantation of primary Neu tumor cells, tumor cells were isolated as described below. Freshly isolated tumor cells were injected into mammary glands #4 and #9 of 5 weeks old Foxn1nu//Foxn1nu females at 10,000 cells per injection, in 100 µl of 3:1 DMEM:Matrigel (Corning, #356234, Corning, NY). Autochthonous and allografted mammary tumors were measured weekly by caliper, tumor volume was calculated as l*w*h/2, where “l” is length, “w” is width, “h” is height (an approximate formula for the volume of ellipsoid). Ganetespib (Selleckchem, #S1159, Houston, TX) in Cremophor vehicle (10% DMSO/18% Cremophor/3.6% dextrose) was injected via tail vein at 80 mg/kg once a week. MK-2206 (Selleckchem, #S1078, Houston, TX) in Captisol vehicle (30% in water) was injected via oral gavage at 20 mg/kg three times a week. Treatments of the allografted animals started when the allografts reached ~90 mm3. All animals were treated humanely and according to the guidelines by the Stony Brook University Institutional Animal Care and Use Committee (IACUC). Sample size was not pre-determined, animals were assigned to treatment groups randomly, no blinding was used.
Isolation of primary Neu tumor cells for mammary gland allografts
Neu tumors were dissected and processed into single tumor cell isolates as previously described33. Tumor cells were grown in DMEM/F12 Medium (Gibco, #10565-018, Gaithersburg, MD) supplemented with 10% FBS for 2 weeks in a humidified chamber at 37 °C with 5% CO2. Tumor cells were detached from plates using accutase (Millipore Sigma, #SCR005, Burlington, MA), resuspended in CNT-Prime epithelial culture medium (CELLnTEC, #Cnt-PR, Bern, Switzerland), and incubated with EpCAM-APC (Invitrogen, #17-5791-82, Carlsbad, CA) and ErbB2/HER2-Alexa Fluor 594 (R&D Systems, #FAB6744T, Minneapolis, MN). The cells were sorted with a FACSAria III Cell Sorter (BD Biosciences, San Jose, CA) and analyzed using Cyflogic software (v1.2.1, CyFlo Ltd, Turku, Finland). The EpCAM-APC and ErbB2/HER2-Alexa Flour 594 double-positive cells were immediately used for transplantation.
Quantitative real-time reverse transcription polymerase chain reaction (RT-PCR)
Total RNA was extracted from the mammary tumors using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Complementary DNA (cDNA) was prepared from 1 µg total RNA using random primers and SuperScript II reverse transcriptase (Invitrogen, Carlsbad, CA). The PCR primers were the following: Hsp90A ttgcttcagtgtcctggtg (F), cctgtttgctgggaatgag (R); Hsp90B cctgctctgtactactactc (F), aatgcctgtgtccaccaaag (R); Hsp70 gatcatcgccaacgaccag (F), ctcgcccttgtagttcacc (R); Hsc70 tggcattgatctcggcacc (F), acgcccgatcagacgtttg (R), Hsp27 cacagtgaagaccaaggaag (F), cctcgaaagtaaccggaatg (R), Hsp40 (aka Dnaja1, DnaJ) aaaacccaatgccacccag (F), tccatgggtgagccaaaac (R). Real-time PCR was performed using QuantiTect SYBR Green PCR kit (Qiagen, Valencia, CA) on an Applied Biosystems QuantStudio 3 Real-time PCR system and analyzed with QuantStudio Design & Analysis software version 1.4 (Applied Biosystems, Waltham, MA). The PCR conditions were: 94 °C for 30 s, 60 °C for 1 min and 72 °C for 1.5 min, 40 cycles. Samples were analyzed in duplicates and normalized to HPRT and to the mean signal for each gene.
Western blot analysis and phospho-RTK array assay
Mammary tumors were minced and resuspended in RIPA buffer (150 mM NaCl, 1% NP-40, 0.5% Deoxycholate, 0.1% SDS and 50 mM Tris HCl, pH 8) with protease inhibitors, sonicated, spun down, and fat was removed before protein quantification. Immunoblots were performed using 20–60 μg of protein using the following antibodies, all from Cell Signaling Technology (Danvers, MA): HER2 (#4290T), p-HER2 (Y1248, #2247T), p-HER2 (Y1221/1222, #2243T), Akt (#4691T), p-Akt (S473, #4060T), p-Akt (T308, #13038T), actin (#3700T). The mouse phospho-RTK analysis was performed using the Proteome Profiler Mouse Phospho-RTK Array Kit (R&D Systems, #ARY014, Minneapolis, MN) according to the manufacturer’s instructions. Signal densities were quantified using NIH Image software and normalized as indicated in the figure legends and also, to the background.
Mass spectrometry
Control, Resistant-3 h and Resistant-24 h tumor samples from Neu mice that acquired Ganetespib resistance in vivo during at least 2 months were analyzed. Each sample contained three biological replicas pooled together. The samples in 8 M urea, 100 mM ammonium bicarbonate, were subjected to reduction (5 mM DTT), alkylation (10 mM iodoacetamide), diluted to 2 M urea and digested with trypsin at 37 °C overnight. Peptides were desalted, lyophilized, resuspended in 0.1% TFA, 50% acetonitrile, 1 M lactic acid and incubated with TiO2 beads for one hour. Peptides not binding to TiO2 beads with collected as a non-phosphorylated, total protein fraction, lyophilized and desalted. Peptides bound to TiO2 were washed with 0.1% TFA, 50% acetonitrile and eluted with 50 mM KH2PO4 pH 10.5, neutralized with 5% formic acid, 50% acetonitrile, lyophilized and desalted. Phospho-peptides and non-phosphorylated (total protein) peptides were both analyzed by liquid chromatography-electrospray ionization tandem mass spectrometry (LC-MS/MS), using orbital trap (Q-Exactive HF; ThermoFisher Scientific, Waltham, MA) and orthogonal quadruple TOF (5600Plus; Sciex, Framingham, MA) instruments followed by protein database searching. HPLC C18 columns were prepared using a P-2000 CO2 laser puller (Sutter Instruments, Novato, CA) and silica tubing (75 µm ID x ~10 cm) and were self-packed with 3 µm Magic AQ C18 resin. Protein abundance and peptide phosphorylation site abundance were established by protein database searching using the ProteomeDiscoverer v2.3 and ProteinPilot v5.01, followed by statistical analysis using JMP12. Three missed tryptic cleavages were allowed, and the posttranslational modifications considered included Ser, Thr and Tyr phosphorylation. Database searches used the mouse UniProt FASTA database (16982 reviewed sequences including common contaminants). False discovery rates of peptide capture experiments were typically <1%.
Statistical analysis
Appropriate statistical tests were used for each dataset, as follows. Mouse overall survival and progression-free survival was analyzed by Kaplan–Meier analysis and Log Rank statistics, the p-values and hazard rates were determined using an online software (www.evanmiller.org/ab-testing/survival-curves.html). Tumor size and mRNA and protein expression levels were analyzed by unpaired two-tailed Student’s t-test. p < 0.05 was considered statistically significant. No blinding was used. Estimate of variation was not performed, normal distribution was not tested for.
Results
Ganetespib is effective against HER2-positive breast cancer in vivo, but eventually evokes acquired resistance
While investigating acquired resistance to Lapatinib, we recently reported that Ganetespib potently suppresses HER2-positive breast tumors in a genetic mouse model, ErbB2;mutant p53 mice30. This is in agreement with clinical trials on patients with Trastuzumab/Lapatinib-resistant disease25,26, the majority of whom (about 70%) also carry mutant TP5330. To extend these studies, we then analyzed Ganetespib in p53-wildtype Lapatinib-resistant ErbB2 mice and found that it was also effective (Fig. 1a). This once again illustrates that HER2 is a much more sensitive Hsp90 client than many others, e.g., mutant p5327,34. We then tested the long-term activity of Ganetespib in two different genetic mouse models of HER2-positive breast cancer, ErbB2 and Neu mice31,32. Both of these models are widely used and represent the two main etiologies of HER2-positive breast cancer in humans: constitutively active HER2 (ErbB2 mice32) and amplified wild-type HER2 (Neu mice31), which is found in the majority of human cases2,3,4. We found that Ganetespib treatment significantly prolonged—approximately doubled—the overall survival of both cohorts (Fig. 1b). Notably, however, both cohorts showed tumor regression/stagnation for an average of 4 weeks (Fig. 1c, d), after which the tumors started to regrow due to acquired Ganetespib resistance. We used these resistant tumors to interrogate the in vivo resistance mechanisms.
Upregulation of compensatory heat shock proteins in Ganetespib-resistant tumors
One predicted mechanism of therapeutic resistance to Hsp90 inhibitors is compensatory activation of alternative heat shock proteins by HSF1 (heat shock factor 1) transcription factor35,36,37. In naïve cells, Hsp90-bound HSF1 is sequestered in the cytoplasm, whereas Hsp90 inhibition activates HSF1, resulting in its nuclear translocation and a quick and transient transactivation of its HSP targets, so called inducible Hsp90A, Hsp70, and Hsp2735,36,37. Therefore, we hypothesized that Ganetespib resistance is mediated by upregulation of these or other HSPs and, consequently, stabilization of their oncogenic clients. Focusing on the Neu mouse model, since it better represents the most common etiology of the human disease2,3,4, we collected mammary tumors from Ganetespib-resistant and control mice at two different timepoints after the last Ganetespib dose (3 h and 24 h) and assessed the expression levels of various HSPs by quantitative RT-PCR (Fig. 2a–f). We used two types of negative controls: (1) untreated tumors (Control) and (2) Ganetespib-sensitive tumors that received Ganetespib only once, which would allow us to “subtract” Ganetespib’s short-term effects from the long-term resistance mechanisms. Not surprisingly, we found upregulation of all inducible HSPs that were tested, Hsp90A, Hsp70, Hsp27 and their co-factor Hsp40, at 3 h post-last Ganetespib dose and their decline at 24 h (Fig. 2c–f). Surprisingly, however, we also found a significant upregulation of so called constitutive HSPs, Hsp90B and Hsc70, whose transcription is driven by core transcription factors (Fig. 2a, b). Moreover, these constitutive HSPs, along with Hsp27, were significantly more upregulated in Ganetespib-resistant than in sensitive tumors (Fig. 2a–c, p < 0.01). In contrast, UDP glucuronosyltransferase 1A (UGT1A), previously implicated in primary—not acquired—Ganetespib resistance in colorectal cancer38, was not upregulated (Fig. 2g). In agreement, mass spectrometry confirmed upregulation of inducible Hsp90, Hsp70 and Hsp27, as well as the endoplasmic reticulum (ER)-specific constitutive Hsp90 and Hsc70 at the protein level (Fig. 2h). Note, protein upregulation was evident at both 3 h and 24 h timepoints, i.e., lasted longer than the mRNA upregulation. These data point towards a broadly upregulated compensatory HSP program as the most upstream mechanism of acquired Ganetespib resistance.
Upregulation of the phospho-RTKs/Akt signaling pathway in Ganetespib-resistant tumors
Many receptor tyrosine kinases (RTKs), besides HER2, are Hsp90 clients (www.picard.ch/downloads/Hsp90interactors.pdf) and also, frequently mediate therapeutic resistance, including in HER2-positive breast cancer21,22,30. Therefore, we analyzed a wide range of activated RTKs in a phospho-kinome assay on Ganetespib-resistant Neu tumor lysates. We found that 54% of all analyzed RTKs were significantly phospho-activated at both 3 h and 24 h after the last Ganetespib dose (21 out of 39), e.g., FGFRs, insulin receptor, insulin-like growth factor receptor, Tie-1/2, TrkA/B, EphA6-8 etc. (Fig. 3a, b). Note that many of them are known Hsp90 clients (Fig. 3b, asterisks). In addition, four RTKs were significantly phospho-activated only at 3 h and three RTKs only at 24 h, while three p-RTKs were significantly downregulated at 3 h, including p-HER2 (data not shown). These data indicate that a broad network of RTKs is stably activated in Ganetespib-resistant tumors.
RTKs, including HER2, signal via two major downstream pathways: the MAPK (Ras-Raf-Mek-Erk) pathway and the PI3K-Akt-mTOR pathway14,22. Therefore, we tested whether these pathways are involved in acquired Ganetespib resistance. First, we analyzed the levels of HER2 itself, which is a client of not only Hsp90A, but also Hsp70, Hsc70 and Hsp2736,39,40. We found that total HER2 and phospho-HER2 (Tyr 1248, Tyr 1221/1222) were still greatly suppressed in the resistant tumors at 3 h and 24 h after the last Ganetespib dose (Fig. 3c, d, h). Furthermore, the levels of HER2 effectors Erk1 and Raf1—also Hsp90 clients—were significantly reduced in the resistant tumors, as was phospho-Raf1 (Fig. 3h, i). As a positive control, known Hsp90 clients Abl1, Map4k4 and Tesk1 were also downregulated (Fig. 3h). In contrast, the levels of Akt/Protein kinase B—the other major RTK effector—were barely affected in Ganetespib-resistant tumors (Fig. 3c, e), even though Akt is a well-established Hsp90 client and therefore, is expected to be downregulated by Ganetespib13,41. Moreover, activated p-Akt, at Thr 308 and Ser 473, were significantly upregulated in the resistant tumors at 3 h and 24 h, respectively (Fig. 3c, f, g). Altogether, this suggests that Akt/p-Akt represents an essential signaling effector of the upregulated p-RTK network as a mechanism of acquired Ganetespib resistance.
Pharmacological inhibition of Akt significantly delays acquired Ganetespib resistance
To test directly whether Akt/p-Akt mediates acquired Ganetespib resistance in vivo, we combined Ganetespib with a suboptimal dose of a pan-Akt inhibitor MK-220642,43. MK-2206 is an orally bioavailable, highly selective, allosteric Akt1/2/3 inhibitor that—in contrast to ATP-competitive inhibitors—does not cause compensatory phospho-activation of Akt44, EGFR and HER245. MK-2206 has been evaluated in 50 Phase I/II clinical trials, 13 of which are on breast cancer including eight on HER2-positive breast cancer. In combination with Trastuzumab or Lapatinib, MK-2206 showed favorable safety and efficacy in patients who progressed on HER2-targeted therapies46,47,48. Using allografts of primary Neu tumor cells into mammary fat pads of immunodeficient female recipients, we found that MK-2206 by itself did not significantly affect progression-free survival at the low dose we used (20 mg/kg three times a week; its half-life is 40 h42,43) (Fig. 4a, c). Ganetespib alone was effective for 9 weeks (median progression-free survival), however, its combination with MK-2206 delayed acquired resistance to a median of 13.5 weeks, i.e., by 50% (Fig. 4b, d). These results indicate that Akt indeed is a principal mediator of acquired Ganetespib resistance in vivo. Based on these data, we propose that clinically, co-targeting Akt together with Ganetespib may represent a therapeutically superior regimen for the treatment of HER2-positive breast cancer.
Discussion
In this study we used autochthonous, immune-competent mouse models to investigate the mechanisms of acquired Ganetespib resistance in vivo. We have focused on HER2-positive breast cancer since it has proven to be highly sensitive to Hsp90 inhibition in the clinic25,26,27. We propose that in naïve HER2-positive breast cancer cells, Ganetespib is highly effective by curbing the Hsp90 support of HER2 itself, its co-receptor EGFR and the effectors Raf, Erk1, Akt, mTOR (Fig. 5a, b)14. However, Ganetespib eventually evokes acquired resistance by upregulating compensatory HSPs, especially constitutive Hsp90B and Hsc70, as well as inducible Hsp27 (Fig. 5c). In agreement, Hsp27 was previously implicated in the acquired resistance to Hsp90 inhibitor 17-AAG in hepatocellular carcinoma49. Although some activation of these HSPs is also observed in Ganetespib-sensitive controls (Fig. 2a–c), it is significantly stronger in the resistant tumors, suggesting an HSP transcriptome rewiring that is actively selected for in the resistant tumor cells. Furthermore, upregulation of wild-type and mutant RTKs is a frequent mechanism of therapeutic resistance in HER2-positive breast cancer21,22. Indeed, we find a wide range of RTKs to be phospho-activated in Ganetespib-resistant tumors (Fig. 3a, b) and propose that their stability and/or activity is actively supported by the upregulated HSPs (Fig. 5c). We did not directly test this, because (1) many RTKs are already established as HSP clients (www.picard.ch/downloads/Hsp90interactors.pdf; asterisks in Fig. 3b)50,51, (2) it would be technically challenging to simultaneously intercept several compensatory HSPs at once to test this idea, especially in our preferred in vivo system, and (3) we sought to find a druggable node downstream (not upstream) of activated p-RTKs as the most clinically impactful goal. To this end, we found that downstream of p-RTKs, the MAPK pathway is still effectively shut down in the resistant tumors (Fig. 3c, d, h, i). In contrast, the Akt pathway is active, as evident by stabilized and phospho-activated Akt (Fig. 3c, e–g) and, most importantly, by the sensitivity of the resistant tumors to pharmacological inhibition of Akt (Fig. 4).
Akt activation is one of the most frequent alterations in human cancers and is associated with enhanced tumor cell survival, proliferation, invasiveness, and poor patient outcomes in breast and other cancers52,53,54. Notably, Akt activation is also associated with resistance to chemo-, radiation and targeted therapies, including in HER2-positive breast cancer19,20,22,55,56,57,58,59,60,61. Akt is usually activated in cancer cells by gene amplification, activating mutations or loss of its inhibitor, tumor suppressor PTEN (phosphatase and tensin homolog)52,55,56,57. On the other hand, Akt is an Hsp90 client and would be expected to be destabilized by Ganetespib13,41. Our finding that, in contrast to HER2, Akt is not destabilized suggests that either Akt is less sensitive to Hsp90 inhibition than HER2, as previously reported14,62,63,64, or that Akt is actively stabilized in the resistant tumors by upregulated HSPs, or both. Supporting an active stabilization mechanism by Hsc70 and/or Hsp27, Akt was previously shown to be stabilized by Hsc70 and destabilized by Hsp7065 and also, activated by Hsp27, both directly and indirectly, via loss of PTEN66,67,68,69,70. Stabilized Akt is likely activated by the upregulated p-RTKs14,22. We did not test these mechanisms directly since it would be technically challenging to intercept all of the identified HSPs (≥3), p-RTKs (≥21) and/or their combinations simultaneously or one by one, especially in vivo. Although this is a limitation of our study, it does not lessen the identified critical role of Akt.
Here, we focused on HER2-positive breast cancer since it has emerged as one of the malignancies that can benefit from Hsp90 inhibition the most25,26,27. Of note, our findings are in agreement with a previous study implicating Akt, as well as the MAPK pathway, in acquired Ganetespib resistance in KRAS-mutant non-small-cell lung cancer71,72. In contrast, another study on triple-negative breast cancer rather found that the JAK-STAT pathway is involved in acquired Ganetespib resistance73. These differences may be explained by the different cancer types investigated, as well as by the fact that the aforementioned studies analyzed only cultured cells, but not genetic models71,72,73. While cell lines-based studies have their benefits, genetic mouse models present a unique opportunity to analyze drug resistance mechanisms in the native tumor microenvironment and in the presence of the functional immune system (known to mediate the effects of Hsp90 inhibitors74,75,76,77), which we made use of.
In sum, here we report that HER2-positive breast cancer, a malignancy highly sensitive to Hsp90 inhibition, acquires Ganetespib resistance by upregulating compensatory HSPs and the RTK-Akt pathway. Since pharmacological inhibition of Akt significantly delays acquired Ganetespib resistance, this establishes Akt as a unifying druggable node downstream of the widely activated HSP and RTK programs and suggests that Akt co-targeting together with Ganetespib may represent a therapeutically superior strategy for the treatment of HER2-positive breast cancer. In addition, our findings may prove valuable in other HER2-driven malignancies, e.g., colorectal, gastroesophageal and lung cancer78,79,80.
References
Jemal, A. et al. Global cancer statistics. CA Cancer J. Clin. 61, 69–90 (2011).
Wolff, A. C. et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J. Clin. Oncol. 31, 3997–4013 (2013).
Varga, Z., Noske, A., Ramach, C., Padberg, B. & Moch, H. Assessment of HER2 status in breast cancer: overall positivity rate and accuracy by fluorescence in situ hybridization and immunohistochemistry in a single institution over 12 years: a quality control study. BMC Cancer 13, 615 (2013).
Mitri, Z., Constantine, T. & O’Regan, R. The HER2 receptor in breast cancer: pathophysiology, clinical use, and new advances in therapy. Chemother. Res. Pract. 2012, 743193 (2012).
Howlader, N. et al. US incidence of breast cancer subtypes defined by joint hormone receptor and HER2 status. J. Natl Cancer Inst. 106, https://doi.org/10.1093/jnci/dju055 (2014).
Slamon, D. J. et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235, 177–182 (1987).
Carr, J. A. et al. The association of HER-2/neu amplification with breast cancer recurrence. Arch. Surg. 135, 1469–1474 (2000).
Ross, J. S. & Fletcher, J. A. The HER-2/neu oncogene in breast cancer: prognostic factor, predictive factor, and target for therapy. Stem Cells 16, 413–428 (1998).
Curtis, C. et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 486, 346–352 (2012).
Nahta, R., Yu, D., Hung, M. C., Hortobagyi, G. N. & Esteva, F. J. Mechanisms of disease: understanding resistance to HER2-targeted therapy in human breast cancer. Nat. Clin. Pract. Oncol. 3, 269–280 (2006).
Nielsen, D. L., Kumler, I., Palshof, J. A. & Andersson, M. Efficacy of HER2-targeted therapy in metastatic breast cancer. Monoclonal antibodies and tyrosine kinase inhibitors. Breast (Edinb., Scotl.) 22, 1–12 (2013).
Zagouri, F. et al. Hsp90 inhibitors in breast cancer: a systematic review. Breast 22, 569–578 (2013).
Jhaveri, K. et al. Heat shock protein 90 inhibitors in the treatment of cancer: current status and future directions. Expert Opin. Invest. Drugs 23, 611–628 (2014).
Lee, H. et al. Ganetespib targets multiple levels of the receptor tyrosine kinase signaling cascade and preferentially inhibits ErbB2-overexpressing breast cancer cells. Sci. Rep. 8, 6829 (2018).
Zuehlke, A. & Johnson, J. L. Hsp90 and co-chaperones twist the functions of diverse client proteins. Biopolymers 93, 211–217 (2010).
Biebl, M. M. & Buchner, J. Structure, function, and regulation of the hsp90 machinery. Cold Spring Harb. Perspect. Biol. 11, https://doi.org/10.1101/cshperspect.a034017 (2019).
Whitesell, L. & Lindquist, S. L. HSP90 and the chaperoning of cancer. Nat. Rev. Cancer 5, 761–772 (2005).
Trepel, J., Mollapour, M., Giaccone, G. & Neckers, L. Targeting the dynamic HSP90 complex in cancer. Nat. Rev. Cancer 10, 537–549 (2010).
Kirkegaard, T. et al. AKT activation predicts outcome in breast cancer patients treated with tamoxifen. J. Pathol. 207, 139–146 (2005).
Wilks, S. T. Potential of overcoming resistance to HER2-targeted therapies through the PI3K/Akt/mTOR pathway. Breast (Edinb., Scotl.) 24, 548–555 (2015).
D’Amato, V. et al. Mechanisms of lapatinib resistance in HER2-driven breast cancer. Cancer Treat. Rev. 41, 877–883 (2015).
Vernieri, C. et al. Resistance mechanisms to anti-HER2 therapies in HER2-positive breast cancer: current knowledge, new research directions and therapeutic perspectives. Crit. Rev. Oncol. Hematol. 139, 53–66 (2019).
Harrison, P. T. & Huang, P. H. Exploiting vulnerabilities in cancer signalling networks to combat targeted therapy resistance. Essays Biochem. 62, 583–593 (2018).
Workman, P., Burrows, F., Neckers, L. & Rosen, N. Drugging the cancer chaperone HSP90: combinatorial therapeutic exploitation of oncogene addiction and tumor stress. Ann. N. Y. Acad. Sci. 1113, 202–216 (2007).
Jhaveri, K. et al. A phase II open-label study of ganetespib, a novel heat shock protein 90 inhibitor for patients with metastatic breast cancer. Clin. Breast Cancer 14, 154–160 (2014).
Jhaveri, K. et al. A phase I trial of ganetespib in combination with paclitaxel and trastuzumab in patients with human epidermal growth factor receptor-2 (HER2)-positive metastatic breast cancer. Breast Cancer Res. 19, 89 (2017).
Jhaveri, K. et al. Biomarkers that predict sensitivity to heat shock protein 90 inhibitors. Clin. Breast Cancer 16, 276–283 (2016).
Workman, P. & van Montfort, R. EML4-ALK fusions: propelling cancer but creating exploitable chaperone dependence. Cancer Discov. 4, 642–645 (2014).
Socinski, M. A. et al. A multicenter phase II study of ganetespib monotherapy in patients with genotypically defined advanced non-small cell lung cancer. Clin. Cancer Res. 19, 3068–3077 (2013).
Yallowitz, A., Ghaleb, A., Garcia, L., Alexandrova, E. M. & Marchenko, N. Heat shock factor 1 confers resistance to lapatinib in ERBB2-positive breast cancer cells. Cell Death Dis. 9, 621 (2018).
Guy, C. T. et al. Expression of the neu protooncogene in the mammary epithelium of transgenic mice induces metastatic disease. Proc. Natl Acad. Sci. USA 89, 10578–10582 (1992).
Muller, W. J., Sinn, E., Pattengale, P. K., Wallace, R. & Leder, P. Single-step induction of mammary adenocarcinoma in transgenic mice bearing the activated c-neu oncogene. Cell 54, 105–115 (1988).
Johnson, S., Chen, H. & Lo, P. K. Isolation of cancer epithelial cells from mouse mammary tumors. Bio Protoc 3, https://doi.org/10.21769/bioprotoc.326 (2013).
Alexandrova, E. M. et al. Improving survival by exploiting tumour dependence on stabilized mutant p53 for treatment. Nature 523, 352–356 (2015).
Zou, J., Guo, Y., Guettouche, T., Smith, D. F. & Voellmy, R. Repression of heat shock transcription factor HSF1 activation by HSP90 (HSP90 complex) that forms a stress-sensitive complex with HSF1. Cell 94, 471–480 (1998).
Ciocca, D. R., Arrigo, A. P. & Calderwood, S. K. Heat shock proteins and heat shock factor 1 in carcinogenesis and tumor development: an update. Arch. Toxicol. 87, 19–48 (2013).
Kijima, T. et al. HSP90 inhibitors disrupt a transient HSP90-HSF1 interaction and identify a noncanonical model of HSP90-mediated HSF1 regulation. Sci. Rep. 8, 6976 (2018).
Landmann, H. et al. UDP glucuronosyltransferase 1A expression levels determine the response of colorectal cancer cells to the heat shock protein 90 inhibitor ganetespib. Cell Death Dis. 5, e1411 (2014).
Powers, M. V., Clarke, P. A. & Workman, P. Dual targeting of HSC70 and HSP72 inhibits HSP90 function and induces tumor-specific apoptosis. Cancer Cell 14, 250–262 (2008).
Kang, S. H. et al. Upregulated HSP27 in human breast cancer cells reduces Herceptin susceptibility by increasing Her2 protein stability. BMC Cancer 8, 286 (2008).
Solit, D. B., Basso, A. D., Olshen, A. B., Scher, H. I. & Rosen, N. Inhibition of heat shock protein 90 function down-regulates Akt kinase and sensitizes tumors to Taxol. Cancer Res. 63, 2139–2144 (2003).
Hirai, H. et al. MK-2206, an allosteric Akt inhibitor, enhances antitumor efficacy by standard chemotherapeutic agents or molecular targeted drugs in vitro and in vivo. Mol. Cancer Therap. 9, 1956–1967 (2010).
Carpenter, R. L. et al. Combined inhibition of AKT and HSF1 suppresses breast cancer stem cells and tumor growth. Oncotarget 8, 73947–73963 (2017).
Okuzumi, T. et al. Inhibitor hijacking of Akt activation. Nat. Chem. Biol. 5, 484–493 (2009).
Crafter, C. et al. Combining AZD8931, a novel EGFR/HER2/HER3 signalling inhibitor, with AZD5363 limits AKT inhibitor induced feedback and enhances antitumour efficacy in HER2-amplified breast cancer models. Int. J. Oncol. 47, 446–454 (2015).
Hudis, C. et al. A phase 1 study evaluating the combination of an allosteric AKT inhibitor (MK-2206) and trastuzumab in patients with HER2-positive solid tumors. Breast Cancer Res. 15, R110 (2013).
Wisinski, K. B. et al. Phase I study of an AKT inhibitor (MK-2206) combined with lapatinib in adult solid tumors followed by dose expansion in advanced HER2+ breast cancer. Clin. Cancer Res. 22, 2659–2667 (2016).
Chien, A. J. et al. A phase 1b study of the Akt-inhibitor MK-2206 in combination with weekly paclitaxel and trastuzumab in patients with advanced HER2-amplified solid tumor malignancies. Breast Cancer Res. Treat. 155, 521–530 (2016).
Cui, Y. et al. HSP27 expression levels are associated with the sensitivity of hepatocellular carcinoma cells to 17-allylamino-17-demethoxygeldanamycin. Future Oncol. (Lond., Engl.) 9, 411–418 (2013).
Jiao, Y., Ou, W., Meng, F., Zhou, H. & Wang, A. Targeting HSP90 in ovarian cancers with multiple receptor tyrosine kinase coactivation. Mol. Cancer 10, 125 (2011).
Wang, M. et al. Development of heat shock protein (Hsp90) inhibitors to combat resistance to tyrosine kinase inhibitors through Hsp90-kinase interactions. J. Med. Chem. 59, 5563–5586 (2016).
Panigrahi, A. R. et al. The role of PTEN and its signalling pathways, including AKT, in breast cancer; an assessment of relationships with other prognostic factors and with outcome. J. Pathol. 204, 93–100 (2004).
Carmona, F. J. et al. AKT signaling in ERBB2-amplified breast cancer. Pharmacol. Ther. 158, 63–70 (2016).
Wang, Q., Chen, X. & Hay, N. Akt as a target for cancer therapy: more is not always better (lessons from studies in mice). Br. J. Cancer 117, 159–163 (2017).
Wendel, H. G. et al. Survival signalling by Akt and eIF4E in oncogenesis and cancer therapy. Nature 428, 332–337 (2004).
Kim, D. et al. AKT/PKB signaling mechanisms in cancer and chemoresistance. Front. Biosci. 10, 975–987 (2005).
Avan, A., Narayan, R., Giovannetti, E. & Peters, G. J. Role of Akt signaling in resistance to DNA-targeted therapy. World J. Clin. Oncol. 7, 352–369 (2016).
Nagata, Y. et al. PTEN activation contributes to tumor inhibition by trastuzumab, and loss of PTEN predicts trastuzumab resistance in patients. Cancer Cell 6, 117–127 (2004).
Berns, K. et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Cancer Cell 12, 395–402 (2007).
Miller, T. W. et al. Hyperactivation of phosphatidylinositol-3 kinase promotes escape from hormone dependence in estrogen receptor-positive human breast cancer. J. Clin. Investig. 120, 2406–2413 (2010).
Musgrove, E. A. & Sutherland, R. L. Biological determinants of endocrine resistance in breast cancer. Nat. Rev. Cancer 9, 631–643 (2009).
Basso, A. D. et al. Akt forms an intracellular complex with heat shock protein 90 (Hsp90) and Cdc37 and is destabilized by inhibitors of Hsp90 function. J. Biol. Chem. 277, 39858–39866 (2002).
Basso, A. D., Solit, D. B., Munster, P. N. & Rosen, N. Ansamycin antibiotics inhibit Akt activation and cyclin D expression in breast cancer cells that overexpress HER2. Oncogene 21, 1159–1166 (2002).
Solit, D. B. et al. 17-Allylamino-17-demethoxygeldanamycin induces the degradation of androgen receptor and HER-2/neu and inhibits the growth of prostate cancer xenografts. Clin. Cancer Res. 8, 986–993 (2002).
Koren, J. 3rd et al. Facilitating Akt clearance via manipulation of Hsp70 activity and levels. J. Biol. Chem. 285, 2498–2505 (2010).
Rane, M. J. et al. Heat shock protein 27 controls apoptosis by regulating Akt activation. J. Biol. Chem. 278, 27828–27835 (2003).
Jomary, C., Cullen, J. & Jones, S. E. Inactivation of the Akt survival pathway during photoreceptor apoptosis in the retinal degeneration mouse. Investig. Ophthalmol. Vis. Sci. 47, 1620–1629 (2006).
Wu, R. et al. Hsp27 regulates Akt activation and polymorphonuclear leukocyte apoptosis by scaffolding MK2 to Akt signal complex. J. Biol. Chem. 282, 21598–21608 (2007).
Zhang, Y. et al. A Targetable molecular chaperone Hsp27 confers aggressiveness in hepatocellular carcinoma. Theranostics 6, 558–570 (2016).
Cayado-Gutierrez, N. et al. Downregulation of Hsp27 (HSPB1) in MCF-7 human breast cancer cells induces upregulation of PTEN. Cell Stress Chaperones 18, 243–249 (2013).
Chatterjee, S., Huang, E. H., Christie, I., Kurland, B. F. & Burns, T. F. Acquired resistance to the Hsp90 inhibitor, ganetespib, in KRAS-mutant NSCLC is mediated via reactivation of the ERK-p90RSK-mTOR signaling network. Mol. Cancer Ther. 16, 793–804 (2017).
Chatterjee, S., Huang, E. H., Christie, I. & Burns, T. F. Reactivation of the p90RSK-CDC25C pathway leads to bypass of the ganetespib-induced G2-M arrest and mediates acquired resistance to ganetespib in KRAS-mutant NSCLC. Mol. Cancer Ther. 16, 1658–1668 (2017).
Mumin, N. H. et al. Overcoming acquired resistance to HSP90 inhibition by targeting JAK-STAT signalling in triple-negative breast cancer. BMC Cancer 19, 102 (2019).
Mbofung, R. M. et al. HSP90 inhibition enhances cancer immunotherapy by upregulating interferon response genes. Nat. Commun. 8, 451 (2017).
Huyan, T. et al. Heat shock protein 90 inhibitors induce functional inhibition of human natural killer cells in a dose-dependent manner. Immunopharmacol. Immunotoxicol. 38, 77–86 (2016).
Tukaj, S. et al. Immunomodulatory effects of heat shock protein 90 inhibition on humoral immune responses. Exp. Dermatol. 23, 585–590 (2014).
Rao, A. et al. Combination therapy with HSP90 inhibitor 17-DMAG reconditions the tumor microenvironment to improve recruitment of therapeutic T cells. Cancer Res. 72, 3196–3206 (2012).
Kavuri, S. M. et al. HER2 activating mutations are targets for colorectal cancer treatment. Cancer Discov. 5, 832–841 (2015).
Vakiani, E. HER2 testing in gastric and gastroesophageal adenocarcinomas. Adv. Anat. Pathol. 22, 194–201 (2015).
Mar, N., Vredenburgh, J. J. & Wasser, J. S. Targeting HER2 in the treatment of non-small cell lung cancer. Lung Cancer (Amst., Neth.) 87, 220–225 (2015).
Acknowledgements
We are thankful to Stony Brook University’s Proteomics Core Facility for performing the mass spectrometry analysis.
Funding
This work was supported by NIH/NCI grant K22CA190653 to EMA and Stony Brook Cancer Center startup funds to EMA.
Author information
Authors and Affiliations
Contributions
C.E.E. contributed to the study design, performed data acquisition, analysis and interpretation, critically read the manuscript and approved its final version. J.D.H. contributed to the study design, performed data acquisition, critically read the manuscript and approved its final version. E.M.A. designed the study, performed data acquisition, analysis and interpretation, and wrote the manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethics statement
No human tissues or samples were used. All mouse work was approved by the Stony Brook University Institutional Animal Care and Use Committee (IACUC), protocol number 924666.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Edited by S. Tait
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Eyermann, C.E., Haley, J.D. & Alexandrova, E.M. The HSP-RTK-Akt axis mediates acquired resistance to Ganetespib in HER2-positive breast cancer. Cell Death Dis 12, 126 (2021). https://doi.org/10.1038/s41419-021-03414-3
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41419-021-03414-3
- Springer Nature Limited