The Hsp90 inhibitor 17-AAG enhanced significantly the radiation sensitivity and induced apoptosis in oral squamous cell carcinoma cell line SAS/neo which has a wild-type p53. On the other hand, the radiation sensitizing effect of 17-AAG was limited in the SAS/Trp248 cells which has a mutated p53 (59). 17-AAG was also shown to inhibit ATP-binding cassette sub-family G member 2 (ABCG2) upregulation, thereby reversing the ABCG2-mediated multidrug resistance (MDR) and in this regard can be used as a chemosensitizer for the treatment of esophageal cancer (60). Furthermore, 17-AAG sensitized esophageal cancer cells to γ-photon radiation, probably through its ability to affect a range of signalling components simultaneously, for example both EGF and IGF-1 (61).

Hsp90 inhibitors, IPI-493 as well as IPI-504, revealed antitumor activity and induced receptor tyrosine kinase KIT downregulation in xenograft model of gastrointestinal stromal tumor (GIST) with heterogeneous KIT mutations (62,63). IPI-493 synergized with tyrosine kinase inhibitors, that are commonly used for the treatment of advanced or imatinib-resistant GIST (63). Treatment effects of IPI-504 were also enhanced and/or more potent in combination with imatinib or sunitinib (62). Similarly a combination of imatinib and Hsp90 inhibitor AT13387 treatment in the imatinib-resistant GIST430 model significantly enhanced tumor growth inhibition over the monotherapies. These results, as well as good tolerance of AT13387, suggest this Hsp90 inhibitor as an excellent candidate for clinical testing in gastrointestinal stromal tumors in combination with imatinib (64). Moreover, by blocking Hsp90, Lang et al (65) disrupted rapamycin-induced activation of alternative signaling pathways in hepatocellular carcinoma and substantially improved the growth-inhibitory effects of mTOR inhibition in vivo.

Hsp90 inhibition may be also an attractive target for innovative treatment of gastroenteropancreatic neuroendocrine tumors alone or in combination with either IGF-1R or mTOR inhibitors (66). Unusually, Hsp90 inhibitor 17-AAG was more responsive in colon cancer cells with BRAF^V600E mutation than selective BRAF^V600E inhibitor PLX4720, probably due to the multiple oncogenic proteins that are Hsp90 clients (67). Indeed, the inhibition of Hsp90 with 17-DMAG abrogated the invasive properties of colon cancer cells and modulated the expression of the antimetastatic factor activating transcription factor-3 and improved the efficacy of oxaliplatin (68). Furthermore, 17-AAG interventions directed to cell cycle arrest in G1 may potentiate oxaliplatin activity in colon cancer cells with impaired p53 (69) and may be critical for its ability to reverse cisplatin resistance (70). Similarly, GA reversed topotecan induced upregulation of the anti-apoptotic protein Bcl-2 in both p53^+/+ and p53^−/− HCT116 cells and sensitised human colon cancer cells to topoisomerase I poisons via depletion of key anti-apoptotic and cell cycle checkpoint proteins (71). On the other hand, the combination of an active metabolite of irinotecan SN-38 and 17-AAG was shown to be synergistic in p53-null, but not in parental HCT116 cells by median effect. Taken together, 17-AAG specifically inhibited the G(2)/M checkpoint in p53-defective cells by downregulation of checkpoint kinases Chk1 and Wee1 (72). Similarly, GA enhanced the effect of ionizing radiation in p53 and p21 null cells (73).

The inhibition of Hsp90 with GA destabilised Cdc25A independent of Chk1/2, whereas the standard drug for pancreas carcinoma treatment, gemcitabine caused Cdc25A degradation through the activation of Chk2. Both agents applied together additively inhibited the expression of Cdc25A and the proliferation of pancreatic carcinoma cells, as well as reduced the resistance of pancreatic carcinoma cells to treatment with gemcitabine (74). Indeed, the combined therapy of 17-DMAG and Akt inhibitor perifosine overcame tumor growth and resistance induced by bone marrow stromal cells and endothelial cells as well as the proliferative effect of exogenous interleukin-6, insulin-like growth factor-I, and vascular endothelial growth factor. The combination induced also apoptosis and growth inhibition in endothelial cells and inhibited angiogenesis (75). Moreover, short interfering RNA (siRNA) was used to knockdown ErbB3 in the resistant cell line AsPC1. Whereas individual treatments with siRNA to ErbB3 or 17DMAG had no effect on radiosensitivity, the combination, which reduced both ErbB2 and ErbB3, resulted in a significant enhancement in AsPC1 radiosensitivity (76). Interestingly, the dietary component from broccoli and broccoli sprouts sulforaphane also potentiated the efficacy of 17-AAG against pancreatic cancer through enhanced abrogation of Hsp90 function (77). For the complete list of oral, esophageal, pancreatic, gastrointestinal and colon cancer cell response to combination of Hsp90 inhibitor and other therapy (treatment) in vitro and/or in vivo see Table II.

Hsp90 inhibitor IPI-504, induced tumor regression in aggressive Nf1-deficient malignancies and KRas/p53-mutant lung cancer mouse models, but only when combined with rapamycin (78). Combinations of low-dose of non-ansamycin inhibitor of Hsp90, ganetespib (STA-9090 a unique resorcinolic triazolone) with MEK or PI3K/mTOR inhibitors, resulted in superior cytotoxic activity than single agents alone in a subset of mutant KRAS cells, and the anti-tumor efficacy of STA-9090 was potentiated by cotreatment with the PI3K/mTOR inhibitor BEZ235 in A549 xenografts in vivo (79). Similar combination of STA-9090 with paclitaxel, docetaxel or vincristine resulted in synergistic antiproliferative effects in the H1975 cells in vitro as wells as enhanced tumor growth inhibition observed in combination with paclitaxel and tumor regressions seen with docetaxel (80). Furthermore, Hsp90 inhibitors have overcame the resistance to EGFR tyrosine kinase inhibitors in subcutaneous xenograft models of H1975 and A549 (81) and enhanced the antitumor activity of paclitaxel and camptothecin, respectively, in breast and colorectal tumor models (82). Hsp90 inhibition overcame the hepatocyte growth factor-triggered resitance to EGFR tyrosine kinase inhibitors and resulted in more successful treatment of patients with EGFR-mutant lung cancers (83). The effectiveness of combined therapy of EGFR tyrosine kinase inhibitors and Hsp90 inhibitors was also confirmed in the treatment of lung cancers co-driven by mutant EGFR containing T790M and MET (84,85). Moreover, 17-AAG decreased the etoposide-induced p38 MAPK-mediated ERCC1 expression (DNA repair capacity) (86) and downregulated the cisplatin-induced thymidine phosphorylase (a key enzyme in pyrimidine nucleoside salvage pathway) expression and Erk1/2 and Akt activation (87) what sensitized lung cancer cells to etoposide and cisplatin, respectively.

Hsp90 inhibitors were also effective in treating crizotinib-resistant tumors harboring secondary gatekeeper mutations within the ALK TK domain (88). Furthermore, Normant et al (89) observed partial responses to administration of IPI-504 as a single agent in a phase II clinical trial in patients with NSCLC, specifically in patients that carry an ALK rearrangement. The experimental data of Solit et al (90) suggested that Hsp90 inhibition in combination with chemotherapy may represent an effective treatment strategy for patients, whose tumors express EGFR kinase domain mutations, including those with de novo and acquired resistance to EGFR tyrosine kinase inhibitors. Acquired resistance limits the efficacy of ALK inhibitors, such as crizotinib and TAE684, however, combining inhibitors of ALK and Hsp90 resulted in synergistic cytotoxicity (91,92).

Hsp90 inhibitor 17-DMAG, as well as purine-scaffold inhibitor PU-H71, were shown to be the most effective as radiosensitizer of lung cancer cells, when administered before radiation, because of its suppression of DNA repair at multiple levels, including BER and ATM-regulated pathways (93,94). Recent results of Kim and Pyo (95) show that combined treatment of 17-AAG and celecoxib at clinically relevant concentrations may significantly enhance the therapeutic efficacy of ionizing radiation therapy. Moreover, very recent combination of 17-DMAG and histone deacetylase inhibitor belinostat synergistically inhibited in vitro proliferation of selected panel of 12 NSCLC cell lines. Importantly, both agents and their combination almost completely prevented TKI-resistant tumor formation (EGFR T790M mutation) in a xenograft model (96). For the above, and even further studies with the response of lung cancer cells to combination of Hsp90 inhibitor and another therapy see Table III.

Hsp90 inhibitor IPI-504 resulted in a marked decrease in the levels of HER-2, Akt, p-Akt, and p-MAPK in trastuzumab-resistant xenografts as early as 12 h after a single dose of IPI-504 inhibitor. IPI-504-mediated Hsp90 inhibition may represent, together with other Hsp90 inhibitors, a novel therapeutic approach in HER-2 overexpressing and/or trastuzumab refractory HER-2-positive cancer (97–100), and provide a promising strategy to overcome the development of resistance against trastuzumab. Very promising strategy to overcome the development of trastuzumab resistance in breast cancer patients seems to be a combination of the Hsp90 inhibitor and histone deacetylase 6 inhibitor carbamazepine (101). GA synergized with carbamazepine to promote HER-2 degradation and inhibited breast cancer cell proliferation. Solit et al (102) found that the inhibition of Hsp90 function sensitized breast cancer xenografts to Taxol through downregulation of Akt kinase. Hsp90 inhibitors may be an effective therapy also to treat aromatase inhibitor-resistant breast cancers and improved efficacy can be achieved by combined use of an Hsp90 inhibitor and an Akt inhibitor (103,104). Moreover, Pashtan et al (105) demonstrated that Hsp90 inhibitor 17-AAG inhibiting multiple signaling activators including ErbB and Src kinases, it does not permit oncogene switching and resulted in a more prolonged and robust inhibition of downstream signaling pathways in breast cancer cells, than the individual tyrosine kinase inhibitors. In this way, Hsp90 inhibitors can prevent breast cancer cells from tyrosine kinase inhibition (105). Another way, to sensitize cancer cells is elimination of RIP1 protein, which can support cell survival due to activation of NF-κB. In this regard, 17-DMAG markedly reduced RIP1 expression and sensitized breast tumor cells to tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis (eliminate the resistance to TRAIL) (106). A recent study of Stecklein et al (107) demonstrated the role of BRCA1, which is associated with the resistance to platinum-based chemotherapeutics and poly(ADP ribose) polymerase (PARP) inhibitors in regulating damage-associated checkpoint and repair responses to Hsp90 inhibitors. The above study pointed out BRCA1 as a clinically relevant target for enhancing sensitivity in refractory and/or resistant malignancies. Furthermore, complete regression of aggressive breast tumors in mice was observed after the combination of anti-angiogenic stress imposed by loss of Id protein and the inactivation of the HIF-1α chaperone Hsp90 (108).

Interestingly, ovarian cancer cells overexpressing HER2 showed 5-fold increased sensitivity to the Hsp90 molecular chaperone inhibitor GA. In contrast, HER2-overexpressing cells showed statistically significant resistance to cisplatin, PI3K inhibitor LY294002 and tyrosine kinase inhibitor emodin (109). On the other hand, GA and 17-AAG may sensitize ovarian cancer cells to cisplatin (110) and paclitaxel (111), respectively, particularly those tumors in which resistance is driven by HER-2 and/or p-Akt.

Pharmacological inhibition of the cell cycle-dependent kinase (essential for G2-M checkpoint) Wee1 and its molecular silencing with siRNA uniformly increased apoptotic activity of the Hsp90 inhibitor 17-AAG in vitro (112). Moreover, Wee1 inhibition synergized with any one of the clinically evaluated Hsp90 inhibitors to inhibit an androgen-independent and invasive human prostate carcinoma cell line PC3 in vitro and in vivo (113). Hsp90 inhibitors trigger a heat shock response associated with increased expression of Hsp90, Hsp70, Hsp27, and clusterin that attenuates drug effectiveness in the treatment of castrate-resistant prostate cancer. However, clusterin inhibitor OGX-011 revealed markedly potentiated antitumor efficacy in xenograft models of human castrate-resistant prostate cancer, leading to an 80% inhibition of tumor growth with prolonged survival compared with Hsp90 inhibitor monotherapy (114). Furthermore, administration of 17-AAG maintained androgen-sensitivity, prevented nuclear localization of endogenous androgen receptor and delayed the progression of LuCaP35 xenograft tumors to castration resistance. Thus, it appears that targeting Hsp90 may lead to new approaches to prevent and/or treat castration-resistant prostate cancer (115). Alternatively, co-administration of hyperthermia and 17-AAG released from plasmonic matrices was a powerful approach for the ablation of malignant cells and can be used in the form of similar combinations of nanoparticles and chemotherapeutic drugs for a variety of malignancies (116). For the complete list of breast, ovarian and prostate cancer cell response to combination of Hsp90 inhibitor and another therapy (treatment) in vitro and/or in vivo see Table IV.

Lamottke et al (117) demonstrated that inhibition of Hsp90 by small molecular mass inhibitors induced cell death in multiple myeloma. They analyzed the effects of the novel, orally bioavailable Hsp90 inhibitor NVP-Hsp990 on multiple myeloma cell proliferation and survival. NVP-Hsp990 inhibitor led to a significant reduction in myeloma cell viability and induced G2 cell cycle arrest, degradation of caspase-8 and caspase-3, and induction of apoptosis. Inhibition of the Hsp90 ATPase activity was accompanied by the degradation of multiple myeloma phospho-Akt and phospho-Erk1/2 and upregulation of Hsp70. In this regard, PI3K/mTOR inhibitor in combination with NVP-Hsp990 mentioned above, rendered the Hsp90 blockade-mediated stress response ineffective and considerably increased the anti-multiple myeloma toxicity (118). Similarly, Chatterjee et al (119) demonstrated that the knockdown of Hsp72 and/or Hsp73 or treatment with VER-155008, decreased protein levels of Hsp90-chaperone clients affecting multiple oncogenic signaling pathways, and acted synergistically with the novel Hsp90 inhibitor NVP-AUY922 in myeloma cell death induction. Inhibition of the PI3K/Akt/GSK3β pathway with siRNA or PI103 decreased expression of the heat shock transcription factor 1 and downregulated constitutive and inducible Hsp70 expression (119). Moreover, NVP-AUY922 in combination with histone deacetylase inhibitors SAHA, NVP-LBH589, melphalan or doxorubicin resulted in synergistic inhibition of viability, with strong synergy (combination index <0.3) in the case of melphalan. Importantly, resistance of the RPMI-8226 cell line and relative resistance of some primary myeloma cells against NVP-AUY922 could be overcome by combination treatment (120). In this manner, the p38 MAPK inhibitor BIRB 796, inhibited protein expression and phosphorylation of Hsp27 induced by Hsp90 inhibitor 17-AAG and enhanced its cytotoxicity (121). Transcription induction of Hsps in response to 17-AAG was also inhibited by the transcription inhibitor actinomycin D. Thus, the combination strategy of 17-AAG and actinomycin D represents a potential approach for the treatment of hematological malignancies (23).

Furthermore, Hsp90 is a promising therapeutic target also in JAK-2 driven cancer, e.g., myeloproliferative neoplasms, B cell acute lymphoblastoid leukemia, including those with genetic resistance to JAK enzymatic inhibitors (122,123). Intriguingly, Hsp90 inhibitor-based antileukemia therapy may override de novo or acquired resistance of acute myelogenous leukemia cells to pan-histone deacetylase inhibitors (124). On the contrary, clinically tolerable doses of tanespimycin (17-AAG) had little effect on resistance-mediating client proteins in relapsed leukemia and exhibits limited activity in combination with cytarabine (Clinicaltrials.gov identifier: NCT00098423) (125).

The Hsp90 inhibitor SNX-7081, which is significantly more potent than 17-AAG synergized and restored sensitivity to fludarabine in seven lymphocytic leukemia cell lines and 23 patient samples, including TP53 mutated cell lines and TP53 or ATM disfunctional patient cells (126). Similar sensitizing effect of Hsp90 inhibitor on chronic lymphocytic leukemia cells was demonstrated by Jones et al (127) as well as Walsby et al (128) who combined GA and NVP-AUY922, respectively, in combination with fludarabine. In this regard, Hsp90 inhibitors represent a potential treatment strategy for fludarabine refractory diseases. A novel strategy to enhance therapeutic response in chronic lymphocytic leukemia may also represent the combination of Hsp90 inhibitor and therapeutic antibody rituximab (129).

Both HL-60 cells transfected with Bcr-Abl and naturally Ph1-positive K562 leukemia cells are resistant to most cytotoxic drugs, but were found to be sensitive to GA. In this regard, GA sensitized Bcr-Abl-expressing cells to doxorubicin and paclitaxel. However, in parental HL-60 cells, GA antagonized the cytotoxic and cell death effects of doxorubicin included inhibition of PARP cleavage and nuclear fragmentation (130). Elimination of mutant BCR-ABL (T315I) kinase seems to be more effective therapeutic strategy for treating BCR-ABL-induced leukemia than the inhibition of BCR-ABL (T315I) kinase activity. Indeed, combination treatment with Hsp90 inhibitor IPI-504 and imatinib was more effective than either treatment alone in prolonging survival of mice simultaneously bearing both wild-type and T315I leukemic cells (131). Simultaneous exposure of BaF3 cells expressing BCR-ABL mutants including T315I to AUY922 and nilotinib was more effective at reducing the outgrowth of resistant cell clones as well as at survival prolongation of mice transplanted with mixture of BaF3 cells expressing wild-type BCR-ABL and mutant forms (132). Similarly, flavopiridol (semisynthetic flavone and selective inhibitor of CDKs) sensitized imatinib-resistant HL/Bcr-Abl cells to GA and it seems that a cocktail of flavopiridol, 17-AAG (or its analogs) and imatinib may be an attractive approach for further investigations (133).

Despite the promising introduction of the proteasome inhibitor bortezomib in the treatment of mantle cell lymphoma (MCL), not all patients respond, and resistance often appears after initial treatment. In bortezomib-resistant cells, cell pretreatment with Hsp90 inhibitor IPI-504, led to synergistic induction of apoptotic cell death when combined with bortezomib (134). For the above, and further studies with a response of multiple myeloma, lymphocytic and myelogenous leukemia cells to combination of Hsp90 inhibitor and another therapy see Table V.

17-AAG and PARP inhibitor olaparib had modest, replication-dependent radiosensitizing effects on T98G glioma cells. Additive radiosensitization was observed through the combination treatment, mirrored by increases in gammaH2AX foci in G(2)-phase cells. Unlike olaparib, 17-AAG did not increase radiation sensitivity of Chinese hamster ovary cells, indicating tumor specificity (135). In addition, 17-AAG sensitized xenografted U87MG cells to cisplatin in nude mice. Indeed, Hsp90-targeted therapy seems to be an effective strategy for potentiating chemotherapy using DNA-crosslinking agents for temozolomide-refractory gliomas (136). Moreover, treatment with subtoxic doses of 17-AAG in combination with tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) induced rapid apoptosis in TRAIL-resistant glioma cells, suggesting an attractive strategy for treating gliomas (137). Interestingly, 17-AAG inhibited the growth of intracranial tumors and synergized with radiation both in tissue culture and in intracranial tumors. On the contrary, 17-AAG did not synergize with temozolomide in any of glioma models (138). The response of glioblastoma cells to combination of Hsp90 inhibitor with other therapy is summarized in Table VI.

17-AAG induced in uveal melanoma cells compared to cutaneous melanoma cells (with low tyrosine kinase activity of ^WTB-Raf) downregulation of ^WTB-Raf protein, followed by Mek/Erk inactivation and a decrease in cyclin D1. In this regard, the combination of 17-DMAG and imatinib mesylate was more efficient, than individual treatments of c-Kit positive uveal melanoma cells (21). The Hsp90 inhibitor XL888 increased Bcl-2 interacting mediator of cell death expression, decreased Mcl-1 expression, and induced apoptosis more effectively than dual mitogen-activated protein-extracellular signal-regulated in most resistant melanoma models. The reversal of the resistance phenotype was associated with the degradation of PDGFRβ, COT, IGFR1, CRAF, ARAF, S6, cyclin D1 and Akt (139). Inhibition of Hsp90 by GA sensitized melanoma cells also to thermosensitive ferromagnetic particle-mediated by hyperthermia with temperature 43°C (140). The Hsp90 inhibitor ganetespib was reported to exhibit robust antitumor efficacy in preclinical model of BRAF^V600E melanoma and could thus readily overcome mechanisms of intrinsic and acquired resistance to selective BRAF inhibitors. The data mentioned above suggest that ganetespib may offer an alternative, and potentially complementary, strategy for therapeutic intervention in mutant BRAF-driven disease (141). The above studies are summarized in Table VII.

Although up to 70% of advanced bladder cancer patients initially show good tumor response to systemic cisplatin-based combination chemotherapy, >90% of good responders relapse and eventually die of the disease. This is attributable to the re-growth of bladder cancer-initiating cells, that have survived chemotherapy. These cancer-initiating cells were more resistant to cisplatin and exhibited more activity in the Akt and Erk oncogenic signaling pathways when compared with their CD44⁻ counterparts. On the other hand, 17-DMAG simultaneously inactivated both Akt and Erk signaling and synergistically potentiated the cytotoxicity of cisplatin against bladder cancer-initiating cells by enhancing cisplatin-induced apoptosis in vitro and in vivo (142). Similarly, low doses of 17-DMAG potentiated antitumor activity of chemoradiotherapy (cisplatin + ionizing radiation) and supported clinical trial of Hsp90 inhibitors to overcome chemoradiotherapy resistance in patients with muscle-invasive bladder cancer (143). The combination of 17-AAG and HIV protease inhibitor ritonavir inhibited renal cancer growth through the inhibition of HSF-1 expression and downregulation of Hsp27, Hsp70 and Hsp90 induced by 17-AAG (144). Recently, Ma et al (145) demonstrated dual targeting of two chaperones Hsp90 and Hsp70 and their combination with conventional anticancer drugs as promising therapeutic option for patients with advanced bladder cancer. The response of bladder and renal cancer cells to combination of Hsp90 inhibitor with another therapy is summarized in Table VIII.

The synthetic Hsp90 inhibitor BIIB021 revealed a strong antitumor effect in head and neck squamous cell carcinoma cells in vitro and in vivo. Its antitumor effect was higher than the effect of ansamycin-based Hsp90 inhibitor 17-AAG and when combined with radiation sensitized the efficacy of radiation therapy (146). The activation of multi-chaperone complex could be a resistance mechanism to the antiproliferative and apoptotic effects induced by tipifarnib (farnesyl transferase inhibitor) and that the combination of tipifarnib with GA is possibly a suitable therapeutic strategy in the treatment of Hsp90-overexpressing head and neck squamous cell cancers (147). Interestingly, the combination of GA with docetaxel was not able to induce any synergistic effects on cell growth inhibition of head and neck squamous cell cancers independently of Hsp90 expression and activation status. On the other hand, the combination of GA with tipifarnib was potentially able to give synergistic effects on the growth arrest of head and neck squamous cell cancers especially when Hsp90 was overexpressed and even more when it was over-activated (147). Sensitizing effect of Hsp90 inhibitor was demonstrated in many other transformed cells, the results of which are shown together with head and neck cancer cells in Table IX.

Recent studies of Lee et al (148) showed that the combination of GA with quercetin or KNK437 sensitized breast cancer stem-like cells characterized by high intracellular aldehyde dehydrogenase activity and higher expression of Hsp90α toward antiproliferation and anti-migration effects of GA. Lee et al (148) also found that knockdown of Hsp27 could mimic the effect of Hsp inhibitors to potentiate the breast cancer stem-like cell targeting effect of GA. Newman et al (149) also supported the use of Hsp90 inhibitor as a cancer stem cell targeting agent and showed, that HSF-1 is an important target for elimination of both cancer and non cancer stem cells in cancer. They found, that low concentrations of the Hsp90 inhibitor 17-AAG eliminate lymphoma cancer stem cells in vitro and in vivo by disrupting the transcriptional function of HIF-1α (149).