Shabason et al have written a thoughtful review of an exciting new class of agents, histone deacetylase (HDAC) inhibitors. While the authors focus primarily on the role of HDAC inhibitors in combination with radiation therapy, we would like to highlight some potential strategies combining these agents with systemic therapies for the treatment of cancer.
Although histones appear to be the most abundant substrate for HDACs, nonhistone proteins such as transcription factors and growth factor receptors are also regulated by these enzymes.[1] In fact, it may well be that the direct actions of HDAC inhibitors on these nonhistone targets hold the key to their success as anticancer agents. Examples of nonhistone proteins whose expression may be altered by HDAC inhibitors include heat shock protein (hsp)90, estrogen receptor (ER)-α, androgen receptor (AR), HER2, p53, and vascular endothelial growth factor (VEGF).
Hsp90 and CML
Hsp90 is a chaperone protein whose function is to stabilize client proteins such as Bcr-Abl, HER2, AKT, and c-raf. Acetylation of hsp90 via inhibition of HDAC6 interferes with this function, thereby leading to the degradation of these oncogenic client proteins.[2] Exploiting this activity against hsp90 has exciting implications for the treatment of a number of different solid and hematologic malignancies. For example, in the case of chronic myelogenous leukemia (CML), mutations within the kinase region of Bcr-Abl confer resistance to treatment with the available tyrosine kinase inhibitors (TKI) such as imatinib(Drug information on imatinib) (Gleevec), dasatinib(Drug information on dasatinib) (Sprycel), and nilotinib(Drug information on nilotinib) (Tasigna). Of particular concern is the T315I mutation, which causes steric hindrance and insensitivity to the standard therapeutic agents. In vitro studies have demonstrated that treatment with HDAC inhibitors in combination with a TKI leads to enhanced apoptosis and depletion of Bcr-Abl levels in imatinib-resistant CML cells, including those harboring the T315I mutation.[3] Based on these data, a phase I study is currently ongoing to evaluate the combination of vorinostat (Zolinza) plus dasatinib in patients with CML in accelerated or blastic phase (ClinicalTrials.gov Identifier: NCT00816283).
HER2 and ER in Breast Cancer
Breast cancer that overexpresses HER2 has been associated with more aggressive tumor biology, altered responsiveness to therapy, and poor clinical outcome including shortened survival.[4] While the development of trastuzumab(Drug information on trastuzumab) (Herceptin), a monoclonal antibody against the HER2 receptor, represents a major advance in the treatment of HER2-positive breast cancer, de novo or acquired resistance to trastuzumab remains a significant problem. Administration of OSU-HDAC42, a novel HDAC inhibitor, leads to acetylation of hsp90 and consequent downregulation of HER2 levels in preclinical models.[5] In addition, treatment of HER2-overexpressing breast cancer cell lines with the HDAC inhibitor LAQ824 leads to enhancement of trastuzumab-induced apoptosis,[6] thus providing a rationale for combining HDAC inhibition with trastuzumab. Because resistance to trastuzumab is often mediated by HER2-independent upregulation of AKT, it is also possible that co-treatment with HDAC inhibitors may delay or even reverse resistance via attenuation of AKT, another client protein of hsp90. A phase I/II trial is currently evaluating vorinostat in combination with trastuzumab for the treatment of HER2-positive metastatic breast cancer (ClinicalTrials.gov Identifier: NCT00258349).
The modulation of ER-α is of particular interest with respect to the treatment of breast cancer. In vitro studies have demonstrated that HDAC inhibitors confer sensitivity to tamoxifen(Drug information on tamoxifen) in both ER-negative and ER-positive cell lines. The mechanisms by which this phenomenon occurs have yet to be fully elucidated, but it has been shown that in ER-positive cell lines, the addition of HDAC inhibitors leads to downregulation of ER-α expression and restoration of antiestrogen sensitivity.[7,8] Conversely, in ER-negative cell lines, treatment with HDAC inhibitors restores ER-α expression and sensitizes cells to tamoxifen.[9,10] These differential actions on ER by this class of agents suggest that there are likely other mechanisms involved to explain the synergy between HDAC inhibitors and antiestrogen therapy, including modulation of ER-β and/or progesterone(Drug information on progesterone) receptor.[8,11]
Munster et al are conducting a phase II study to evaluate the role of vorinostat in restoring tamoxifen sensitivity in patients with ER-positive metastatic breast cancer treated with prior aromatase inhibitor therapy.[12] At the time of preliminary data presentation, 24 patients were evaluable for response; 17 had received prior adjuvant tamoxifen. The response rate was 21% and stable disease rate was 33%. It should be noted, however, that several of these patients might have responded to tamoxifen alone, as it is well known that simply switching classes of antiestrogen agents can be an effective strategy of inducing clinical response. Nevertheless, combining an HDAC inhibitor with endocrine therapy seems to be a promising strategy that warrants further investigation.
AR and Prostate Cancer
Inhibiting AR signaling is an important therapeutic objective in the treatment of prostate cancer, including castrate-resistant prostate cancer.[13] Welsbie et al demonstrated that the HDAC inhibitors decreased levels of the AR protein by suppressing transcription of AR (as opposed to increasing degradation of AR by inhibition of Hsp90).[14] However, inhibition of AR signaling did not solely depend on lowering of AR mRNA and protein. Rather, it appeared that vorinostat and panobinostat (Faridak) inhibited transcription of genes under the control of the AR, including PSA and TMPRSS2, by blocking assembly of the transcription complex after the AR binds to the enhancers of target genes. In this study, HDAC inhibition of AR signaling was dose-dependent and relatively high drug concentrations were required to achieve inhibition.
As Welsbie et al point out, two recent trials of HDAC inhibitors as single agents have failed to demonstrate any prostate-specific antigen (PSA) responses, perhaps due to insufficient plasma drug levels. Bradley et al treated 27 patients with 400 mg of vorinostat orally daily. While no PSA decline ≥ 50% was observed, patients experienced significant adverse events including fatigue (81%), nausea (74%), and anorexia (59%), suggesting that higher doses may need to be administered on an intermittent schedule for the sake of tolerability.[15] Similar results were reported by Molfie et al, who treated 35 patients with weekly romidepsin (Istodax).[16] While two patients achieved a radiographic and PSA partial response lasting ≥ 6 months, overall single-agent romidepsin showed insufficient activity. In additional to alternate dosing schedules, future studies of HDAC inhibitors in prostate cancer may employ drug combination strategies.
Marrocco et al reported minimal cell death when a low dose of vorinostat was used to treat androgen-responsive prostate cancer (LNCaP) cells.[17] However, the combination of subtherapeutic, low-dose vorinostat and the AR antagonist bicalutamide(Drug information on bicalutamide) was synergistic, leading to significant apoptotic cell death. Recently, Liu et al demonstrated that low concentrations of panobinostat increased concentrations of a transcriptional repressor protein leading to a decline in AR levels.[17] Furthermore, treatment of bicalutamide-resistant cells with standard doses of panobinostat and bicalutamide synergistically inhibited cell growth. This combination is currently being tested in an ongoing phase I/II clinical trial (ClinicalTrials.gov Identifier: NCT00878436).
Synergistic Combinations
As Shabason et al point out in their review, HDAC inhibitors as monotherapy appear to have minimal clinical activity, and their biggest role will likely be in combination with other therapeutic agents. Several studies have demonstrated synergy when HDAC inhibitors are combined with various cytotoxic drugs including anthracyclines, taxanes, topoisomerase inhibitors, and other DNA-damaging agents.[18] For example, in vitro data performed on non–small-cell lung cancer (NSCLC) cells has shown enhanced DNA damage when vorinostat is combined with carboplatin(Drug information on carboplatin) and increased microtubule stability when combined with paclitaxel(Drug information on paclitaxel).[19] On the basis of these findings, Ramalingam et al conducted a randomized phase II study of carboplatin plus paclitaxel with or without vorinostat in advanced-stage NSCLC.[20] The response rate was 34% with vorinostat vs 12.5% with placebo (P = .02), and there was a trend toward improved progression-free survival (6.0 vs 4.1 months; P = .48) and overall survival (13.0 vs 9.7 months; P = .17) in favor of vorinostat. Grade 4 thrombocytopenia was more frequent with vorinostat (18% vs 3%; P < .05). Nausea, vomiting, fatigue, dehydration, and hyponatremia were also more common with the addition of vorinostat.
In vitro and preclinical models indicate that HDAC inhibitors also show synergistic activity when combined with agents from a variety of other classes, including antiangiogenic drugs, DNA hypomethylating agents, and proteasome inhibitors.[18] Studies are ongoing to further investigate the optimal manner in which to combine these therapies.
Finding Relevant Targets
As discussed by Shabason et al, if we are to design rational therapies, we need to demonstrate that the intended target is in fact being modulated by the drug under study and that this modulation corresponds with clinical outcomes. Because of convenience, peripheral blood mononuclear cells (PBMCs) are often tested for levels of histone 3 (H3) and histone 4 (H4) acetylation after treatment with HDAC inhibitors. However, while most patients demonstrate increased acetylation of H3 and H4 in their PBMCs after exposure to HDAC inhibitors, no consistent correlation has been observed between levels of acetylation and clinical activity.[21,22] This lack of association may be a result of the fact that PBMCs are not the ideal surrogate for tumor tissue; alternatively, perhaps we have yet to discover the most relevant targets of this class of agents. Fotheringham et al have identified HR23B, a gene responsible for shuttling ubiquinated cargo proteins to the proteasome, as a potential biomarker. The level of HR23B appears to influence the response of tumor cells to HDAC inhibitors, suggesting that these agents also influence proteasome activity.[23] Other investigators are evaluating gene signatures that may identify tumors most likely to respond to HDAC inhibitors.[24,25]
Conclusions
The study of HDAC inhibitors is still in its infancy. While this class of agents holds great promise as anticancer therapy, we have yet to learn how best to administer these drugs. In this era of personalized medicine, we strive to individualize therapy so that maximal benefit may be achieved and unnecessary toxicity minimized. Our ability to do so depends on furthering our understanding of the various mechanisms by which HDAC inhibitors exert their effects, elucidating the optimal sequence and schedule of administration, and identifying individuals who are most likely to benefit from this particular therapy.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
