Background
Despite recent advances in our understanding of the biology of multiple myeloma (MM) and the development of novel agents and therapeutic strategies [
1], the prognosis of MM patients remains poor, and resistance to traditional and new drugs frequently occurs [
2]. Alkylators have underpinned the treatment of MM for more than 50 years [
3], but short- and long-term toxicity of these drugs remains a concern. Considerable effort has been made to develop new molecules [
4] and strategies [
5,
6] which could improve the activity and decrease the toxicity of these agents. In this context, EDO-S101 [
7] is a first-in-class compound derived from a molecule of the alkylator bendamustine that has been linked to a histone deacetylase inhibitor (HDACi) radical with potent inhibitory activity of both class I and II HDAC [
7].
Bendamustine [
8] is a bifunctional molecule that combines the alkylating activity of the mustard group with an anti-metabolite purine analog structure. It is approved for newly diagnosed MM patients, based on a randomized phase III study that compared bendamustine and prednisone (BP) with the standard melphalan and prednisone (MP) [
9]. However, bendamustine is mainly used in the relapsed setting, either in monotherapy [
10] or in combination with bortezomib [
5,
6] or immunomodulators [
11,
12].
On the other hand, HDACi have been described to inhibit histone deacetylase proteins (HDACs) and other non-histone proteins [
13]. HDAC proteins are enzymes that remove acetyl groups from an
N-acetyl-lysine amino acid on a histone, allowing the histones to wrap the DNA more tightly. HDAC proteins can also be called lysine deacetylases (KDAC), to describe their function rather than their target, which also includes non-histone proteins [
14]. HDACs are deregulated in many cancers thereby affecting the expression of tumor suppressors and oncogenes [
15]. Two of these HDACi have been tested in phase III trials in combination with bortezomib in relapsed MM patients: vorinostat [
16] and panobinostat [
17]; particularly, panobinostat potentiated the activity of the proteasome inhibitor, leading to the recent approval of panobinostat in combination with bortezomib and dexamethasone for the treatment of relapsed MM patients.
The rationale for using EDO-S101 arises from the hypothesis that histone acetylation induced by the novel radical would result in a more open chromatin structure which would be particularly susceptible to the alkylating effect of bendamustine. Preclinical data have shown the synergy of panobinostat with melphalan [
18] and that of entinostat with bendamustine [
19]. Unfortunately, this effect has not been observed in the clinical setting, partly due to the hematological toxicity of these combinations [
20,
21]. Our hypothesis is that this hybrid molecule may be able to overcome these caveats.
In the present work, we show the potent activity of EDO-S101 in MM cell lines, in MM cells from patients, in a subcutaneous plasmacytoma xenograft model, and in the clinically predictive VK*MYC murine model. This activity was found to be mediated through the potent induction of DNA damage, deacetylase inhibitory activity, and the simultaneous impairment of DNA damage repair.
Discussion
EDO-S101 represents a new fusion principle in which bendamustine and a deacetylase radical inhibitor have been melted into a new molecule in which functional properties cannot be separated. One hypothesis underlying the design of EDO-S101 is that the combined administration of both in the same compound would increase the efficiency of both mechanisms, while decreasing toxicity and providing more convenient administration. In our studies, EDO-S101 was demonstrated to be more potent than bendamustine, retaining and increasing the alkylating activity with an added deacetylase effect. Moreover, EDO-S101 showed efficacy in all MM cell lines tested (median IC
50 = 3.1 μM) and fresh plasma cells from untreated and refractory patients (median IC
50 = 5 μM), independently of the p53 mutational state. It was equally efficient in cell lines resistant to conventional anti-myeloma treatments, such as dexamethasone (MM1R) and melphalan (RPMI-LR5, U266-LR7), indicating that this compound is, at least in vitro, more potent than previous alkylators and could be used to overcome drug resistance. Most importantly, our in vivo data demonstrate that EDO-S101 is also active in in vivo MM models, including the genetically engineered Vk*MYC mouse model, recognized to predict drug response and clinical efficacy in MM. Moreover, it is also remarkable that EDO-S101 is the only drug identified with single-agent activity in the multidrug-resistant Vk12653 transplant model of relapsed/refractory MM (Fig.
5f).
Some previous preclinical studies have shown the synergy of HDACi and alkylators in MM [
19,
42] and other B cell lymphoproliferative disorders [
43]. However, toxicity has been a concern in the clinical settings. For example, the serious hematological toxicity observed with the combination of panobinostat and melphalan in MM precluded the demonstration of clinical activity in this particular clinical trial [
21]. In our ex vivo experiments, we observed a therapeutic window for plasma cells as compared with lymphocytes, which is further confirmed in in vivo studies, in which we showed the clear efficacy of the compound with an acceptable toxicity. This balance was positive, as there was a significant survival benefit in all settings, suggesting that potentially associated secondary effects, such as infections, was not a clear concern for this molecule. However, the safety profile will only be completely defined in the currently ongoing phase I clinical trial in hematological malignancies.
The mechanistic rationale for EDO-S101 was that the HDACi activity would lead to a less compacted structure of the chromatin, making DNA more susceptible to the action of the alkylating molecule. Our in vitro and in vivo studies demonstrated the DNA damage with an increase in H2AX phosphorylation and DNA fragmentation. Attending to the DNA damage pathway and DNA damage response (DDR), EDO-S101 also resulted more active and potent than its precursor, bendamustine, supporting the initial rationale for development.
The pan-HDACi effect of EDO-S101 was also demonstrated by the hyperacetylation of α-tubulin (substrate of the class II deacetylase HDAC6) [
44] and histones 3 and 4 (substrates of the class I deacetylases HDAC1 and HDAC2). Moreover, we have also demonstrated that EDO-S101 inhibits DSB repair by HR, which is consistent with previous reports showing a reduction in HR efficiency after treatment with HDACis. In this regard, it has been recently demonstrated the role of HDAC8 in DSB repair, as it co-localizes with RAD51 at DNA damage sites after irradiation, and HDAC8 inhibition resulted in a decrease in RAD51 promoter activity [
45]. This could explain the defect in the recruitment of RAD51 to DSBs in irradiated cells pretreated with EDO-S101 as has been shown in this work.
One important consideration for any new agent is whether it synergizes with other standards of care in MM. In this regard, EDO-S101 showed synergy with bortezomib in vitro and in vivo. Several studies have previously reported the preclinical and clinical synergy of alkylating agents plus bortezomib [
5,
41] and of bortezomib plus HDACi [
17]. Two main mechanisms may explain, at least partially, this synergy: first, the potentiation of the DNA damage induction evidenced by higher γH2AX levels, and second, the simultaneous inhibition of the proteasome induced by bortezomib and the aggresome inhibition exerted by EDO-S101; last, one event evidenced by the increase in α-tubulin acetylation [
46]. These circumstances would lead to a great accumulation of unfolded and misfolded proteins [
47] which would contribute to cell death.
Conclusions
In summary, our results demonstrate the in vitro, ex vivo, and in vivo anti-myeloma efficacy of EDO-S101 through its HDACi and alkylating activity. The particular mechanism of action of EDO-S101, involving interlaced pathways of potent induction of DNA damage, deacetylase inhibitory activity, and the simultaneous impairment of DNA damage repair, supports the clinical evaluation of this agent in MM patients both in monotherapy and in combination with bortezomib.
Acknowledgements
Not applicable.