Drain the lysosome: Development of the novel orally available autophagy inhibitor ROC-325

ABSTRACT

Although macroautophagy/autophagy is a key contributor to malignant pathogenesis and therapeutic resistance, there are few FDA-approved agents that significantly affect this pathway. We used medicinal chemistry strategies to develop ROC-325, an orally available novel inhibitor of lysosomal-mediated autophagy. Detailed in vitro and in vivo studies in preclinical models of renal cell carcinoma demonstrated that ROC-325 triggered the hallmark features of lysosomal autophagy inhibition, was very well tolerated, and exhibited significant superiority with respect to autophagy inhibition and anticancer activity over hydroxychloroquine. Our findings support the clinical investigation of the safety and preliminary efficacy of ROC-325 in patients with autophagy-dependent malignancies and other disorders where aberrant autophagy contributes to disease pathogenesis.

Although there is overwhelming preclinical evidence demonstrating that autophagy functions as a mechanism to maintain bioenergetic homeostasis under stress conditions including those imposed by hypoxia, nutrient deprivation, and radio- and pharmacological therapy in a diverse range of solid and hematological cancers, the chloroquines (chloroquine [CQ] and hydroxychloroquine [HCQ]) remain the only FDA-approved drugs that inhibit autophagy that have been evaluated in a significant number of cancer clinical trials. While adding HCQ to other anticancer regimens was generally shown to be safe in initial clinical studies, preliminary efficacy was reported to be modest regardless of the specific combination. The underlying basis for the disconnect between the promising efficacy observed in earlier preclinical studies of these regimens and the somewhat disappointing results in the context of early phase human trials is not entirely clear. One distinct possibility is that the maximum tolerated dose of HCQ in these settings may not have resulted in complete inhibition of autophagic activity. This was suggested by pharmacodynamic assays conducted in parallel with these clinical trials. However, the small number of patients in each cohort prohibited testing this hypothesis in a statistically significant manner. Nevertheless, these initial clinical findings highlighted the need to develop better agents to antagonize autophagic proteolysis to rigorously investigate the potential benefit of autophagy inhibition as an anticancer strategy.

We used medicinal chemistry approaches to generate a series of new dimeric compounds containing modified HCQ and lucanthone (Miracil D) scaffolds with the goal of generating new drugs with better autophagy-directed potency, tolerability, and anticancer efficacy than HCQ. Our efforts led to the development of ROC-325, a novel water soluble compound containing HCQ and lucanthone structural motifs (Fig. 1). Preclinical studies showed that ROC-325 exhibits superior lysosomal autophagic inhibition and approximately 10-fold greater single-agent anticancer activity against a broad range of tumor types compared with HCQ. We chose to focus intensive studies of the mechanism of action and efficacy of ROC-325 in models of renal cell carcinoma (RCC) for 3 primary reasons. First, the high sensitivity of A498 cells in our preliminary compound screens. Second, the clinical efficacy we observed in a patient with RCC treated with a combination of HCQ and vorinostat after failing 7 lines of prior therapy in our recently conducted investigator-initiated clinical trial. Finally, a report published in this journal (Lebovitz et al., 2015) suggesting that alterations in the autophagy interactome in patients with renal cancer may negatively affect overall survival.

Figure 1.

Chemical structure of ROC-325.

In addition to demonstrating that treatment with ROC-325 promoted lysosomal deacidification, the accumulation of autophagosomes, and disruption of autophagic flux, our study also showed that genetic impairment of the critical autophagy genes ATG5 and ATG7 blunted its anticancer efficacy. This indicates that autophagic inhibition is a major component of ROC-325s mechanism of action. In vivo investigation of the tolerability, efficacy, and pharmacodynamics of ROC-325 showed that oral administration to mice bearing RCC xenografts was very well tolerated and exerted single-agent anticancer activity that was superior to a higher dose of HCQ. Immunohistochemical analyses of tumor specimens from mice treated with ROC-325 and HCQ affirmed its superior efficacy as oral delivery of ROC-325 leads to significantly better in vivo autophagy disruption, improves inhibition of tumor cell proliferation, and elevates levels of tumor cell apoptosis.

Our collective findings support further investigation of ROC-325 as a novel agent for autophagy-dependent malignancies and other diseases where aberrant autophagy is a driver of disease pathogenesis. However, several important questions remain open with respect to optimally positioning autophagy inhibitors as targeted anticancer agents. A key challenge for all investigators in the field is identifying specific tumor types that are more dependent upon autophagy or therapeutic regimens where its inhibition could achieve synthetic lethality. Significant efforts have been applied to addressing both of these issues, but there is still much more work to be done to achieve more definitive clarity.

An equally important question that still remains relatively understudied regards where precisely to target the pathway for maximal therapeutic benefit. It is well known that CQ/HCQ disrupt the pathway at the most distal point—the lysosome. Given that no true targeted inhibitors of upstream elements of the autophagy cascade have been clinically tested, one cannot definitely conclude whether proximal pathway disruption would be equivalent to or superior to distal (lysosomal) perturbation. Indeed, several pharmaceutical companies have active pipeline programs focused on developing specific antagonists of upstream autophagy regulators including ATG7 and the ULK kinases.

After careful consideration of potential pharmacological approaches to impair autophagy, we chose to focus on creating novel compounds that trigger lysosomal disruption rather than attempting to engineer small molecule inhibitors of potential upstream regulators. We hypothesized that the individual targeting of proximal pathway regulators would likely be less therapeutically beneficial than hitting the primary point of pathway convergence (the lysosome) based on the complexities of the autophagic cascade itself. There are many different combinations of interconnecting factors that can be sequentially activated to stimulate autophagy. The reality is that there is not a simple linear series of regulators that are universally involved in autophagy regulation initiated by all stimuli. The context-dependent variability in contributing autophagy regulators combined with the possibility for other factors to compensate for the antagonism of individual upstream components is an important issue to consider when targeting this pathway. We think these factors are likely to result in the maintainence of some level of pathway activity when individual upstream regulators are impaired. The lack of a complete understanding of the functional roles of all upstream regulators and how they might change in different genetic or biochemical scenarios adds another layer of complexity to the investigation of this approach. Given that all autophagy roads lead to the lysosome, we propose that antagonizing lysosomal integrity and/or function will yield greater therapeutic benefit than the targeted inhibition of any single autophagy-regulating gene. Clinical trials that test the safety and efficacy of second-generation lysosomal antagonists like ROC-325, and specific inhibitors of genes that regulate autophagy will ultimately demonstrate which strategy is superior. Until then, we will continue to focus our research efforts on the optimization of pharmacological approaches to “drain the lysosome.”

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by grants from the Scott Hamilton CARES Initiative (JSC) and the National Cancer Institute (R01CA172443, JSC; R01CA190789, STN; and P30CA023074)