This single-arm, multi-center phase Ib study sought to evaluate the efficacy of ME-344 combined with bevacizumab in patients with refractory mCRC. In total, 23 patients were enrolled rapidly over a period of 4 months, indicating an unmet need for patients with refractory disease. Treatment was found to be safe, without new safety signals. Unfortunately, the clinical activity was limited, achieving a median PFS of 1.9 months, estimated 16-week PFS rate of 30.6%, and a median overall survival of 6.7 months.

Angiogenesis has a crucial role in tumor growth and metastases, and angiogenic inhibiting drugs have become a major therapeutic approach in several common epithelial malignancies such as colorectal, lung, kidney and liver cancers [19, 20]. Two main types of antiangiogenics have been developed, monoclonal antibodies aimed VEGF or VEGF receptors and the oral small molecule multi-tyrosine kinase inhibitors (TKIs) involved in normal cellular functions and oncogenesis [21]. However, single agent activity is modest, and development of resistance is common, with several mechanisms involved [22]. Various strategies have been studied to enhance antiangiogenic therapy, including combination with chemotherapy, targeted therapy, or immune-checkpoint inhibitors, the latter based on the known immunosuppressive effects of VEGF, which can be reversed with VEGF inhibition [23].

Another approach attempts to capitalize on hypoxia correction following antiangiogenic therapy leading to downregulation of glycolysis and switch to long-term reliance on mitochondrial respiration for tumor survival. In vivo, agents like ME-344 can induce synergistic tumor control when combined with antiangiogenics, leading to metabolic synthetic lethality, as shown in combination with TKIs in the PyMT breast cancer and Lewis lung carcinoma models [14]. This effect was subsequently documented in a randomized phase 0/1 window of opportunity study in 42 patients with newly diagnosed HER2-negative breast cancer, where patients were initially treated with a single dose of bevacizumab with pre- and post-dose FDG-PET, followed by randomization to ME-344 + bevacizumab or placebo. Those patients whose tumors had evidence of vascular normalization by FDG-PET (and confirmed by confocal microscopy and HIF1α staining) had a greater decrease in Ki67 [15].

We conducted this study in patients with refractory mCRC given the poor outcomes with standard of care therapy and the established use of antiangiogenics [24]. Because we had no clinical data on ME-344 in combination with bevacizumab in mCRC, this study was designed as having two stages, with the second stage opening to enrollment if at least 20% of patients enrolled in the first stage were progression free at week 16, doubling the rate observed in the phase 3 study of regorafenib in refractory mCRC [3]. While the observed 16-week PFS was 25% in cohort 1, meeting the predefined protocol criteria to open cohort 2, further enrollment to the study was halted by the sponsor to instead prioritize the development of an extended-release formulation which may provide longer exposure to ME-344 and possibly better synergy with antiangiogenics. ME-344 plasma concentrations in this study were generally comparable to exposures previously reported with single agent ME-344, as well as ME-344 in combination with topotecan, with a mean half-life of approximately 6 h [10, 11].

Patients enrolled in the study were heavily pretreated with a median of 4 prior lines of therapy, including bevacizumab in all. In this population, treatment was well tolerated with no drug-related adverse events resulting in treatment discontinuation. The estimated median PFS of 1.9 months and median overall survival of 6.7 months are comparable to that reported with either TAS-102 or regorafenib alone. Of note, patients enrolled in the regorafenib phase 3 study, were all TKI-naïve, whereas 26% of patients enrolled in this study had disease that progressed after TKI therapy [3]. There are no published data with bevacizumab monotherapy in this setting. Thus, it is not possible to determine how the clinical activity might differ from bevacizumab alone. The bevacizumab dose of 5 mg/kg every 14 days was selected based on the prior experience in the breast cancer study and the same as administered with the FOLFOX and FOLFIRI regimens, but lower than 10 mg/kg dose that was evaluated in the phase 3 Study E3200 comparing FOLFOX4 with or without bevacizumab or bevacizumab alone [25]. It is conceivable that a higher bevacizumab dose may have achieved greater synergy with ME-344 leading to a higher response rate. Further, as seen in the phase 0/1 breast cancer study, pre-selection of patients with post-bevacizumab vascular normalization, as judged by FDG-PET, may have enriched for a patient population that benefitted from this strategy.

The untargeted serum metabolomics study revealed considerable variation in baseline levels of metabolites across subjects; however, there was an acute decrease of ≥ 20% in several purine nucleotides and amino acids, including glutamate, serine, ornithine, proline, asparagine, and hypoxanthine. All these metabolites have been associated with proliferation of CRC cell lines or tumors [18]. The 20% cutoff was chosen empirically for this exploratory study, and it is possible that a higher cutoff may have been more clearly associated with better outcomes. The goal of this sub-study was to identify potential metabolic biomarkers of response to ME-344 that could serve as the basis for a targeted panel of metabolites to interrogate in future studies, which could be comprised of a panel of 6 to 18 of the identified metabolites. As noted, the observed metabolites largely returned to near baseline levels prior to dosing on day 15. Thus, while this analysis suggests modulation of metabolic pathways relevant to mitochondria, it also raises concern that exposure may be suboptimal, with a half-life of 6 h and dosing being just weekly.

In conclusion, this study showed that ME-344 in combination with bevacizumab was well tolerated, with no increased toxicity compared to what was reported with each agent alone. Some activity was seen with a degree of disease stabilization in heavily pre-treated patients, not expected with bevacizumab alone. Further investigation can be considered, preferably using an ME-344 extended-release formulation that would increase exposure to study drug and potentially improve clinical benefit.