Phase 1 study of MRX34, a liposomal miR-34a mimic, in patients with advanced solid tumours

Abstract

Background: In this first-in-human, Phase 1 study of a microRNA-based cancer therapy, the recommended Phase 2 dose (RP2D) of MRX34, a liposomal mimic of microRNA-34a (miR-34a), was determined and evaluated in patients with advanced solid tumours.

Methods: Adults with various solid tumours refractory to standard treatments were enrolled in 3 + 3 dose-escalation cohorts and, following RP2D determination, expansion cohorts. MRX34, with oral dexamethasone premedication, was given intravenously daily for 5 days in 3-week cycles.

Results: Common all-cause adverse events observed in 85 patients enrolled included fever (% all grade/G3: 72/4), chills (53/14), fatigue (51/9), back/neck pain (36/5), nausea (36/1) and dyspnoea (25/4). The RP2D was 70 mg/m² for hepatocellular carcinoma (HCC) and 93 mg/m² for non-HCC cancers. Pharmacodynamic results showed delivery of miR-34a to tumours, and dose-dependent modulation of target gene expression in white blood cells. Three patients had PRs and 16 had SD lasting ≥4 cycles (median, 19 weeks, range, 11-55).

Conclusion: MRX34 treatment with dexamethasone premedication demonstrated a manageable toxicity profile in most patients and some clinical activity. Although the trial was closed early due to serious immune-mediated AEs that resulted in four patient deaths, dose-dependent modulation of relevant target genes provides proof-of-concept for miRNA-based cancer therapy.

Clinical trial registration: NCT01829971.

Fig. 1. MRX34 pharmacodynamics.

a As measured by qRT-PCR, combined relative mRNA expression (%) pre- and post-treatment (5 timepoints) of five miR-34a target oncogenes (BCL2, CTNNB1, DNAJB1, FOXP1, HDAC1) in white blood cells (WBCs) from patients shows dose-dependent downregulation with increasing MRX34 dose from 50 (n = 4) to 70 (n = 16) to 93 (n = 16) to 110 mg/m² (n = 9). Average expression in pre-dose samples was set as 100%. b Next generation sequencing (RNA-Seq) measurement of relative mRNA expression (%) pre- and post-treatment (24 h) for validated miR-34a target genes shows similar results to qRT-PCR. Dose-dependent knockdown is suggested by the increased range of downregulated genes seen with increasing MRX34 dose from 50 (n = 4) to 70 (n = 6) to 93 (n = 9) to 110 mg/m² (n = 11). Validated miR-34a target genes contain miR-34a binding sites in their respective 3′ UTRs for which regulation by miR-34a has been experimentally verified. c Sylamer analysis plot shows a statistically significant enriched miRNA signature for miR-34a for putative target genes, suggesting specific activity of the miR-34a mimic against its target genes in patient-derived WBCs. In this analysis, genes affected by MRX34 at the 24-hr time point relative to baseline in the pre-dose samples for all dosing cohorts were sorted based on RNA-Seq measured expression from most downregulated to most upregulated. The 3′ UTR sequences of these genes were then scanned by the Sylamer algorithm for miRNA binding sites. The x-axis of the plot represents the sorted gene list from downregulated to upregulated. The y-axis represents an enrichment score of the seed sequence binding sites. P-values are based on simulations for seed ‘CTGCCA’. The plot also includes results of Sylamer analyses performed for house-keeping miRNAs, miR-24 and miR-191, which failed to show significant enrichment of downregulated target genes for these miRNAs, further indicating specific gene-directed activity of MRX34.

Fig. 2. Chromogenic in situ hybridisation (CISH) staining of pre- (baseline) and post-MRX34 treatment (treated) liver biopsies from patients with various advanced solid tumours.

Compared to pre-treatment staining, post-treatment results show variably increased miR-34a staining (dark blue/purple) in tumour tissue with localisation to the cellular cytoplasm. In all cases, a lesion in the liver was biopsied for CISH analysis. Uveal melanoma: tumour biopsy was taken 12 days after first dose (93 mg/m²); the biopsy presented spindle cell (bottom) and polygonal-shaped (top) melanoma. SCLC: tumour biopsy was taken 3 days after MRX34 dosing (93 mg/m²). HCC: tumour biopsy was taken 3 days after MRX34 dosing (70 mg/m²). GIST: tumour biopsy was taken 4 days after MRX34 dosing (93 mg/m²).

Fig. 3. Responses in patients treated with MRX34.

a Confirmed PR in a 32-year-old male with N-ras mutated, KIT/BRAF wild, PD-L1⁺ acral melanoma that was initially treated by thumb amputation. Following progression with multiple metastases, the patient was treated unsuccessfully with adoptive T-cell therapy with high-dose IL-2, ipilimumab, pembrolizumab, and CVT chemotherapy. MRX34 treatment was initiated and index lesion size reductions of 39 and 54% were observed after cycles 4 and 6, respectively. MRX34 was discontinued at the patient’s request after completion of cycle 7, after which the PR lasted an additional 7 months with no other treatment for a total duration of response of 65+ weeks. b Confirmed PR in a 56-year-old male with clear cell renal carcinoma who had been refractory to sunitinib, temsirolimus, and bevacizumab. The patient received three cycles of MRX34. Due to rising liver enzymes, treatment was discontinued at that time, and a liver biopsy showed immune hepatitis, but no tumour. PR was noted 3 months after MRX34 discontinuation and confirmed at 4.5 months. The response lasted 54 weeks.