UCLA

BACKGROUND: We evaluated the non-cyclic dinucleotide stimulator of interferon genes agonist MK-2118 ± pembrolizumab in patients with advanced solid tumors or lymphomas. METHODS: This first-in-human study (NCT03249792) enrolled patients with refractory, advanced solid tumors or lymphomas. Patients received intratumoral (IT) MK-2118 100-20,000 µg (arm 1), IT MK-2118 900-15,000 µg plus intravenous (IV) pembrolizumab 200 mg every 3 weeks (Q3W; arm 2), or subcutaneous (SC) MK-2118 5000-150,000 µg plus IV pembrolizumab 200 mg Q3W (arm 4); arm 3 (visceral injection of MK-2118) was not pursued. IT dosing used an accelerated titration design and modified toxicity probability interval method; SC dosing (arm 4) was started subsequent to arms 1 and 2. Primary objectives were safety/tolerability. MK-2118 pharmacokinetics were a secondary endpoint; objective responses and biomarkers were exploratory endpoints. RESULTS: 140 patients were enrolled (arm 1, n=27; arm 2, n=57; arm 4, n=56). Grade 3/4 treatment-related adverse events occurred in 22%, 23%, and 11% of patients, respectively, but no maximum tolerated dose was identified up to MK-2118 20,000, 15,000, and 150,000 µg across the 3 arms. Dose-dependent increases in MK-2118 systemic exposure were observed following IT and SC administration. Objective responses were seen in 0%, 6%, and 4% of patients, respectively. IT MK-2118 led to dose-dependent changes in STING-based blood RNA expression levels, interferon-gamma, interferon-gamma‒induced protein 10, and interleukin-6; SC MK-2118 did not generate dose-related immune responses. CONCLUSION: IT MK-2118 ± pembrolizumab and SC MK-2118 plus pembrolizumab had manageable toxicity and limited antitumor activity. IT but not SC administration demonstrated systemic immune effects.