MK-1084 Alone and Plus Pembrolizumab Shows Early Safety, Activity in KRAS G12C+ Solid Tumors

Carlos Rojas

MK-1084 had an acceptable toxicity profile and elicited responses when administered as a monotherapy in patients with KRAS G12C–mutated previously treated solid malignancies and in combination with pembrolizumab (Keytruda) in those with treatment-naive non–small cell lung cancer (NSCLC), according to updated data from a phase 1 study presented at the 2024 ESMO Targeted Anticancer Therapies Congress.¹

In arm 1, 98% of patients who received MK-1084 monotherapy at all doses (n = 55) experienced at least 1 AE; these effects were grade 3 for 35% of patients, grade 4 for 4% of patients, and grade 5 for 5% of patients. A total of 4 patients discontinued study treatment because of toxicity. Moreover, 60% of patients experienced at least 1 AE that was related to treatment; 7% of these AEs were grade 3 and 2% were grade 4. No dose-limiting toxicities (DLTs) were observed.

In arm 2, all patients who received MK-1084 at all doses (n = 31) plus pembrolizumab experienced 1 or more AEs; these effects were grade 3 (32%), grade 4 (10%), and grade 5 (6%). A total of 7 patients in this arm discontinued any study drug because of toxicity. Additionally, 87% of patients experienced 1 or more treatment-related AEs and these effects were grade 3 (39%), grade 4 (3%), and grade 5 (3%). One patient who received MK-1084 at a daily dose of 400 mg plus the immunotherapy experienced DLTs in the form of grade 3 increased aspartate and alanine aminotransferase (AST; ALT).

Efficacy data indicated that in arm 1, single-agent MK-1084 at all doses (n = 55) elicited an objective response rate (ORR) of 29% (95% CI, 18%-43%), with a median time to response of 1.4 months (range, 1.2-6.2). The disease control rate (DCR) was 82% in these patients. When broken down by dose, those who received a daily dose ranging from 25 mg to 200 mg total (n = 33) experienced an ORR of 30% (95% CI, 16%-49%) and a DCR of 79%. In those who received a daily dose of 400 mg or higher total, the ORR was 27% (95% CI, 11%-50%) and the DCR was 86%.

In arm 2, MK-1084 given at all doses plus pembrolizumab (n = 27) led to an ORR of 70% (95% CI, 50%-86%) and a DCR of 85%. The median time to response was again 1.4 months (range, 1.2-4.2). When broken down by dose, those who received the investigative agent at a total daily dose ranging from 25 mg to 200 mg (n = 23) paired with pembrolizumab experienced an ORR and DCR of 70% (95% CI, 47%-87%) and 87%, respectively. In those who received MK-1084 at a total daily dose of 400 mg or higher plus pembrolizumab (n = 4), the ORR was 75% (95% CI, 19%-99%) and the DCR was (75%).

“Antitumor activity for MK-1084 monotherapy was observed in patients with previously treated solid malignancies whose tumors harbored KRAS G12C mutations; responses occurred in patients with colorectal cancer [CRC], NSCLC, and other tumor types at 50 mg to 800 mg total daily doses for MK-1084,” Dr Carlos Rojas, of Centro de Investigacion Clinica, Bradford Hill, in Santiago, Chile, said in a presentation of the data. “Antitumor activity for MK-1084 in combination with pembrolizumab was observed in patients with previously untreated NSCLC with a PD-L1 tumor proportion score [TPS] of at least 1% across all doses evaluated.”

Although KRAS G12C mutations occur frequently in several cancers, they have the highest incidence in those with NSCLC, according to Rojas, who added that inhibition of this target could encourage a proinflammatory tumor microenvironment, which is conducive to success with immunotherapeutic approaches. “ML-1084 is a selective KRAS G12C–GDP inhibitor that has shown promising antitumor activity in preclinical studies,” he said.

The open-label, global, phase 1 study enrolled patients into two arms. Those in arm 1 had locally advanced solid tumors and had previously received at least 1 therapy, and those in arm 2 had treatment-naive metastatic NSCLC with a PD-L1 TPS of 1% or higher. All patients needed to be at least 18 years of age, have confirmed KRAS G12C–mutated disease that was also measurable by RECIST v1.1 criteria, and an ECOG performance status of 0 or 1.

Those enrolled in arm 1 received single-agent MK-1084 (n = 55) at the following once-daily dose levels: 25 mg (n = 3), 50 mg (n = 5), 100 mg (n = 8), 200 mg (n = 9), 400 mg (n = 9), or 800 mg (n = 4) with optional twice-daily dose levels of 50 mg (n = 8) and 200 mg (n = 9). Those in arm 2 received MK-1084 at once-daily doses of 25 mg (n = 3), 50 mg (n = 8), 100 mg (n = 5), 200 mg (n = 8), or 400 mg (n = 4), or a twice-daily dose of 50 mg (n = 3), in combination with pembrolizumab given at 200 mg every 3 weeks.

The primary end points of the study were to examine DLTs, adverse effects (AEs), and AEs that led to treatment discontinuation. A key secondary end point was ORR.

The median patient age in arms 1 and 2, respectively, was 64.0 years (range, 36-81) and 68.0 years (49-78); 62% vs 52% of patients were male and more than half of patients in both arms had an ECOG performance status of 1 (65.5% vs 74.2%). In arm 1, 22% of patients had NSCLC, 67% had CRC, 2% had anal cancer, 2% had cervical cancer, 2% had melanoma, 2% had mucinous neoplasm of the appendix, 2% had neuroendocrine cancer, and 2% had pancreatic cancer. In arm 1, 24% had prior 1 line of therapy, 33% received 2 prior lines, 16% received 3 prior lines, and 22% had 4 or more lines; this information was unknown for 5% of patients.

The median follow-up time in arm 1 was 11.1 months (range, 2.1-21.5) and the median exposure duration was 5.7 months (range, 0.2-15.2). At the time of the data cutoff date of November 3, 2023, 38% of patients were still receiving treatment and 62% had discontinued. The most common reason for treatment discontinuation was disease progression (55%) followed by AEs (7%). For arm 2, the median follow-up time was 7.0 months (range, 0.4-17.2) and the median exposure duration was 6.0 months (range, 0.4-15.9). At data cutoff, 84% of patients were still receiving treatment and 16% had discontinued due to disease progression (10%) or AEs (6%).

Additional data showed that within arm 1, those with CRC (n = 37) experienced an ORR of 24% (95% CI, 12%-41%) with a DCR of 78% and a median time to response of 2.6 months (range, 1.2-6.2). Those with NSCLC (n = 12) experienced an ORR of 42% (95% CI, 15%-72%), a DCR of 83%, and a median time to response of 1.4 months (range, 1.2-1.4). Those with other solid tumors (n = 6) experienced an ORR and DCR of 33% (95% CI, 4%-78%) and 100%, respectively; the median time to response in this group was also 1.4 months (range, 1.3-1.5).

In arm 2, those with a PD-L1 TPS ranging from 1% to 49% (n = 13) achieved an ORR of 54% (95% CI, 25%-81%) and a DCR of 77%; the median time to response in this group was 1.4 months (range, 1.2-4.2). In those with a PD-L1 TPS of 50% or higher (n = 14), the ORR and DCR experienced with the doublet was 86% (95% CI, 57%-98%) and 93%, respectively. Again, the median time to response in this group was 1.4 months (range, 1.2-2.6).

Duration of response data were not mature at the time of the analysis, according to Rojas.

In arm 1, the treatment-related AEs (TRAEs) that had an incidence of 10% or higher in either arm included increased AST (grade 1/2, 15%; grade 3/4, 4%), increased ALT (13%; 4%), nausea (16%; 0%), diarrhea (11%; 2%), fatigue (13%; 0%), pruritus (7%; 0%), rash (4%; 0%), and increased alkaline phosphatase (ALP; 4%). In arm 2, the TRAEs that had an incidence of at least 10% in either arm comprised increased ALT (grade 1/2, 26%; grade 3-5, 13%), increased AST (23%; 10%), pruritus (23%; 3%), diarrhea (23%; 0%), rash (10%; 3%), fatigue (13%; 0%), increased ALP (13%; 0%), and nausea (10%; 0%).

“The study has been expanded to additional combinations evaluating MK-1084 in patients with KRAS G12C­–mutated cancer,” Rojas concluded.

Reference

Rojas CI, Lugowska I, Juergens R, et al. 44O Updated results from a phase I study evaluating the KRAS G12C inhibitor MK-1084 in solid tumors and in combination with pembrolizumab in NSCLC. ESMO Open. 2024;9(suppl 102273). doi:10.1016/j.esmoop.2024.102273