Adagrasib Delivers Durable CNS-Specific Activity in KRAS G12C+ NSCLC

Joshua K. Sabari, MD

Adagrasib (MRTX849) led to an intracranial objective response rate (ORR) of 32% and a median intracranial duration of response (DOR) that was not reached in patients with KRAS G12C–mutant non–small cell lung cancer (NSCLC) and active, untreated central nervous system (CNS) metastases, according to findings from the phase 1b portion of the phase 1/2 KRYSTAL-1 trial (NCT03785249) that were presented at the 2022 ASCO Annual Meeting.

Additionally, the median overall survival (OS) was not reached.

“CNS metastases from KRASG12C-mutant NSCLC are common and are associated with a poor prognosis. Adagrasib demonstrated encouraging and durable CNS-specific activity in patients with KRASG12C-mutant NSCLC and active untreated CNS metastases,” Joshua K. Sabari, MD, from Perlmutter Cancer Center, New York University Langone Health, said in his presentation.

For patients with this type of mutated NSCLC, about 27% to 42% will have CNS metastases at diagnosis, with a 5-month median OS for patients with active and untreated CNS involvement. Targeted therapies that penetrate the CNS improve outcomes for patients with these types of complications. Adagrasib has previously shown clinical activity in treated, stable CNS metastases, and tumor regression in animal models.

Patients with advanced solid tumors and a KRASG12C mutation were given adagrasib as monotherapy or combined with pembrolizumab (Keytruda), cetuximab (Erbitux), or afatinib (Gilotrif) on KRYSTAL-1, a phase 1/2 multiple expansion cohort trial. The data that were presented look at the phase 1b cohort of patients with NSCLC and active, untreated CNS metastases treated with 600 mg of oral adagrasib twice daily. Safety and clinical activity were evaluated for this portion of the trial, including systemic and intracranial ORR by blinded independent central review (BICR), OS, DOR, and progression-free survival (PFS).

There were 25 patients enrolled, although not all were evaluable for efficacy of adagrasib. The median age in this cohort was 66 years (range, 47-89), and 52% of patients were female. A majority of patients were white (84%), had ECOG performance status of 1 (72%), and were current or former smokers (96%). About half of patients had received 1 prior line of therapy (20%), and the rest had received 0 (12%), 2 (20%), or 3 or more (16%) lines of therapy. Of the patients who were evaluable for efficacy (n = 19), 15 had target CNS lesions and 4 had non-target lesions.

Among the overall evaluable population (n = 19), the ORR was 32%; for patients with only non-target lesions (n = 4) and target lesions with or without non-target lesions (n = 15), it was 50% and 27%, respectively. The complete response rate was 16% in the overall group, 50% in the non-target lesion group, and 7% in the target lesion group. Partial responses were also seen in 16% overall and 20% in patients with target lesions, but not in any patients with only non-target lesions.

Fifty-three percent of patients in both the overall and target lesion groups had stable disease (SD), while 50% had SD in the non-target lesion group. Investigators observed progressive disease in 11% of patients overall, 13% with target lesions, and none with non-target lesions. The disease control rate (DCR) was 84%, 80%, and 100%, for patients in the overall, target lesion, and non-target lesion groups.

“[In] the waterfall plot for BICR, intracranial disease control was broad, with intracranial tumor shrinkage of any magnitude occurring in 80% of radiographically evaluable patients with target lesions. One additional patient demonstrated 100% regression in the target lesions, however, [the patient] was non-evaluable due to the scans being performed too early,” Sabari noted.

There were objective intracranial responses observed in 32% of all patients (95% CI, 12.6%-56.6%). The intracranial DCR was 84% (95% CI, 60.4%-96.6%). Fourteen of the 16 patients who responded had concordance between systemic and intracranial disease control. Additionally, the systemic ORR by RECIST v1.1 was 37% (95% CI, 16.3%-61.6%) and the systemic DCR was 79% (95% CI, 54.4%-93.9%).

At median follow-up of 6.6 months, the median OS had not been reached. The median intracranial PFS was 4.2 months (95% CI, 3.8-not estimable [NE]). The median intracranial DOR had also not been reached (95% CI, 4.1-NE). Of the responders, 83% were still on therapy at the data cutoff of December 31, 2021.

An average K_{p, uu} of 0.47 was seen in 2 patients who had cerebrospinal fluid collected, which exceeds the values for patients who have received tyrosine kinase inhibitors that have shown CNS penetration and antitumor activity in CNS metastases. “Of note, both patients had evidence of CNS regression in their lesions via imaging compared to baseline. [One patient also] demonstrated a decrease in the edema compared to baseline,” Sabari said.

According to the safety profile, grade 1 or 2 treatment-related adverse events (TRAEs) occurred in 60% of the 25 patients who were evaluable for toxicity. There were no grade 4 or 5 TRAEs observed with adagrasib monotherapy. The most frequent grade 3 TRAEs were vomiting in 12%, and diarrhea, alanine transferase, and dizziness in 8% of patients each. Dose reductions/interruptions due to TRAEs were needed for 48% of patients, and discontinuation was needed in 1 patient (4%). CNS-specific TRAEs occurred in 6 patients, 5 reporting dizziness and 1 reporting grade 1/2 aphasia and insomnia.

“Adagrasib is the first KRASG12C inhibitor to demonstrate clinical activity in patients with KRASG12C-mutated NSCLC with both treated as well as active, untreated CNS metastases,” Sabari stated. “An expanded-access program is open and enrolling patients with KRASG12C-mutant solid tumors, including patients with active and untreated CNS metastases.”

Reference

Sabari JK, Spira AI, Heist RS, et al. Activity of adagrasib (MRTX849) in patients with KRASG12C-mutated NSCLC and active, untreated CNS metastases in the KRYSTAL-1 trial. J Clin Oncol. 2022;40(suppl 17):LBA9009. doi:10.1200/JCO.2022.40.17_suppl.LBA9009