During a Targeted Oncology™ Case-Based Roundtable™ event, Devika Das, MD, discussed the role of KRAS inhibitors and the outcome of the KRYSTAL-1 trial in patients with non–small cell lung cancer. This is the first of 2 articles based on this event.
Targeted Oncology: What do the National Comprehensive Cancer Network (NCCN) guidelines recommend for patients with non–small cell lung cancer (NSCLC) and a KRAS G12C mutation?
DAS: Per NCCN guidelines…in patients with the KRAS G12C mutations who have adenocarcinoma or squamous cell carcinoma in the second line—so they have progressed on frontline therapy—we have 2 drugs currently approved: sotorasib [Lumakras] and adagrasib [Krazati].1 One was approved in 2021, the other in 2022. Both of the approvals are for second line and [later] in patients in this scenario.
Why were these drugs studied in the second line and later, whereas other targeted inhibitors are moving to the frontline setting?
A lot of them started in second and third line because it takes a lot of justification to bring any drug to frontline. With KRAS G12C inhibitors, it's because we've had so many treatments in preclinical studies that have not worked. There was more reluctance in bringing it up front, knowing that all the drugs historically tested in trials were not effective. That’s one of the reasons. We'll go through some of the data where even in second and third line or the subsequent lines, the data are not [strong enough]. It's not the FLAURA trial [NCT02296125] of osimertinib [Tagrisso]. That's not the level of data we're seeing where [want to] take this to the front line. There are ongoing trials trying to do that, but even the data in subsequent lines are not as impressive as we want it to be. They're great drugs, but not to that level where I'm ready to jump to front line.
What is the significance of the KRYSTAL-1 study (NCT03785249) in this setting?
KRYSTAL-1 is a phase 2 study with adagrasib. It's a registrational phase 2 cohort. It was used as a monotherapy in patients with metastatic disease who had prior therapy with a PD-L1 [inhibitor] plus or minus chemotherapy, in sequence or together. They permitted inclusion of patients with stable CNS [central nervous system] metastases. Patients received 600 mg twice a day [in the form of] a capsule, and patients were fasting. The primary end point was overall response rate [ORR], and the secondary end points were duration of response [DOR], progression-free survival [PFS], and overall survival [OS].
Patients were predominantly white [83.6%], had ECOG performance status of 0 or 1, and were mostly former or current smokers.2 Interestingly, the histology they included in this trial is predominantly adenocarcinoma [with] a small percent of squamous cell carcinoma [2.6%]. Most of the patients had metastatic disease [vs 11.2% with locally advanced disease]. For some of the patients, this was their [fifth] line of therapy. In [12.1%] of patients on the study [adagrasib was the fifth or later] line of therapy. It’s unusual for a lot of clinical trials to have that. They included a [mix] of patients with bone, CNS, adrenal, and liver metastases.
What were the efficacy outcomes of this trial?
The ORR with the single agent in this heavily pretreated patient population was 42.9%. One patient had a complete response, but most patients had partial responses [42.0%] and stable disease [36.6%], which I think is decent if somebody is [treated in the] third line or later. There was disease control in 79.5% of patients, the median PFS was 6.5 months, and the median OS was 12.6 months at the time of data cutoff [median follow-up: 12.9 months for PFS; 15.6 months for OS].
The KRAS G12C mutation is associated [with CNS metastases]. Anywhere between 27% to 40% of patients have CNS metastases at presentation or at some point during the disease course.3 I think that's a good way to test for efficacy of these drugs, and that's often where a lot of our patients fail. In preclinical models, adagrasib showed a clinically meaningful penetration into the cerebrospinal fluid with a brain-to-plasma concentration of 0.47, which is comparable with a lot of the other TKIs that have approval in patients with CNS metastases, so it is a good drug with CNS penetration.3
They looked at a post hoc evaluation of the best intracranial response in these patients, and 33.3% of patients had confirmed intracranial response.2 The median duration of intracranial response was 11.2 months, and the median intracranial PFS was 5.4 months in  patients who had CNS metastases at baseline. These are data signaling this could be a good drug for patients who have CNS metastases.
As an aside, the group looked at a cohort of about 25 patients who had untreated CNS metastases and put them on adagrasib and had very similar good responses.4 These were patients with metastases that were not treated with radiation. They excluded patients with cerebellar or brainstem metastases or leptomeningeal disease, but most of the brain metastases were included and saw some good responses.
What toxicities were seen with adagrasib in this trial?
We [sometimes think that if] it's an oral medicine, it's relatively safer. That's not always the case. These drugs come with a significant amount of toxicity that you expect with tyrosine kinase inhibitors [TKIs]. For most patients, it was gastrointestinal toxicities, where approximately 70% of patients had some form of diarrhea.2 Fatigue was a [common] adverse event, and…I want to highlight the liver toxicity. Hepatotoxicity was seen at any grade in 37% of patients but grade 3 and 4 [occurred in] 10%, which is a significantly high amount of hepatotoxicity and something that we should be aware of when giving these patients these drugs.
How do you weigh whether to use a KRAS G12C inhibitor in a patient with brain metastases vs radiotherapy or other approaches?
I extrapolate a lot of this from how I treat patients with EGFR mutations. As of now, the data that we're seeing is off of a phase 2 study, and the 25 patients whom I talked about are a subset of those patients. I would not change current practice based on that. The way these data help me is, if somebody has CNS metastases, I would…send them for radiation therapy referral or neurosurgery referral based on what's appropriate to treat that CNS metastasis. But then if I have to pick a drug to give that patient, would I prefer docetaxel vs an oral TKI, one of these 2 drugs [sotorasib and adagrasib]. I would prefer the oral TKI because I know docetaxel does not have good CNS penetration, but these drugs do, and there could be some benefit in that patient population. It is something that I would use after I treat the CNS metastasis. There are a lot of patients who might have micrometastatic disease in the CNS that's not always visible on the MRI. If this is a patient where I am suspecting that highly, then I would be more inclined to pick the TKI if they can take it.
1. NCCN. Clinical Practice Guidelines in Oncology. Non–small cell lung cancer, version 5.2023. Accessed November 27, 2023. https://tinyurl.com/ydsfxfaz
2. Jänne PA, Riely GJ, Gadgeel SM, et al. Adagrasib in non-small-cell lung cancer harboring a KRASG12C mutation. N Engl J Med. 2022;387(2):120-131. doi:10.1056/NEJMoa2204619
3. Sabari JK, Velcheti V, Shimizu K, et al. Activity of adagrasib (MRTX849) in brain metastases: preclinical models and clinical data from patients with krasg12c-mutant non-small cell lung cancer. Clin Cancer Res. 2022;28(15):3318-3328. doi:10.1158/1078-0432.CCR-22-0383
4. Negrao MV, Spira AI, Heist RS, et al. Intracranial efficacy of adagrasib in patients from the KRYSTAL-1 trial with KRASG12C-mutated non–small-cell lung cancer who have untreated CNS metastases. J Clin Oncol. 2023;41(28):4472-4477. doi:10.1200/JCO.23.00046