Skip to main content
Premium Trial:

Request an Annual Quote

ESMO Data Deepens Understanding of KRAS Inhibitor Activity in Lung, Colorectal Cancer


PARIS – Researchers at the European Society for Medical Oncology Congress on Monday reported the first head-to-head data pitting Lumakras (sotorasib) against chemotherapy in KRAS G12C-mutated non-small cell lung cancer, a setting in which Amgen has accelerated approval for the KRAS inhibitor from the US Food and Drug Administration since May 2021.

That approval in the US and then in Europe made Lumakras the first KRAS inhibitor to reach the market for KRAS G12C-mutated metastatic NSCLC. Mirati is hot on Amgen's heels with adagrasib, and earlier this year, the FDA accepted the firm's new drug application for its KRAS inhibitor in the same setting and is expected to make a decision by Dec. 14.

But both Amgen's and Mirati's ambitions for their drugs go beyond NSCLC. At ESMO, researchers also presented data on the activity of Lumakras and adagrasib as single agents and combination therapies in KRAS G12C-mutated metastatic colorectal cancer, a patient population with particularly poor outcomes and sorely in need of new treatments.

Lumakras vs chemo

The Phase III CodeBreaK-200 trial produced the first data showing that a KRAS inhibitor had greater benefit than docetaxel chemo in patients with KRAS G12C-mutant metastatic NSCLC. Amgen announced last week that it had "productive discussions with regulators about [the CodeBreaK-200] trial design." The firm could use the data to convert Lumakras' accelerated approval to a full approval in the US as a second-line treatment for KRAS G12C-mutated advanced lung cancer.

The study enrolled 345 patients; 169 patients were evaluable on the Lumakras arm, and 151 patients were evaluable on the chemo arm. Twice as many patients responded to Lumakras (28.1 percent) than they did to chemo (13.2 percent). Eight of 10 patients treated with Lumakras saw a reduction in tumor size compared to 63 percent on chemo. The disease control rate was 82.5 percent on Lumakras versus 60.3 percent on docetaxel.

The one-year progression-free survival rate was 24.8 percent for Lumakras-treated patients versus 10.1 percent in the chemo arm. In the Lumakras arm, the median progression-free survival was 5.6 months versus 4.5 months in the chemo arm.

However, median overall survival was not significantly different between treatment arms. On Lumakras, the median overall survival was 10.6 months compared to 11.3 months on docetaxel. Patients in the chemo arm were allowed to cross over to Lumakras once they progressed, which may have reduced the statistical power. In presenting the results from CodeBreaK-200, Melissa Johnson, program director of lung cancer research at the Sarah Cannon Cancer Institute, noted that the study was not powered to assess survival.

"This first-in-class KRAS G12C inhibitor showed a significant improvement in primary endpoint of progression-free survival, with a 34 percent risk reduction in terms of disease progression," Johnson said. "These findings support sotorasib as a new second-line standard in the treatment of patients with NSCLC harboring KRAS G12C mutations."

While the results showed an improvement over chemo, which remains the standard of care in this setting, Natasha Leighl, leader of the thoracic medical oncology group at the Princess Margaret Cancer Center in Toronto, voiced some disappointment in a discussion of the results. Lumakras "performed a little bit less well than we had hoped. The magnitude of benefit is a concern; it's not what we'd hoped," Leighl said. She noted the lack of survival benefit even at 18 months of follow-up, suggesting that the crossover built into the study from the chemo arm to the Lumakras arm may not fully account for the lack of survival benefit.

"Could we be missing other things? Perhaps imbalances in concomitant poor prognostic alterations, such as KEAP1?" she posited. A previous study of Lumakras in NSCLC found that patients with a co-occurring KEAP1 mutation were less likely to see long-term benefit from treatment with the drug.

Leighl also suggested that more research is needed to better understand mechanisms of primary and acquired resistance to KRAS G12C inhibitors and improve response rates. Johnson also acknowledged that resistance was a concern for KRAS inhibitors.

"We've known about the presence of KRAS mutations in lung cancer for 30 years, and we've been trying to drug them for much longer," Johnson said. "The hyperactivity of the MAPK pathway is a concern, and, as we know from other drugs and other tumor types, resistance develops very quickly. Perhaps a combination regimen [with a KRAS inhibitor] will be more effective."

As Amgen works to prove Lumakras benefit in KRAS-mutated NSCLC against the standard of care, Mirati is also building its data on adagrasib. The firm presented data from a single-arm trial of the drug in KRAS G12C-mutated NSCLC earlier this year, where half of the patients responded to treatment. That data also showed the drug had activity in NSCLC patients with brain metastases. Among 19 patients with active, untreated central nervous system metastases, 32 percent experienced an intracranial response on adagrasib. 

There are many other players evaluating KRAS inhibitors in metastatic NSCLC. For example, Novartis presented data in April from its KRAS G12C inhibitor, JDQ443, which demonstrated a 28.2 percent response rate in an early-stage study.

KRAS inhibitors in colorectal cancer

Outside of lung cancer, both Amgen and Mirati are exploring their drugs in combination with EGFR-inhibiting monoclonal antibodies in colorectal cancer. Lumakras was paired with Amgen's Vectibix (panitumumab) in the CodeBreaK-100 study and adagrasib with Eli Lilly's Erbitux (cetuximab) in the KRYSTAL-1 trial.

"In KRAS G12C colorectal tumors, we know now [they] require EGFR and MAPK blockade in order to have increased activity of these two compounds," explained Maria Elena Elez Fernandez, principal investigator of the gastrointestinal tumor group at Vall d'Hebron Institute of Oncology, in discussing the data from these trials.

In the colorectal cancer cohort of the CodeBreaK-100 study, the Lumakras-Vectibix combination demonstrated a 30 percent response rate, with 12 partial responses among 40 evaluable patients with KRAS G12C-mutated metastatic colorectal cancer. The disease control rate, including responses and stable disease, was 93 percent.

Yasutoshi Kuboki, a gastrointestinal oncologist at National Cancer Center Hospital East in Japan, said 88 percent of patients saw some reduction in the size of their tumor from the combination treatment. The median progression-free survival was 5.7 months at a median follow-up of 11 months. At six months, 41.7 percent of patients had not progressed, and at nine months 12.3 percent still hadn't progressed on treatment.

"There appeared to be no differences in response based on the tumor location on the left or right side," Kuboki said in a presentation of the CodeBreaK-100 data. Of 27 patients with left-sided colorectal tumors, 30 percent responded, while 31 percent of 13 patients with right-sided tumors did. There was also no difference in median progression-free survival between the two groups.

At 8.8 months of follow-up, median overall survival was not yet reached in the colorectal cancer cohort in CodeBreaK-100. The six-month overall survival rate was 91.5 percent, while at nine months, 82.5 percent of patients were alive.

Kuboki noted that previously reported data from a colorectal cancer cohort that received Lumakras as a monotherapy showed a "modest benefit," but "when combined with panitumumab, the confirmed objective response rate increased by threefold," he said.

"Similarly, the progression-free survival was 5.7 months in the current [combination] study compared to four months reported for sotorasib monotherapy," Kuboki continued. After the disappointing monotherapy results, Amgen pivoted to focus on combination approaches in this cancer type.

Amgen is already enrolling patients for the registrational, randomized Phase III CodeBreaK-300 study evaluating Lumakras plus Vectibix against investigator's choice of standard treatment in patients with KRAS G12C-mutant metastatic colorectal cancer.

The story at Mirati is similar. In testing out monotherapy and combination approaches with adagrasib in colorectal cancer, the latter has shown more promise. At ESMO, researchers presented results from its Phase II KRYSTAL-1 study evaluating adagrasib with or without Erbitux in KRAS G12C-mutant metastatic colorectal cancer patients.

"We have very limited late-line options [in colorectal cancer], and we need to do better in order to serve our patients," Samuel Klempner, an associate professor at Massachusetts General Hospital and Harvard Medical School, said in a presentation at ESMO. "Adagrasib with or without cetuximab continues to demonstrate encouraging clinical activity in this heavily pretreated patient population of G12C-mutant colorectal cancer."

In the monotherapy portion of the study involving 40 patients, the response rate was 19 percent with eight partial responses. The disease control rate was 86 percent, and 79 percent of patients saw some tumor shrinkage. The median progression-free survival with single-agent adagrasib was 5.6 months and median overall survival was 19.8 months.

However, the combination of adagrasib and Erbitux demonstrated greater benefit in this patient population. Among 28 patients, 46 percent responded, with 13 partial responses. The disease control rate was 100 percent, meaning there were no patients with progressive disease on the combo treatment. All but two patients (93 percent), saw a reduction in tumor size.

Median progression-free survival in the combo group was 6.9 months. The median overall survival was 13.4 months with 84 percent of patients alive at six months and 61 percent alive at 12 months.

Mirati is continuing to explore both the monotherapy and combination approaches in colorectal cancer. The firm is conducting the registrational Phase III KRYSTAL-10 trial evaluating adagrasib and Erbitux against chemo as a second-line treatment for KRAS G12C-mutant advanced colorectal cancer. There is also a separate, potentially registrational monotherapy cohort of adagrasib in third-line metastatic colorectal cancer in the KRYSTAL-1 study, noted Klempner.

"We have good benchmark data, and validating a good molecular target with a good drug in an earlier-line setting is a paradigm that has been successful in multiple targets," Klempner said of the KRYSTAL-10 study. "Later lines, where we're all desperate for new therapies, is the lower hanging fruit, but … moving it earlier and getting more patients exposed to these therapies and giving them the opportunity for deep response is a worthwhile endeavor."

Elez Fernandez also mentioned the need for larger studies and greater fine-tuning of KRAS inhibitors in colorectal cancer, such as better patient selection or identifying biomarkers of response.

"If we take a look to the evolution of the incorporation of targeted therapies in metastatic colorectal cancer, the development of KRAS inhibitors has been really fast," she said. "The works that have been presented today are the result of the incorporation of new paradigms in the development of therapeutics that provide the possibility to design better treatment options for these patients."