NEW YORK – Based on the efficacy and safety observed with Lumakras (sotorasib) and checkpoint inhibitors, Amgen is refining its combination treatment strategies in first-line advanced non-small cell lung cancers bearing a KRAS G12C mutation.
At the World Conference on Lung Cancer in Vienna this weekend, researchers presented data from a cohort in the Phase Ib CodeBreaK 100/101 master protocol trial, in which KRAS G12C-mutated advanced NSCLC patients received Lumakras with Merck's anti-PD-1 drug Keytruda (pembrolizumab) or Genentech's PD-L1 inhibitor Tecentriq (atezolizumab). While less than a third of the patients responded to these KRAS-checkpoint inhibitor combinations, there were unexpectedly high rates of treatment-related adverse events, primarily asymptomatic elevations of the liver enzymes AST and ALT.
A total of 58 patients received these combinations and were followed for a median of 12.8 months. The 29 percent response rate with Lumakras-Keytruda or Lumakras-Tecentriq fell short of the 40.7 percent previously reported for Lumakras monotherapy.
Still, Amgen hasn't soured on exploring combination strategies with its KRAS inhibitor. The drugmaker took some encouragement from its trial of Lumakras with Revolution Medicines' SHP2 inhibitor RMC-4630, since all three patients with NSCLC who were KRAS inhibitor naïve experienced disease control.
Based on these results and data from other trials of Lumakras, Amgen is laying out different paths for the agent in the first-line setting using patients' PD-L1 status. While the company has sufficient evidence to move forward with a Phase III trial of Lumakras with chemotherapy in the first-line NSCLC setting involving PD-L1-negative patients, it will continue studying Lumakras combinations with Keytruda and a SHP2 inhibitor in patients with some level of PD-L1 expression in CodeBreaK expansion trials.
The KRAS G12C mutation has been a breakthrough opportunity in targeting the Teflon-like KRAS protein. KRAS is the most commonly mutated gene in the RAS oncogene family and KRAS mutations are believed to be the most common drivers in human cancer. However, attempts over the past 40 years to target KRAS with drugs have not been successful.
Inside tumor cells with the KRAS G12C mutation, however, the change of glycine to cysteine at position 12 on the amino acid chain offers a "handle" for KRAS inhibitors to grab. In 2021, Amgen was the first to market with its KRAS inhibitor Lumakras for G12C-mutated NSCLC, and the company has reported that oncologists are widely prescribing it when their patients' KRAS G12C mutation status is known in a timely fashion.
However, the reported median progression-free survival of 6.3 months reflects the cancer's ability to evade KRAS inhibition, which has spurred Amgen's efforts to try to stave off progression longer through combination treatment strategies. Some mechanisms believed to help cancer cells evade KRAS inhibition include immune escape through upregulation of PD-L1 expression and adaptive reactivation of KRAS by SHP2. These two hypotheses provided Amgen the rationale for pairing Lumakras with PD1/PD-L1 inhibitors or a SHP2 inhibitor.
In a call with market analysts on Monday, Bob Li, a Memorial Sloan Kettering Cancer Center oncologist and a CodeBreaK lead investigator, said that Amgen's preclinical data from tumor cells and mouse models showed that a Lumakras combination with anti-PD1 or -PD-L1 agents "led to dramatic tumor shrinkage and durable response," prompting the company to explore that strategy.
Amgen used a flexible dose-finding design, anticipating there might be some toxicities. "We know that, historically, combining targeted therapy and immunotherapy has been challenging," said Li, who presented the latest CodeBreaK findings at WCLC.
The trial design permitted Lumakras doses to be escalated and de-escalated ranging from 960 mg down to 120 mg "as clinically appropriate." It also included both concurrent combination regimens, in which both drugs were started at the same time, and lead-in regimens, in which Lumakras was given for three weeks or six weeks as a single agent, and then followed by combination therapy with Keytruda or Tecentriq.
Treatment-related adverse events led about half of those receiving the highest doses of Lumakras with Keytruda or Tecentriq in the concurrent groups to stop taking the combined therapies. Li attributed the weaker efficacy results to the high rate of discontinuations plus the fact that two-thirds of the patients had been previously treated with PD-1 or PD-L1 inhibitors.
Although the liver enzyme elevations were reversible with dose modifications and corticosteroid treatment, Li said that "from a drug development point of view," the concurrent high-dose regimen for Lumakras with immunotherapy "is not a tenable combination for further development."
In comparison with the high-dose concurrent regimens, patients leading in with a low dose of Lumakras alone for three or six weeks before adding Keytruda or Tecentriq had fewer grade 3 or 4 adverse events, especially elevated liver enzymes. For example, Li said the rate of grade 3 or higher elevated liver enzymes was around 20 percent in this group. "Due to this experience, a low dose of sotorasib as lead-in, followed by a combination with pembrolizumab, is now being further studied as a potential first-line treatment for these patients," Li said. Amgen has said that it will pursue a lead-in schedule starting with a 240 mg dose of Lumakras for several weeks, after which Keytruda will be added.
Li said there remains significant unmet need for first-line NSCLC patients with KRAS G12C mutations treated with chemotherapy or chemotherapy plus immunotherapy. "It's just not good enough that the progression-free survival is nine months," Li said. "We hope to really improve the duration of benefit [in terms of] progression-free survival and duration of response endpoints. And we are seeing a glimmer of hope in the Phase I study of this combination. We're seeing a longer duration of clinical benefit in some patients."
Among the 17 patients who responded to these combinations, the duration of response was 17.9 months.
Regarding the choice to move forward with Keytruda rather than Tecentriq, Amgen Executive VP of R&D David Reese said, looking at the data from the trial, and taking into account the small numbers of patients in each subgroup of the trial, "I don't think you can conclude that there's any real difference in the profiles between the PD-1 versus PD-L1 inhibitors. We'd like to move forward with the [drug] more commonly used in the first-line setting."
Li added, "This is more about practice patterns, and pembrolizumab just has such a dominant role in the first-line setting … I have no opposition to studying atezolizumab or nivolumab or any other PD-L1 [inhibitor], from a scientific standpoint. This is a pretty practical choice."
Amgen also presented results of a dose-finding study of Revolution Medicines' SHP2 inhibitor, RMC-4630, with Lumakras in KRAS G12C-mutated NSCLC, CRC, or other solid tumors. A total of 27 patients in the study received 960 mg of Lumakras with doses ranging from 100 to 200 mg of RMC-4630. Among the 11 patients with NSCLC, there was a 27 percent rate of partial response. However, most of them were "heavily pretreated," according to Li. Among the three patients that were naïve to KRAS inhibitors, all three were in the group that had a partial response. Seven of the 11 NSCLC patients achieved disease control.
"We saw promising and durable clinical activity in the KRAS G12C inhibitor-naïve setting in non-small cell lung cancer," said Li, noting that based on this finding, the Lumakras/RMC-4630 combo has been advanced into a Phase II study.