FDA Approval Summary: Sotorasib for KRAS G12C-Mutated Metastatic NSCLC

Locally advanced or metastatic non–small cell lung cancer (NSCLC) with KRAS G12C mutation is a genetically distinct form of lung cancer that is not curable with available therapy and comprises approximately 14% of advanced NSCLC cases. The Kirsten rat sarcoma proto-oncogene (KRAS) G12C mutation leads to persistent downstream signaling to pathways promoting cell growth and division including the RAF-MEK-ERK pathway. KRAS has been described historically as an “undruggable target,” with multiple failed therapeutic efforts (1). Previous challenges to successful targeting of mutant KRAS included the risk of significant toxicity from off-target wildtype KRAS inhibition, the high affinity of KRAS for GTP which thereby activates the signaling protein, and the perceived lack of suitable drug binding pockets (2, 3).

Unlike patients with NSCLC harboring other oncogenic driver mutations (e.g., EGFR and anaplastic lymphoma kinase), who are generally never-smokers, patients with KRAS-mutated NSCLC frequently have a smoking history (1, 4). KRAS-mutated NSCLC is often associated with high programmed death-ligand1 (PD-L1) expression, high tumor mutational burden, and responsiveness to first-line immunotherapy. This last feature is distinct from other subtypes of lung cancer harboring a targetable mutation (5, 6).

Prior to the approval of sotorasib, patients with KRAS G12C-mutated NSCLC were treated with therapies approved for NSCLC without a targetable mutation. First-line standard therapies include immunotherapy alone or in combination with platinum-based chemotherapy. The addition of immunotherapy to lung cancer treatment regimens has improved the median overall survival of patients with KRAS G12C-mutated NSCLC from 10–20 months with chemotherapy alone to 21–28 months (7, 8). However, the vast majority of patients with KRAS G12C-mutated NSCLC progress on first-line chemo-immunotherapy. Second-line treatment options are docetaxel monotherapy or in combination with ramucirumab, both intravenous therapies with limited efficacy (9).

Sotorasib is the first FDA-approved drug for advanced, KRAS G12C-mutated NSCLC. Herein, we provide a summary of FDA's review of the marketing application that led to the approval of sotorasib for patients with advanced NSCLC that harbors a KRAS G12C mutation and progressed after at least one line of therapy.

Sotorasib was granted Orphan Drug Designation for the treatment of KRAS G12C-mutated NSCLC on May 1, 2019 and received Breakthrough Therapy Designation on December 7, 2020. The NDA was submitted on December 16, 2020, and used the Real-Time Oncology Review pilot program. The application voluntarily utilized the Assessment Aid to facilitate FDA's review (10).

KRAS is an oncogene that encodes the small guanosine 5′-triphosphate hydrolase (GTPase) signal-transduction protein KRAS (11). It is a member of the RAS family of proteins, which function as molecular switches by alternating between inactive GDP-bound and active GTP-bound states. Binding of GTP to RAS activates downstream signaling through multiple pathways including RAF/MEK/ERK pathway (12). Intrinsic GTPase and GDP–GTP exchange rates can vary among the different RAS mutants. The KRAS G12C-activating mutation reduces GAP-mediated GTP hydrolysis resulting in accumulation of active GTP-bound KRAS (13, 14). KRAS-G12C exhibits near-wild-type intrinsic GTPase activity, thus enabling the development of covalent inhibitors that bind to the GDP-bound state of KRAS-G12C (15).

Previously unexploited utilization of a surface groove in KRAS G12C led to the development of sotorasib (1). Sotorasib is an oral small-molecule inhibitor of the RAS GTPase family that binds irreversibly to the P2 pocket in the inactive GDP-bound form of KRAS. Sotorasib forms an irreversible, covalent bond with the unique cysteine of KRAS G12C, locking the protein in an inactive state (1). Sotorasib thereby blocks KRAS signaling, inhibits cell growth in vitro, tumor growth in vivo, and promotes apoptosis only in KRAS G12C tumor cell lines (16).

FDA reviewed subject-level data collected in CodeBreaK 100 (Study 20170543) and healthy volunteer studies to characterize the pharmacokinetics, pharmacodynamics, food effect, and drug–drug interactions of sotorasib. Systemic exposures (i.e., AUC_0–24h and C_max) of sotorasib after daily administration of doses ranging from 180 to 960 mg were similar (Table 1; Fig. 1). In addition, overall response rates (ORR) ranging from 25%–50% were observed in patients with NSCLC who were treated with lower doses of sotorasib in the dose-escalation portion of the trial (n ranging from 3–16 per dose level; Table 2).

Sotorasib is primarily metabolized by CYP3A4 and is mainly eliminated via the fecal route with minor renal excretion. Sotorasib may be taken with or without food. No clinically significant differences in the exposure of sotorasib were identified based on age (28–86 years), sex, body weight (37–158 kg), mild and moderate renal impairment or mild hepatic impairment. Sotorasib USPI recommends specific dose modification strategies for concomitant use of sotorasib with CYP3A inducers and substrates, gastric acid–reducing agents, and P-glycoprotein substrates with a narrow therapeutic window.

The approval of sotorasib was based upon the results of CodeBreaK 100, a global, multicenter, multicohort, dose-escalation, and dose-expansion trial that enrolled 427 patients with advanced KRAS G12C-mutated solid tumors, including 250 patients with NSCLC, as of September 1, 2020. All enrolled patients received at least one prior systemic therapy for their malignancy except in the treatment-naïve NSCLC cohort. Oral doses of 180, 360, 720, and 960 mg daily were studied in the dose-escalation portion of the trial. Patients in the dose-expansion portion of the study received sotorasib 960 mg daily in 21-day cycles.

The primary efficacy endpoint of CodeBreaK 100 to support marketing approval was the ORR as determined by blinded independent central review (BICR); duration of response (DOR) was a key secondary efficacy endpoint.

The primary efficacy analysis population included 124 patients with KRAS G12C-mutated NSCLC who had at least one measurable lesion at baseline by RECIST 1.1 and received sotorasib 960 mg daily. Patients were primarily current or former smokers (93%), and most had received an immune checkpoint inhibitor (91%) or platinum-based chemotherapy (90%) previously. A total of 81% of patients had progressive disease after both platinum-based chemotherapy and an immune checkpoint inhibitor. The majority of the patients in the primary efficacy population were White (82%) with African American and Asian patients represented at 2% and 15%, respectively.

The ORR of the primary efficacy population as determined by BICR was 36% (95% CI, 28–45). The median DOR was 10.0 months (range, 1.3+ to 11.1) with 58% of responders experiencing a DOR ≥6 months (Table 3). These results are based upon updated data provided to FDA at the time of new drug application (NDA) submission.

Toxicological assessment of sotorasib was performed in rats and Beagle dogs. In the 3-month rat toxicology study, renal toxicity was most notable, and characterized by renal tubular degeneration/necrosis and increased kidney weight with biomarkers of renal tubular injury. Partial recovery was observed after 2 months. Increases in cysteine S-conjugate β-lyase pathway metabolism in the rat kidney compared with human (17) may make rats more susceptible to renal toxicity due to local formation of a putative sulfur-containing metabolite (18). Renal toxicity was not identified in the 3-month dog toxicology study, although sotorasib induced findings in the liver, pituitary gland, and thyroid gland in dogs that may be due to an adaptive response to hepatocellular enzyme induction and secondary hypothyroidism (16, 19, 20).

Safety analyses were conducted in all 204 patients with NSCLC enrolled in Study 20170543 who received at least one dose of sotorasib 960 mg daily in the fed or fasted state. Among patients in this primary safety population, 39% were exposed for at least 6 months and 3% were exposed for at least 1 year. The most common (>10%) adverse events (AE) were diarrhea (42%), musculoskeletal pain (35%), hepatotoxicity (25%), nausea (26%), fatigue (26%), cough (20%), vomiting (17%), constipation (16%), dyspnea (16%), abdominal pain (15%), edema (15%), decreased appetite (13%), pneumonia (12%), and rash (12%). Serious AEs occurred in 50% of patients treated with sotorasib, the most common of which were pneumonia (8%), hepatotoxicity (3.4%), and diarrhea (2%). Permanent discontinuation of sotorasib due to an AE occurred in 9% of patients; the most common reason for discontinuation was hepatotoxicity (4.9%).

Interstitial lung disease/pneumonitis and hepatotoxicity were identified as potentially significant safety issues. Interstitial lung disease/pneumonitis occurred in 3 patients in the primary safety population (1.5%), with one fatal case. All 3 patients had received prior treatment with an immune checkpoint inhibitor; however, the contribution of sotorasib could not be ruled out.

In FDA's safety review, hepatotoxicity was defined as a group including preferred terms alanine aminotransferase increased, aspartate aminotransferase increased, blood bilirubin increased, drug induced liver injury, hepatitis, hepatotoxicity, liver function test increased, and transaminases increased. While hepatotoxicity was one of the most commonly observed AEs, only 12% of hepatotoxicity events were severe (grade 3 or 4), and there were no fatal events. Of the 50 patients in the primary safety population who experienced an event of hepatotoxicity, 46% (n = 23) required either a dose reduction or interruption, which resulted in resolution in 91% of the cases. In addition, 5% of the 357 patients treated with sotorasib 960 mg daily with an event of hepatoxicity were treated with systemic corticosteroids as part of its management. In all cases, steroids were administered in combination with dose modifications so the degree to which steroids contributed to hepatotoxicity mitigation is unclear.

While preclinical studies with sotorasib were concerning for a potential nephrotoxicity signal, renal toxicity events in the primary safety population were rare, mild to moderate, and did not require dose reduction or discontinuation.

The FDA review team considers the safety profile of sotorasib to be acceptable for the intended population in the context of a life-threatening disease. Information in the Warnings and Precautions and Dosage and Administration sections of the product labeling provides guidance to clinicians on monitoring and managing toxicities associated with sotorasib (16).

The approval of sotorasib provides a new treatment option for patients with KRAS G12C-mutated NSCLC for whom a targeted therapy has been elusive for decades (Table 4). While the use of immunotherapies has improved overall survival for patients with KRAS G12C-mutated NSCLC, later lines of therapies have resulted in minimal benefit (7, 8, 21). To confirm the clinical benefit of sotorasib in patients with KRAS G12C-mutated advanced NSCLC with progression after at least one prior therapy, Amgen is conducting CodeBreaK 200 (Study 20190009) which compares the efficacy of sotorasib as a single agent to docetaxel, in the same refractory population as CodeBreaK 100, with progression-free survival as a primary endpoint. Although the clinical benefit of therapeutics granted accelerated approval for an oncologic indication is often confirmed in trials evaluating an earlier line of therapy, there may be inherent challenges in combining tyrosine kinase inhibitors with standard first-line therapies for NSCLC (immunotherapy with or without platinum-based chemotherapy) due to the additive toxicity of the regimens. Furthermore, the addition of immunotherapy to platinum-based chemotherapy for the treatment of patients with NSCLC without targetable oncologic driver mutations has resulted in a markedly improved overall survival over platinum-based chemotherapy alone, offering patients with KRAS G12C-mutated NSCLC an effective first-line therapy.

As a part of the accelerated approval, a post-marketing requirement (PMR) was issued for a dose optimization study to investigate the safety and efficacy of a lower sotorasib daily dose as compared with the 960 mg daily dose. Sotorasib exhibited similar systemic exposures with multiple daily dosing across all dose levels (180–960 mg) administered in CodeBreaK 100. Tumor responses were observed at lower sotorasib doses; however, the small numbers of patients enrolled at the lower dose levels resulted in wide and overlapping 95% confidence intervals.

During the NDA review period, agreement on the design of a PMR study to evaluate the antitumor activity and safety of sotorasib administered at 240 mg daily (two 120 mg tablets) compared to the approved dose (960 mg; eight 120 mg tablets) was reached. If a lower dose of sotorasib is found to be equally effective, patients may benefit from a reduced daily pill burden from the eight pills a day required for 960 mg of sotorasib. FDA is developing Project Optimus to investigate strategies to improve dose optimization of oncology therapies (22).

To address the limited numbers of African American patients in CodeBreaK 100, a post-marketing commitment was issued for the submission of results of clinical trials enrolling a sufficient number of African American patients with KRAS G12C-mutated NSCLC to further characterize the safety and efficacy of sotorasib in this patient population (23). Of note, a total of 15% of the study population was Asian, which the FDA considers to be sufficiently representative of the overall population of Asian patients with KRAS G12C-mutated NSCLC. In Asian countries, the KRAS G12C mutation has been reported as having a low incidence (∼4% in an analysis of 11,951 tumor samples collected in China; ref. 24).

As a part of the Agency-wide effort to assess the impact of the COVID-19 pandemic on clinical trials, FDA evaluated the protocol changes and protocol deviations that occurred in CodeBreaK 100 due to the pandemic. Notable protocol modifications included permitting alternative sites for imaging and laboratory assessments, telehealth assessments for AEs, and direct shipment of sotorasib to patients' homes. In the primary efficacy population, 13 patients (10%) had ≥ 1 important protocol deviation related to the COVID-19 pandemic, including missing data (excluding missing data for incorrect treatment, dose, or other treatment compliance; 10 patients, 8%), received an incorrect, incomplete, or partial dose of sotorasib (3 patients, 2.4%), and off-schedule procedures (1 patient, 0.8%). These study adaptations did not affect overall efficacy endpoints or interpretation of study results.

This application was reviewed under FDA's Project Orbis, in collaboration with several international Orbis partners. Australia's Therapeutic Goods Administration, Brazil's Health Regulatory Agency, Health Canada, and the United Kingdom's Medicines and Healthcare Products Regulatory Agency.

Treatment with sotorasib results in a response rate that represents a clinically meaningful improvement over available therapies for second- and later-line treatment. This is a landmark approval for a population with a significant unmet medical need for which targeted therapies have remained elusive for decades. Mutant KRAS has been difficult to target due to its structure (25) and the results of CodeBreaK 100 are notable. Although the approved dose of sotorasib 960 mg daily may not be fully optimized, the overall risk-benefit profile of this novel treatment is positive at the currently approved dose. Future directions for improvement on the treatment options for patients with KRAS G12C-mutated NSCLC include the study of combination therapies as first-line treatment options for patients with metastatic disease, assessment of the efficacy of KRAS G12C inhibitors in the central nervous system, and the development of therapies and regimens targeting resistance mutations that may develop after treatment with sotorasib. The landmark development and approval of sotorasib is a significant advance for patients with KRAS G12C-mutated NSCLC, providing a new and long-awaited treatment option for a refractory population of patients with lung cancer.

No disclosures were reported.

The Editor handling the peer review and decision-making process for this article has no relevant employment associations to disclose.