Lung · Cheat sheet
KRAS G12C in non-small cell lung cancer
2026 evidence summary · Last updated 2026-05-12 · Reviewed against AMP/ASCO/CAP 2017
At a glance
- Frequency: ~13% of lung adenocarcinomas in US/EU populations. Less common in colorectal cancer (~3%) and other tumors.
- HGVS: p.Gly12Cys / c.34G>T.
- Mechanism: Glycine-to-cysteine substitution at codon 12 enables covalent binding of selective KRAS G12C inhibitors.
- Approved post-platinum (NSCLC): sotorasib (Lumakras) FDA-approved May 2021; adagrasib (Krazati) FDA-approved December 2022.
- Critical caution: STK11 and KEAP1 co-mutations are negative modifiers for both KRAS G12C inhibitors and ICI.
Biology and mechanism
KRAS is a membrane-anchored GTPase that cycles between active (GTP-bound) and inactive (GDP-bound) states. The G12C substitution introduces a reactive cysteine residue at codon 12 in the switch-II pocket. This cysteine is the molecular handle that selective KRAS G12C inhibitors (sotorasib, adagrasib) covalently bind, locking the protein in its inactive GDP-bound conformation. Other KRAS codon 12 substitutions (G12D, G12V, G12A) lack this reactive cysteine and require different inhibitor chemistry.
KRAS sits upstream of the RAF/MEK/ERK and PI3K/AKT cascades. Inhibiting KRAS G12C reduces signaling through both axes, but feedback reactivation through wild-type KRAS or through RAS-pathway bypass (HER2, MET, EGFR amplification) is common and underlies most acquired resistance.
Epidemiology
KRAS G12C is the most common KRAS substitution in NSCLC, accounting for roughly 40% of KRAS-mutant lung adenocarcinomas. It is strongly associated with smoking history. Squamous histology and never-smoker adenocarcinoma show much lower prevalence. In colorectal cancer, KRAS G12C accounts for ~3% of all KRAS mutations and has its own approval landscape distinct from NSCLC.
FDA-approved therapies (NSCLC, post-platinum)
| Drug | Approval | Source | Notes |
|---|---|---|---|
| Sotorasib (Lumakras / Lumykras) | FDA accelerated approval May 2021; updated label per CodeBreaK 200 confirmatory data | openFDA label; CodeBreaK 100 | Standard dose 960 mg daily. Hepatotoxicity is the most common dose-limiting toxicity; baseline and serial LFTs are required. |
| Adagrasib (Krazati) | FDA accelerated approval December 2022 | openFDA label; KRYSTAL-1 | Prolonged QT-interval risk requires baseline ECG and electrolyte management. Different toxicity profile from sotorasib. |
First-line context (NSCLC)
KRAS G12C is not currently approved for first-line single-agent KRAS-G12C-inhibitor therapy in NSCLC. First-line in 2026 remains platinum-based chemotherapy + immune checkpoint inhibitor combinations, or ICI monotherapy when PD-L1 expression is high. Active investigation includes:
- Sotorasib + ICI combinations (CodeBreaK 100, 101, 200 series).
- Adagrasib + ICI combinations (KRYSTAL-7).
- Sotorasib or adagrasib added to chemotherapy in earlier lines.
Positioning is changing fast. Verify current label and trial readouts at decision time.
Resistance pathways and management
Acquired resistance to KRAS G12C inhibitors emerges in most patients within months. Mechanisms cluster into several patterns:
| Mechanism | Notes |
|---|---|
| Secondary KRAS mutations (Y96D, R68S, H95Q) | Restore KRAS activity by altering the switch-II pocket. Detection in liquid biopsy at progression supports resistance attribution. |
| Acquired EGFR / HER2 / MET amplification | Bypass pathways. Combination strategies under investigation. |
| Histologic transformation (adeno-to-squamous) | Re-biopsy can be diagnostic. Treatment shifts accordingly. |
| EMT / non-genetic mechanisms | Active investigation; harder to detect on a molecular panel. |
| Activating mutations in PI3K, MAP2K1, or BRAF | Reactivate downstream signaling. |
Co-mutations that change treatment selection
- STK11 (LKB1) co-mutation: negative modifier for both KRAS G12C inhibitors and ICI. Worse response and shorter PFS in retrospective and prospective series.
- KEAP1 co-mutation: similar negative-modifier signal for ICI; KEAP1 alterations also affect oxidative stress response and chemo sensitivity.
- Concurrent STK11 + KEAP1: worst response signal among these co-mutations. Strong consideration for clinical trial enrollment.
- TP53 co-mutation: common; generally does not change G12C-inhibitor selection but affects overall prognosis.
Detection and companion diagnostics
| Test | Specimen | Indication |
|---|---|---|
| QIAGEN therascreen KRAS RGQ PCR Kit | Tissue | Companion diagnostic for sotorasib |
| Agilent Resolution ctDx FIRST | Plasma | Companion diagnostic for adagrasib |
| FoundationOne CDx, Guardant360 CDx | Tissue / plasma | Comprehensive panels report KRAS G12C as part of standard solid-tumor profiling |
Tissue is preferred when feasible. Plasma is appropriate for monitoring and for patients without sufficient tissue for repeat NGS. Concordance is good but not perfect — interpret discordance carefully.
Trial pointers
ClinicalTrials.gov is the source of truth. Useful filter combinations for KRAS G12C NSCLC:
- "KRAS G12C" + "Lung Cancer" + recruiting → typically 30–60 active studies at any time.
- High-volume sponsors include Amgen, Mirati/Bristol Myers Squibb, and Revolution Medicines (RAS-MAPK pipeline including pan-KRAS and pan-RAS approaches).
- Search "G12C immune checkpoint inhibitor combination" for ongoing combination readouts.
- Search "pan-RAS inhibitor lung cancer" for next-generation pan-KRAS or pan-RAS programs.
Pharmacogenomic considerations
Sotorasib is metabolized primarily by CYP3A4 with secondary contribution from UGT enzymes. Strong CYP3A4 inducers (rifampin, carbamazepine, phenytoin, St. John's wort) substantially reduce sotorasib exposure; alternative therapy or dose adjustment is required.
Adagrasib carries QT-interval prolongation risk. Concurrent QT-prolonging medications require baseline ECG, magnesium and potassium repletion, and serial monitoring. Avoid concurrent strong CYP3A4 inhibitors; concurrent strong CYP3A4 inducers reduce exposure.
Citation pointers (identifier-only, license-clean)
- CIViC EID6184 — KRAS G12C + sotorasib.
- PubMed 33606975 — Hong et al., CodeBreaK 100, NEJM 2020.
- PubMed 35658005 — Jänne et al., KRYSTAL-1, NEJM 2022.
- AMP/ASCO/CAP 2017 — Li MM et al., J Mol Diagn 2017;19(1):4-23.
- openFDA labels for sotorasib (Lumakras), adagrasib (Krazati).
Caveats
- Decision-support reference, not CAP/CLIA-validated diagnostic.
- KRAS G12C in colorectal cancer follows a different treatment landscape; adagrasib + cetuximab (KRYSTAL-1 CRC cohort) has distinct positioning. Drug-gene match depends on tumor type.
- The first-line context for KRAS G12C is changing fast as ICI combination data matures. Verify current label and trial readouts at decision time.
- Trial-recruiting status changes daily.
Frequently asked questions
- Is sotorasib or adagrasib first-line in KRAS G12C NSCLC?
- Neither is currently approved as first-line single-agent therapy. FDA-approved indications are for previously treated locally advanced or metastatic disease (post-platinum). First-line in 2026 remains platinum-based chemotherapy + ICI, or ICI alone in PD-L1-high. Combination ICI + KRAS G12C inhibitor trials may change this.
- How do STK11 and KEAP1 co-mutations affect KRAS G12C treatment?
- STK11 and KEAP1 co-mutations are negative modifiers for both KRAS G12C inhibitors and ICI. Their presence shapes prognosis discussions and supports consideration of clinical trial enrollment.
- What companion diagnostics are FDA-approved for KRAS G12C NSCLC?
- QIAGEN therascreen KRAS RGQ PCR Kit (tissue) is the companion diagnostic for sotorasib. Agilent Resolution ctDx FIRST (plasma) is the companion diagnostic for adagrasib. Most comprehensive NGS panels also detect KRAS G12C.
Need to interpret a vendor NGS report?
UNMIRI's pathologist tool unifies Foundation Medicine, Tempus, Caris, Guardant, and other vendor reports into one normalized view. Free for individual clinical use.