Lung cancer is the second most diagnosed cancer worldwide and the leading cause of cancer death, with non-small-cell lung cancer (NSCLC) accounting for over 80% of total cases ^[1]. Prior to 2002, cytotoxic chemotherapy was the only treatment for advanced NSCLC ^[2]. Prognosis was dismal, with 5-year overall survival rates of 16.9% across all stages in 2000 ^[3]. According to recent data, the availability and use of targeted therapies and immune checkpoint inhibitors have improved NSCLC prognoses significantly to a 5-year survival rate of 28% across all stages in 2023 ^[1]. The incidence of NSCLC has also been dropping worldwide over the last decade, largely due to declining rates of smoking ^[1]. Despite this, NSCLC remains a deadly condition, with significant associated morbidity and mortality for its sufferers. A large proportion of lung cancers present with advanced or metastatic disease at diagnosis, precluding curative options of treatment ^[1]. For the 30% or so of patients with a surgically resectable disease at diagnosis ^[1], over half may experience a recurrence of their disease post-operatively ^[4].

Improved understanding of the cell signalling pathways implicated in lung cancer tumorigenesis has led to developments in technologies allowing for the detection of actionable genetic mutations and, subsequently, the development of drugs to block these driver mutations. Lung cancers can accumulate a large mutational burden, and several targetable pathways have been identified for lung adenocarcinoma. These include EGFR, ALK, ROS1, HER2, MET, RET, BRAF, NTRK, and NRG1 fusions ^[5]. Systemic drug therapies targeting these pathways have yielded impressive and clinically significant improvements in outcomes, and some have even now replaced chemotherapy as first-line treatment strategies. The epidermal growth factor receptor (EGFR) mutation pathway is perhaps the most well-known success story highlighting this, with its first inhibitor, gefitinib, seeing Food and Drug Administration (FDA) approval in 2003 ^[6]. Since then, rapid advancement has been made in targeted therapy for this pathway, leading to the approval of osimertinib, a third-generation EGFR inhibitor, in 2018 for first-line management of advanced NSCLC ^[7]. Not only has osimertinib shown impressive OS and PFS benefits in the metastatic setting, but also its use in the adjuvant setting has now been shown to yield an OS benefit ^[8].

One common pathway, however, that has previously proved challenging to target directly has been mutations involving Kirsten rat sarcoma virus (KRAS). Previously considered impossible to target effectively, recent advances have led to the development of small-molecule-targeted inhibitors of this oncogene with significant implications for the future of NSCLC treatment in this space ^[9].

2. Kirsten Rat Sarcoma Virus (KRAS)

KRAS is a commonly encountered mutated oncogene in human cancers [11]^[10]. Through production of the K-Ras protein (a guanine triphosphatase), it regulates cell signalling via the RAS/mitogen activated protein kinase (MAPK) pathway by cyclically binding to guanosine triphosphate (GTP) [11,12]^[10][11]. This represents KRAS’s active state, which acts as a cellular “on switch”, affecting downstream pathways involved in cellular growth and differentiation, before converting the GTP to guanosine diphosphate (GDP) and becoming inactive once again [11,12]^[10][11]. Mutations of KRAS are expressed in multiple tumour types, most notably in NSCLC at a rate of approximately 25–30% [13]^[12]. In its mutated form, KRAS remains GTP-bound and, as such, active due to decreased GTP hydrolysis, with increased activity resulting in tumour growth. In lifelong non-smokers, over half (56%) have the KRAS G12V mutation, where KRAS is kept in an activated GTP-bound state [13]^[12]. In smokers or ex-smokers, KRAS mutations typically occur on codon 12, with a substitution of glycine for cysteine at this location (G12C) being the most frequent, found in 42% of patients [13]^[12]. Among the KRAS mutant cancers, there is evidence to support that the G12C mutation confers a worse prognostic outcome. A multicentre retrospective review from Japan published in 2021 compared survival in patients with KRAS G12C mutant metastatic colorectal cancers and those with other KRAS mutations. It showed a significantly worse progression-free survival (PFS) and overall survival (OS) in the G12C cohort compared to those of other mutations (median PFS 9.4 versus 10.8 months, p = 0.015; median OS 21.1 versus 27.3 months, p = 0.015) [14]^[13]. In lung cancers, the prognostic significance of KRAS mutations has been the source of much debate, with some studies suggesting a shorter OS in KRAS mutant lung cancers but others suggesting the contrary. The 2015 TAILOR study demonstrated a significantly worse median overall survival between KRAS mutant and KRAS wildtype patients with advanced NSCLC who had previously been treated with platinum-based chemotherapy [15]^[14]. As such, the presence of a KRAS mutation confers an overall negative prognostic outcome for NSCLC patients and represented an area of unmet need in targeted therapy for many decades. It is becoming increasingly apparent that complex KRAS downstream interactions and co-mutations also influence tumour signalling in lung cancer. There are three co-mutations of particular interest in the recent literature: TP53, Kelch-like ECH-associating protein 1 (KEAP1), and STK11/NRF2, of which the latter two are considered equivalent mutations. KRASG12C has also been associated with ERBB2 and ERBB4 mutations [16]^[15]. Co-mutation with TP53 occurs in 38–42% of patients, with KEAP1 in 8.1–27%, and with STK11 11.8–29% of patients [16^{[15][16][17][18]},17,18,19], with some variability as to the frequency of KEAP1 and STK11. TP53 co-mutations were independently associated with high PD-L1 expression (odds ratio (OR), 6.36; 95% confidence interval (CI), 1.84–22.02; p = 0.004) [18]^[17]. STK11/NRF2 protects cells against carcinogens and oxidants. STK11/NRF2 is a transcription factor that mediates the induction of phase-2 detoxification enzymes. Kelch-like ECH-associating protein 1 (KEAP1) binds to NRF2 and represses NRF2 transcriptional activity [20]^[19]. In one study, KEAP1 and STK11 had no association with PD-L1 status, while another suggested it may be associated with a low PD-L1 expression but a high TMB score [18]^[17]. KRAS co-mutations have been studied to determine their effects on prognosis in lung cancer patients. It is consistent that TP53, while repeatedly associated with a poorer prognosis in other tumour streams, has no association with decreased overall survival in KRAS mutant lung cancer [17,18]^[16][17]. STK11 and KEAP1 have been associated with a poorer prognosis in KRAS mutant lung cancer [17,18]^[16][17]. Both STK11 and KEAP1 have been associated with shorter overall survival under platinum-based chemotherapy [18]^[17]. In Arbour et al.’s study using multivariate analysis only, KEAP1 appeared to effect prognosis in KRAS mutant lung cancer (hazard ratio (HR), 1.96; 95% CI, 1.33–2.92; p < 0.001), and the authors argued that the high proportion in concurrent KEAP1 and STK11 mutations may have skewed the univariate analysis results for STK11. In this same review, KEAP1 was associated with a decreased overall survival under both chemotherapy and immunotherapy, while there was no effect from TP53 or STK11 mutations [17]^[16]. STK11 and KEAP1 mutations do not seem to influence the prognosis in KRAS wildtype lung cancers [19]^[18].

3. KRAS Detection

As new treatment options for patients with KRAS-mutated lung cancers become available, the need for improved methods of testing for KRAS mutations becomes more important [21]^[20]. Previously, KRAS testing was performed on tumour tissue, where DNA is extracted from formalin-fixed paraffin-embedded (FFPE) tissue blocks, followed by a variety of polymerase chain reaction (PCR)-based testing methods including sanger sequencing and pyrosequencing. Sanger sequencing of PCR-amplified DNA was previously considered the gold-standard technique. One criticism of this method is its relatively modest limit of detection. Multiple studies have demonstrated that the Sanger sequencing method requires a minimum of 15% to 50% of the sample DNA to contain the KRAS mutation before reliable detection is achieved [22]^[21]. As sequencing technologies can lack sufficient sensitivity, efforts have been made to look into alternative methods of KRAS testing. Whilst Sanger sequencing and next-generation sequencing (NGS) share some key principles, NGS provides a much higher sequencing volume through its ability to process millions of reactions in parallel. This results in a high-throughput, high-sensitivity method of testing, allowing for testing with more rapid turnarounds at a reduced cost, with many genome sequencing projects which would have taken many years with Sanger sequencing completed within hours using NGS [23]^[22]. Despite this, NGS-based assays for genomic alterations in tumours can still take up to 12–15 days [24]^[23], a suboptimal timeframe for the treatment of NSCLC. Moreover, due to the complexity of sample processing for NGS, bottlenecks can occur in the management, analysis, and storage of datasets [23]^[22]. Recently, the Idylla oncology assay by Biocartis was launched to complement NGS testing. Currently, the Idylla platform enables the detection of KRAS hotspot mutations, along with BRAF, EGFR, and NRAS mutations with rapid turnaround times of less than 3 h. Unlike NGS-based assays, the Idylla system uses formalin-fixed paraffin-embedded (FFPE) tissue samples without the need for DNA extraction with a fully automated interpretation of the results [24]^[23]. Moreover, several studies have demonstrated and confirmed the improved validity, accuracy, and concordance of the Idylla system in comparison to those of NGS in detecting KRAS, EGFR, BRAF, and NRAS hotspot mutations.