Detection of MET exon 14 skipping mutations in non-small cell lung cancer: overview and community perspective

ABSTRACT

Introduction: Non-small cell lung cancer (NSCLC), which accounts for the majority of lung cancer diagnoses in the United States, has many known driver mutations, including MET exon 14 skipping mutation (METex14). The detection of oncogenic driver mutations in NSCLC and the development of drugs to target these alterations, including METex14, has created the need for accurate and reliable testing, of which next-generation sequencing (NGS) is the gold standard. However, detection of METex14 in patients with NSCLC can be challenging due to the complex biology of METex14 and the abilities of different NGS platforms to detect METex14.

Areas covered: This review provides an overview of METex14 biology, discusses the optimal platforms for the detection of METex14 in NSCLC, and provides an overview of the use of NGS in the community setting.

Expert opinion: Broad molecular testing is crucial for identifying actionable oncogenic drivers in NSCLC. METex14 is a complex oncogenic driver mutation requiring carefully optimized platforms for proper detection. To identify patients eligible for targeted therapies – including therapies targeting novel oncogenic drivers, such as MET inhibitors – community oncologists need to be aware of both the use of NGS platforms and the differences in their capabilities to detect certain oncogenic drivers.

KEYWORDS:

• Capmatinib
• clinical decision-making
• crizotinib
• next-generation sequencing
• savolitinib
• tepotinib

1. Introduction

Non-small cell lung cancer (NSCLC) is a molecularly heterogeneous disease with many known driver mutations [1,2]. In recent years, many therapies have been approved for the treatment of NSCLC, which accounts for approximately 85% of lung cancer diagnoses in the United States [2,3]. Many of these treatments target oncogenic drivers, such as EGFR and ALK [2].

MET exon 14 skipping mutation (METex14) has emerged as an oncogenic driver for NSCLC, occurring in approximately 3% to 4% of NSCLC cases [1,2,4–9]. METex14 typically occurs in the absence of other driver mutations and is an adverse prognostic factor in patients with NSCLC [1,4,5,10]. A meta-analysis of patients with NSCLC found that among different histologies, METex14 occurred only in ~1% of squamous cell carcinoma, ~2% of adenocarcinoma, ~6% of adenosquamous cell carcinoma, and ~13% of pulmonary sarcomatoid carcinoma cases [6]. In particular, the prevalence of METex14 in pulmonary sarcomatoid carcinomas can be substantial, ranging between 4.9% and 31.8% [4,6,11,12]. Patients with METex14 NSCLC are significantly older than patients with NSCLC without METex14 as well as those with EGFR, KRAS mutations, or ALK rearrangements. These patients are also more likely to be female and less likely to have a smoking history than other patients with NSCLC [6].

Studies have shown that MET inhibitors prolong overall survival of patients with METex14 NSCLC; in contrast, immunotherapy agents have shown modest overall response rates in patients with METex14 NSCLC [13–15]. Capmatinib, a selective MET inhibitor, was approved in 2020 for the treatment of adult patients with metastatic NSCLC whose tumors have a mutation that leads to METex14, based on results from the GEOMETRY mono-1 trial (NCT02414139) [16,17]. Additionally, tepotinib, another MET inhibitor, was recently approved for the treatment of adult patients with metastatic NSCLC harboring MET exon 14 skipping alterations, based on results from the VISION trial (NCT02864992) [18,19]. Other targeted therapies, such as crizotinib and savolitinib, are being investigated for the treatment of METex14 NSCLC in clinical trials (Table 1).

Table 1. Active studies with MET-targeted therapies in advanced or metastatic METex14 NSCLC: summary of ongoing trials [17–22,44,77]

	GEOMETRY mono-1	PROFILE 1001	NCT02897479	VISION
MET inhibitor	Capmatinib	Crizotinib	Savolitinib	Tepotinib
Phase and trial design	Phase II, single arm, multicohort	Phase I, single arm, multicohort	Phase II, single arm	Phase II, single arm, multicohort
Number of patients	Cohort 4: 69 Cohort 5b: 28 Cohort 6: 31 Cohort 7: 23 Safety set: 364	69 (METex14)	70	Efficacy set: 152 Safety set: 255
MET testing modality	Central lab testing by tissue-based RT-PCR	Local testing, predominantly with tissue-based samples	Local testing confirmed by central testing	Central lab testing by liquid biopsy (DNA based) or tissue biopsy (RNA based)
Prior therapies allowed	Cohorts 5b and 7: no prior systemic therapy for advanced/metastatic disease Cohort 6: 1 line of systemic therapy for advanced/metastatic disease; prior IO allowed Cohort 4: 1 or 2 lines of systemic therapy for advanced/metastatic disease; prior IO allowed No prior MET-targeted therapy	No prior MET-targeted therapy Prior IO allowed	No prior MET inhibitor therapy Prior systemic therapies allowed	No prior MET-targeted therapy Prior IO allowed
Primary endpoint	ORR^a	ORR	ORR^a	ORR^a
Secondary endpoints	DOR,^a,b TTR,^a,b PFS, OS, disease control, safety, pharmacokinetics	DOR, TTR, PFS, OS, safety	DCR, DOR, TTR, PFS, 6‑month PFS rate, OS, safety, tolerability	ORR,^b DOR, PFS, OS, objective disease control, safety, HRQOL
References	Wolf J 2020 [17], Wolf J 2019 [21]	Drilon A 2020 [20]	Lu S 2020 [44]	Tepmetko [prescribing information] [18], Paik PK 2020 [19], Paik PK 2019 [77], Paik PK 2021 [22]

^aBlinded independent review committee. ^bInvestigator.

DCR, disease control rate; DOR, duration of response; HRQOL, health-related quality of life; IO, immuno-oncology; MET, mesenchymal–epithelial transition; METex14, MET exon 14 skipping mutation; NSCLC, non-small cell lung cancer; ORR, overall/objective response rate; OS, overall survival; PFS, progression-free survival; RT-PCR, reverse transcriptase–polymerase chain reaction; TTR, time to response.

It is important to note that MET amplification, which is distinct from METex14, is an additional MET alteration that has been identified as an oncogenic driver associated with poor prognosis [11,23]. Unlike the genetic alterations leading to METex14, MET amplifications are caused by an increased copy number of the MET gene [1,24].

With the availability of targeted therapies, the field of oncology has evolved rapidly from the use of single-gene tests to broad molecular testing, enabling clinicians to identify actionable markers and high-risk patients in a cost-efficient manner [25,26]. However, studies have shown that there are still barriers to adoption of NGS testing and utilization of broad molecular testing results in guiding treatment decisions in the community setting. Accurate and reliable tests are needed for identification of NSCLC patients with METex14, who may benefit from these emerging therapies. Due to the heterogeneous nature of NSCLC, next-generation sequencing (NGS) platforms are a preferred method for screening NSCLC, as NGS enables the simultaneous detection of multiple actionable biomarkers [1,2,27]. Here we will review the biology of METex14, discuss the abilities of available NGS platforms to detect METex14, and highlight the use of and the need for NGS testing in the community setting.

2. METex14 biology

MET is a proto-oncogene that encodes the c-Met tyrosine kinase protein [28,29]. The MET signaling pathway plays key roles in embryogenesis, wound healing, and organ regeneration; overactivation of this pathway has been observed in many types of cancer [28,29]. Several different types of alterations can occur within the MET gene that can result in exon 14 loss [1,30]. Exon 14 encodes a juxtamembrane domain, a key regulatory region of the MET receptor; loss of this region results in increased oncogenic signaling [1,28,30]. Based on previous research, it is believed that without the juxtamembrane domain, MET receptors have increased stability and tyrosine kinase activity, which can lead to cell proliferation and tumor growth (Figure 1A) [31–35]. METex14 occurs most commonly in lung cancers, and the mutations are highly diverse [1,10,34,36,37]. Alterations that lead to METex14 include base substitutions and insertions or deletions (indels) at the splice acceptor site, base substitutions and indels at the splice donor site, base substitutions and indels in intronic noncoding regions immediately adjacent to the splice acceptor site, and whole-exon deletion [10,36]. In a recent study of 1387 patients with NSCLC, more than 500 distinct genetic alterations leading to METex14 were identified [10]. These alterations leading to METex14 were found at the splice donor site (42%), the splice acceptor site (~5%), the polypyrimidine tract (15%), and both the splice acceptor site and the polypyrimidine tract (13%) [10]. All of these alterations result in the fusion of exon 13 and exon 15 in the messenger RNA transcript (Figure 1B). The loss of exon 14 due to these underlying genetic alterations may lead to increased MET stability and promote oncogenesis [1,6,8,31,34].

Figure 1. Biology of MET exon 14 skipping mutation (METex14). (A) Left, Exon 14 of the MET gene encodes the juxtamembrane domain of the MET receptor. Right, METex14 eliminates this key regulatory region, which could lead to increased stability and increased kinase activity of the MET receptor and, ultimately, upregulated signaling of the MET pathway [34]. (B) Alterations leading to METex14 result in a messenger RNA in which exons 13 and 15 are fused [10,34,36,45]. Arrows indicate the primary regions of interest for sequencing to detect METex14. HGF, hepatocyte growth factor; MET, mesenchymal–epithelial transition; mRNA, messenger RNA; P, phosphorylation; Ub, ubiquitin. Cortot AB et al, Exon 14 Deleted MET Receptor as a New Biomarker and Target in Cancers, Journal of the National Cancer Institute, 2017, 109, 5, djw262, adapted by permission of Oxford University Press

Some studies have also identified an association between METex14 and MET amplification, but data are inconclusive. These studies note significant associations between METex14 and MET amplification, with a co-occurrence rate between 0% and 40.5% [4,8,10,17,24,31,37–40]. However, amplification is difficult to discern across studies because there is no standard definition; the use of a low threshold for defining MET amplification will produce overlap with a large number of known dominant oncogenes [4,9,24,38,41].

In addition, METex14 is not correlated with MET overexpression, whereas mixed results surround the correlation between MET amplification and MET overexpression [7,11,24]. Quantification of MET protein levels by immunohistochemistry has shown that increases in MET levels may be present even in the absence of METex14 or MET amplifications [24,31,42]. Most cases of MET protein overexpression are driven by other oncogenic drivers that promote tumor growth and progression [4].

3. Detecting METex14 using NGS-based approaches

METex14 has been associated with poor prognosis in patients with NSCLC (Table 2) [4,5,13,14,43]. To improve patient outcomes, METex14 needs to be detected accurately and reliably, given the approval of capmatinib and pending approvals of additional METex14-targeted therapies [16,19,44]. Although multiple NGS platforms are available for screening mutations, not all are optimized for detecting METex14 [27,45,46]. One factor to consider prior to choosing a NGS platform is whether mutations, indels, gene fusions, or gene copy number alterations will be assessed [47]. For the detection of novel gene alterations, such as METex14, amplicon-based and hybrid capture–based platforms are 2 commonly used NGS approaches [47], but the detection rates may vary significantly between platforms [27,45,46].

Table 2. Association of METex14 with poor overall survival [4,5,13,14,43]

	METex14 without MET inhibitor	METex14 with MET inhibitor	Disease stage	Publication
Median OS (95% CI), months [n]	8.1 (5.3–NR) [32]^a	24.6 (12.1–NR) [25]^a	IV	Awad MM 2019 [13]
	10.9 (7.4–16.9) [50]^a	25.3 (18.8–40.9) [34]^a	IV	Wolf J 2018 [14]
	14.8 (9.0–NR)[29.4]^b,c	20.8 (12.6–NR) [26]^c	IIIB/IIIC/IV	Wolf J 2020 [43]
OS hazard ratio (95% CI)^d	1.993 (1.023–3.882)		I/II/III/IV	Tong JH 2016 [4]
	2.93 (1.05–8.21)		I/II/III/IV	Yeung SF 2015 [5]

^aOS was defined as time between stage IV diagnosis and death.

^bEach patient was assigned a weight proportional to the probability of being in the GEOMETRY mono-1 group, resulting in an altered composition of the weighted real-world group; the same weights were applied for the OS analysis. OS in the real-world group was adjusted for left truncation.

^cOS was defined from start of treatment to censoring or death; real-world patients were considered at risk from the date of meeting the inclusion criteria.

^dThe Kaplan-Meier method was used to estimate the survival rates between wild type and METex14; the Cox proportional hazards model was employed for univariate survival analyses.

MET, mesenchymal–epithelial transition; METex14, MET exon 14 skipping mutation; NR, not reached; OS, overall survival.

Amplicon-based methods rely on the enrichment of target sequences by multiplex polymerase chain reaction (PCR) amplification (Figure 2A) [47]. In 2017, Poirot et al [27] published an in silico assessment of the METex14 detection rate of commercial amplicon-based cancer panels and found that no panel detected more than 63% of METex14; only 2 of the 7 panels assessed achieved this detection rate. Of the 7 panels, 4 were predicted to have METex14 detection rates of <25%. Analysis of the commercial amplicon-based panels also revealed that different panels were predicted to detect different types of METex14 events. Only 3 of 7 panels were predicted to find splice donor-site and/or Y1003 mutations at a rate >99%, whereas 2 other panels were predicted to detect acceptor-site or branch-site mutations only at a rate as high as 65% [27]. Another study found that DNA-based amplicon assays performed between 2013 and 2018 had a METex14 detection rate of 1.3%, whereas RNA-based amplicon assays performed between 2016 and 2018 had a detection rate of 4.2%. Among samples tested by both amplicon-based assays, 60% that were positive according to the RNA-based assay were negative according to the DNA-based amplicon assay [45].

Amplicon-based approaches may fail to identify METex14 based on the way they are constructed. The diversity of alterations leading to METex14 can lead to allele dropout and produce false negatives if there is a single-nucleotide polymorphism or short indel in the primer region, because the primer will be mismatched and will not bind [27,47]. Indels that substantially affect amplicon size may also evade detection [47]. Large deletions, such as those over 40 base pairs in length, can go undetected with amplicon-based methods [48]. Finally, poor sequencing quality at the ends of amplicons may lead to miscalling of variants [47]. However, with careful optimization, such as improved primer coverage of exon 14 and fragment-length analysis, detection rates with amplicon-based methods can be improved [46,48,49].

Figure 2. Overview of the amplicon-based and hybrid capture–based NGS methods. (a) Amplicon-based NGS methods utilize PCR primers that are designed to amplify targeted regions of interest in the genome [45,47]. Alterations (indels or base substitutions) that affect primer binding within the target region or are located outside of the target region (such as within intronic regions flanking MET exon 14) may go undetected. (b) Hybrid capture–based NGS methods use oligonucleotide probes to isolate larger regions of interest in the tumor genome, including regions flanking the targeted area [47]. Allele dropout is less likely to occur, as long probes used for hybridization can tolerate mutations within the genomic DNA. Furthermore, DNA sequences outside of the target region (such as intronic regions flanking MET exon 14) can be captured and sequenced. NGS, next-generation sequencing; PCR, polymerase chain reaction. Adapted with permission from Elsevier: Journal of Molecular Diagnostics, 19/3, Jennings LJ et al, Guidelines for Validation of Next-Generation Sequencing–Based Oncology Panels: A Joint Consensus Recommendation of the Association for Molecular Pathology and College of American Pathologists, 341–365, copyright 2017

In contrast to amplicon-based platforms, hybrid capture–based platforms are constructed using long biotinylated oligonucleotides that hybridize to target regions in the genome and enable sequencing of flanking regions (Figure 2B) [47]. These probes are significantly longer than the PCR primers used in amplicon-based assays and can therefore tolerate the presence of mismatches in the binding site without interfering with target hybridization; this bypasses the issue of allele dropout that arises in amplicon-based methods [47,50]. The hybrid capture method generally isolates large fragments of DNA, including regions that flank the target of interest [47]. Given the diversity of METex14 events and the potential for these alterations to occur in regions flanking exon 14, the ability to probe larger regions of interest makes the hybrid capture approach more amenable to detecting the alterations leading to METex14 [36,47]. Additionally, because of the favorable properties of hybrid capture–based assays, this methodology has been used to capture other alterations leading to oncogenic gene fusion events [51,52]. However, it should be noted that hybrid capture–based platforms may not be suitable for small tissue samples, because they require relatively large amounts of DNA, ranging from 50 to 250 ng, whereas amplicon-based platforms require about 10 ng of DNA input [53–55].

4. RNA-based versus DNA-based approaches for testing METex14

Both RNA-based and DNA-based approaches have been used to detect METex14 in NSCLC [1,27,39,45,46,56]. DNA approaches detect mutations that are predicted to lead to exon 14 skipping, whereas RNA approaches detect the direct fusion of exon 13 and exon 15 transcripts [45]. Studies have found that RNA-based methods detect METex14 events at a higher rate than DNA-based approaches [45,46,56]. In a comparison of single-gene testing methods, the mRNA-based quantitative reverse transcriptase (RT)-PCR demonstrated 100% sensitivity in detecting METex14, compared with 61.5% sensitivity using conventional DNA-based Sanger sequencing [56].

Two studies found that RNA-based NGS methods outperformed DNA-based NGS methods. In a study by Davies et al [45], METex14 was detected at a rate of 1.3% using a DNA amplicon–based NGS assay, compared with 4.2% using an RNA-based NGS approach (P < 0.01). Similarly, Jurkiewicz et al [46] reported a METex14 detection rate of 2.5% (16/644 lung cancer tumors) when a DNA amplicon–based NGS panel was used, compared with a detection rate of 3.9% (25/644 lung cancer tumors) using a NGS-based RNA fusion panel.

The higher accuracy of RNA-based assays compared with that of DNA-based assays can be attributed to the easily detectable direct fusion events between exon 13 and exon 15 found in RNA transcripts [39]. In contrast, DNA-based assays need to be able to detect many potential point mutations or deletions leading to METex14 that occur not only within the MET gene but also in neighboring intronic regions [39,45]. Variants detected in the RNA-based panel were likely to include alterations in the intron 13 splice acceptor site, or other sites relevant to splicing but not covered by the DNA panel, whereas METex14 detected by the DNA-based NGS panel included only alterations at or around the intron 14 splice donor site, which suggests that the DNA panel had insufficient coverage [46]. DNA-based testing results should be verified by RNA-based testing unless the alteration identified by DNA-based testing has been previously validated [50,57]. Although high levels of accuracy have been observed with RNA-based approaches, it is important to note that one limitation is the ability to obtain enough high-quality RNA for NGS assays, which can be challenging because RNA is prone to degradation.

Notably, in some cases, DNA-based NGS methods can match the sensitivity of RNA-based assays depending on the platform used. A recent retrospective analysis of NSCLC patient samples from the GEOMETRY mono-1 study found that a hybrid capture NGS approach identified 99% of METex14 events previously detected by using a single-gene quantitative RT-PCR approach [39]. METex14 in 72 of 73 patients with NSCLC were successfully detected, including 41 with canonical METex14 events, 20 with indels in the intron preceding exon 14, 10 with base substitutions in the intron following the 3′ end of exon 14, 11 with indels in the intron following the 3′ end of exon 14, and 1 with a rearrangement involving MET exon 14. Thus, this study demonstrated that a wide range of METex14 events occurring within and surrounding exon 14 can be detected using a DNA-based hybrid capture NGS approach [39].

5. Optimization of METex14 testing in the community setting

METex14 is estimated to occur in ~3 to 4% of patients with NSCLC [1,4–8]. If METex14 events are detected at a lower rate within a practice or institution, the testing methodology may need to be assessed for the ability to detect METex14 and for whether it has been optimized for detecting METex14. Due to the complexity of METex14, certain types of alterations in patients with NSCLC may elude detection on various NGS platforms, particularly amplicon-based NGS platforms, as demonstrated by multiple studies [27,45–47].

When using an amplicon-based NGS assay to detect METex14, optimization may be required. For example, it has been shown that panel primers can increase intronic coverage by targeting both the 3′ splice site of intron 13 and the 5′ splice site of intron 14 to ensure detection of a wide range of METex14 events [46]. One group improved their METex14 detection rates from 0.3% to 2.2% after optimizing their amplicon-based NGS assays to improve coverage of METex14 in combination with fragment-length analysis [49]. The fragment analysis–based approach paired with amplicon-based NGS has been further validated by Loyaux and colleagues [48] as a method to detect METex14.

6. Use of NGS testing in the community setting

To assess the use of broad molecular profiling in the community setting, a 3-year study of a large community-based oncology network was conducted from 2013 to 2015 [58]. Despite the fact that both the contemporary National Comprehensive Cancer Network (NCCN) guidelines and the College of American Pathologists/International Association for the Study of Lung Cancer/Association for Molecular Pathology (CAP/IASLC/AMP) guidelines recommended routine testing for EGFR and ALK alterations [59,60], only 59% of patients were tested for EGFR and ALK alterations in this patient population [58]. Contemporary NCCN guidelines also recommended the testing of 5 additional biomarkers [59]; however, only 8% of patients were tested for all 7 NCCN-recommended biomarkers in this analysis [58]. It should be noted that in a recent German study, the average biomarker testing rate before start of first-line treatment was 92.2% in 2019 for patients with nonsquamous NSCLC [61].

Another retrospective review assessed the use of targeted therapy among patients with NSCLC who underwent broad molecular testing between January 2011 and January 2018 [62]; 85% of patients who underwent initial screening (N = 28,998) were treated in the community setting. Of the 1260 patients with an advanced NSCLC diagnosis who had an NCCN-listed driver alteration (EGFR, ALK, ROS1, MET, BRAF, RET, or ERBB2), only 48.3% (n = 609) received NCCN-recommended therapy after the advanced diagnosis. Importantly, the study showed that receiving a targeted therapy (n = 575) was significantly associated with longer overall survival compared with not receiving a targeted therapy (n = 560) for patients with a mutation in an NCCN-listed gene (median overall survival, 18.6 months [95% CI, 15.2–21.7] vs 11.4 months [95% CI, 9.7–12.5]; P < 0.001) [62].

Several factors may play a significant role in limiting the use of broad molecular testing. The prioritization of tissue specimens for histopathologic diagnosis and PD-L1 expression, which may leave little tissue remaining for broad molecular testing, could be one limiting factor [63]. In some cases, patients with advanced disease may have initiated nontargeted therapy prior to receiving NGS testing results due to the turnaround time needed [63].

Several studies have shown that physician attitudes and knowledge may also play a role in both the use of broad molecular testing for diagnosis and utilization of testing results in guiding treatment decisions [64–69]. A recent study found that community oncologists were unsure about or incorrectly matched the molecular alteration to the targeted therapy up to 69% of the time, including for some FDA-approved therapies [68]. Another survey of 100 practicing oncologists showed that 36% found NGS testing extremely important, whereas 51% found NGS testing only moderately important [70].

One of the largest surveys, conducted by the IASLC from May to December 2018, evaluated 2537 respondents from 102 countries on their perceptions of current molecular testing. It found that among respondents who request tests and treat patients, 33% were unaware of the most recent guidelines for molecular testing. When molecular tests were used, over 90% tested for EGFR mutation and ALK rearrangement, whereas less than 30% tested for METex14. Few providers were using multiplex assays or including all suggested biomarkers within multiplex tests [69]. Oncologists may not be choosing the ideal testing platform due to the number of commercial platforms available and lack of knowledge surrounding these platforms. No studies to date have investigated oncologists’ understanding of the capabilities and limitations of different testing platforms.

Physician attitudes may be shaped by training, experience in using broad molecular testing, and patient volume. In a study using data from the 2017 National Survey of Precision Medicine in Cancer Treatment assessing confidence levels regarding use of broad molecular testing, oncologists who had received prior genomics training reported higher confidence levels than those who did not [67]. Similarly, oncologists who had prior experience using genomic tests, including gene expression and NGS tests, reported the highest confidence levels, with 60.1% being very confident and 35.6% being moderately confident in using broad molecular testing to guide patient management [67].

Taken together, the results from these studies underscore the need for increased awareness of broad molecular profiling use in the community setting and education on patient outcomes for those who receive the appropriate targeted therapy. Considering that the NCCN guidelines for NSCLC treatment recommendations rely on testing for METex14 and other biomarkers [71], it is critical for oncologists to incorporate broad molecular testing into their clinical decision-making. To guide treatment decisions and improve patient outcomes, oncologists should have a working knowledge of the newly approved therapies and appropriate genomic testing and incorporate precision medicine into their practices.

7. Conclusions

Detection of METex14 in patients with NSCLC can be challenging due to the complex alterations leading to the loss of exon 14 and the abilities of different NGS platforms to detect these alterations. The use of broad molecular testing in patients with NSCLC is crucial for matching their tumors with the appropriate targeted therapy, especially as the list of actionable oncogenic drivers continues to grow. Although guidelines recommend routine testing of biomarkers for patients with NSCLC, not all physicians test their patients for actionable biomarkers. For patients to have the best outcome, increased awareness and education on broad molecular profiling and genetic testing is needed within the physician community.

8. Expert opinion

Broad molecular testing is recommended in clinical practice guidelines for newly diagnosed NSCLC [71,72]. The approval of capmatinib provides a rationale to routinely include METex14 in these panels [16]. METex14 is an established oncogenic driver in NSCLC, with a prevalence of about 3% to 4% in NSCLC cases [1,4–8].

Multiple studies have found that treatment with MET-targeted therapies is associated with better outcomes in patients with METex14 [13,14,43]. In a recent real-world analysis, comparing clinical outcomes of treatment-naive patients with METex14 advanced NSCLC who received first-line therapy with capmatinib in the GEOMETRY mono-1 study and those of matched patients who received other standard-of-care first-line therapies, patients who received first-line capmatinib had longer median progression-free survival than patients with first-line chemotherapy and/or immunotherapy (12.0 months vs 6.2 months) [43]. Another real-world analysis of patients with METex14 found that patients receiving a MET inhibitor had longer overall survival compared with patients who did not receive a MET inhibitor (25.3 months and 10.9 months, respectively) [14]. Finally, in a retrospective study, patients with METex14 treated with single-agent immunotherapy had only modest overall response rates, with a median progression-free survival of 1.9 months and median overall survival of 18.2 months [15].

Importantly, NGS panels used for broad molecular profiling are not equivalent in their ability to detect METex14. Choosing the correct test is critical because biopsies tend to have minimal tumor content and using a single sample for multiple tests may not be possible [73–75]. Many amplicon-based NGS panels are not properly optimized to detect METex14, and the detection rate may be as low as 0.3% [49]. One study found that a DNA-based assay had a METex14 detection rate of 1.3% [45]. Additionally, an in silico study comparing 7 different DNA amplicon-based assays revealed that the detection of known METex14 ranged from 3% to 63% [27]. The hybrid capture approach is a preferred method for detecting alterations leading to METex14 and other oncogenic gene fusions because the assay is able to isolate large fragments of DNA – including regions surrounding the target of interest – due to the probes being significantly longer than PCR primers, which avoids allele dropout [45,47,51,52]. However, this method also requires bioinformatic tools optimized to detect these events, which could be an additional obstacle for oncologists and laboratories looking to switch testing panels [47]. Another obstacle oncologists face when choosing the correct testing platform is the duration between the test and when the patient can begin an appropriate targeted treatment. A real-world analysis investigating testing patterns and clinical outcomes in patients with METex14 found that NGS results had a median time from specimen collection to reported results of 27 days (interquartile range, 27–54 days) [76]. The majority of patients initiated first-line therapy before receiving their NGS reports [76]. Given the moderate response METex14 patients have with immunotherapy treatment, it is critical for patients to receive the correct targeted therapy as quickly as possible. Finally, clinical oncologists may also be hesitant to run certain tests due to concerns related to costs associated with molecular profiling and insurance coverage [68].

Future research for patients with METex14 includes combination studies with MET inhibitors, such as capmatinib, and immunotherapy. Additional studies are currently being conducted on the use of MET inhibitors in patients with MET amplification [17,77,78]. Further research is needed to explore the role of treatment sequencing in patients with METex14 treated with first-line MET inhibitors. Physicians may seek to retest patients for new molecular alterations upon disease progression. Preclinical studies have shown that resistance mutations to one type of MET inhibitor may remain sensitive to another class of MET inhibitor [79,80]. However, these findings will require clinical confirmation.

In 5 years, as targeted therapies move into earlier disease settings [81–84], the practice norms should evolve to include routine broad molecular profiling of all NSCLCs, regardless of stage.

Article highlights

• Non-small cell lung cancer (NSCLC) is a heterogeneous disease with numerous histologies and many known oncogenic drivers

• The approval of MET tyrosine kinase inhibitors has made MET exon 14 skipping mutation (METex14) an actionable oncogenic driver in NSCLC

• METex14 occurs across all histologies of NSCLC and can occur through a diverse set of underlying genetic alterations in and around exon 14 in the MET gene

• Carefully optimized next-generation sequencing assays are required to properly detect the diverse alterations that lead to METex14

• In the community setting, guideline-recommended biomarkers are not tested consistently

• Surveys have found that community oncologists vary in confidence and understanding of different biomarker assays

• Increased awareness and understanding of broad molecular profiling is critical in the community

Declaration of interest

Medical writing assistance was utilised in the production of this manuscript, provided by Chameleon Communications and funded by Novartis Pharmaceuticals Corporation.

J Subramanian has received personal fees for advisory boards and speakers bureaus from AstraZeneca, Boehringer Ingelheim, and Eli Lilly; personal fees for advisory boards from Novartis, Pfizer, G1 Therapeutics, Jazz Pharmaceuticals, and Cardinal Health; and grants from Paradigm and Biocept.

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors thank Lisa Hang of Chameleon Communications International, New York, NY, for providing medical writing assistance, which was funded by Novartis Pharmaceuticals Corporation, East Hanover, NJ, in accordance with Good Publication Practice (GPP3) guidelines (http://www.ismpp.org/gpp3).

Additional information

Funding

This work was supported by Novartis Pharmaceuticals Corporation, East Hanover, NJ, in accordance with Good Publication Practice (GPP3) guidelines (http://www.ismpp.org/gpp3).