Mesenchymal Epithelial Transition (MET) T alterations are reported in several solid tumors and are considered driver mutations in the carcinogenic process ^[1][26]. Three different genomic states can lead to the deregulation of this pathway: gene amplification, mutations and fusions ^[2][25]. These alterations result in tyrosine kinase activation and ligand-independent downstream signaling.

MET amplification is found in 1–5% of untreated  non-small cell lung cancer (NSCLC) ^[3][4][27,28], and in 5–20% of EGFR mutated tumors with acquired resistance to EGFR-TKIs ^[5][29]. Gene amplification can be detected by FISH or NGS panels, however, to date there is no consensus on the actual definition of MET amplification. Studies have tested several cut-offs for FISH positivity: MET– to–chromosome 7 centromere ratio (MET/CEP7) values of 1.8 or higher, 2.0 or higher, 2.2 or higher and 5 or higher ^[6][7][30,31]. Literature data seem to suggest that a FISH MET/CEP7 ratio of 5 or higher could be the optimal cut-off for defining positivity, as a high-level MET amplification strictly correlates with oncogenic-dependance and there is no overlap with other oncogene drivers, and therefore treatment sensitivity ^[8][32].

Capmatinib is a highly potent and selective inhibitor of MET receptor that showed in vitro and in vivo activity in preclinical cancer models with diverse types of MET aberrations ^[9][33]. Its clinical efficacy and safety were analyzed in the prospective, international, multicohort, open-label, phase 2 trial, GEOMETRY mono-1. The trial included naïve or pretreated patients with stage IIIB and IV NSCLC with no EGFR mutation or ALK fusion, tested positive for MET amplification or MET exon 14 skipping mutation. MET amplified NSCLCs were classified according to gene copy number (GCN) in tumor tissue as follows: GCN less than 4, 4–5, 6–9 and 10 or higher. Notably, limited efficacy was registered in patients with a GCN less than 10, with an overall response rate (ORR) ranging from 7–12% and a median progression-free survival (PFS) from 2.7–3.6 months. Among tumors with a GCN of 10 or higher, an objective response was registered in 29% of pretreated, and in 40% of treatment naïve patients, although the overall response was lower than the prespecified threshold set for a clinically relevant activity ^[10][34].

Treatment options for advanced NSCLC with MET amplification also include the MET inhibitor, crizotinib. Efficacy data is derived from small cohorts of phase 1 and 2 trials. The phase 1 PROFILE 1001 trial enrolled 38 MET positive (with a MET/CEP7 ratio ≥ 1.8) NSCLC patients. Consistent with the findings of the GEOMETRY mono-1 trial, response rates were greater in patients with high MET amplification: ORR of 38% in MET/CEP7 ratio ≥4.0, 14.3% in MET/CEP7 ratio 2–4 and 33.3% in MET/CEP7 ratio >1.8, with a median PFS of 6.7, 1.9 and 1.8 months, respectively ^[11][35]. Similarly, the phase 2 METROS trial enrolled a cohort of 26 MET deregulated NSCLC patients, 16 of them with a MET amplification, 9 with a MET exon 14 (METex14) skipping mutation and 1 patient with cooccurrence of the two alterations. The ORR was 27% (all partial responses), stable disease was registered in 42% of cases, with a global disease control rate (DCR) of 69%. Although treatment response rates were promising for a pretreated population, survival outcomes were poor: at a follow up of 21 months, median PFS and OS were 4.4 and 5.4 months, respectively ^[12][36].

To date, neither capmatinib nor crizotinib have been approved by the FDA or EMA for pretreated NSCLC with high-level MET amplification.

Exon 14 encodes the 47-amino acid juxtamembrane domain of the MET receptor, a key regulatory region preventing MET overexpression and thus oversignalling ^[13][37]. The genomic events underlying the mis-splicing of MET exon 14 are complex and include several types of alterations, such as point mutations, insertions or deletions. The specific mechanism of carcinogenesis has not been fully elucidated yet, however, the loss of this region results in an impaired MET receptor degradation and aberrant activation of the signaling pathway ^[14][38].

In NSCLC, METex14 is observed in approximately 2–4% of cases ^[15][16][22,39] as tested by DNA or RNA NGS. Given the diversity of alterations that may lead to MET exon 14 skipping and the potential location of these alterations in the MET gene, the optimal testing technique is still matter of debate ^[17][23]. The phase 2 trial GEOMETRY mono-1, included 97 patients tested positive for a MET exon 14 skipping mutation. In this subset of patients, capmatinib showed substantial antitumor activity. An overall response was observed in 41% of patients who had received one or two prior lines of therapy and in 68% of treatment-naïve patients; the median duration of response (DoR) was 9.7 months and 12.6 months, respectively. The median PFS was 5.4 months in pretreated and 12.4 months in untreated patients ^[10][34]. Notably, responses to capmatinib were rapid, with the majority of patients showing response at the first radiological evaluation. No difference in response to the study drug was observed according to the specific genetic alteration causing METex14 skipping mutation.

Tepotinib activity was assessed in the multi-cohort, phase 2 VISION trial, conducted on advanced NSCLC patients with evidence of METex14 skipping mutation detected on tissue or liquid biopsy, who received up to two courses of previous therapy for metastatic disease. Globally, the response rate confirmed by independent central review (ICR) (primary study endpoint) was 46%, with a median duration of response of 11 months. The investigator-assessed response rate was 56%, similar to previous lines of therapy ^[18][40].

Following the results of the GEOMETRY mono-1 and VISION trials, the FDA approved capmatinib and tepotinib for adult patients with advanced NSCLC harboring a METex14 skipping mutation on 6 May 2020, and 13 February 2021, respectively ^[19][41]. On 16 February 2022, tepotinib received marketing authorization valid throughout the European Union ^[20][42]. On 22 April 2022, the Committee for Medicinal Products for Human Use (CHMP) adopted a positive opinion, recommending the granting of capmatinib authorization, however European Medical Agency (EMA) final approval is still awaited ^[21][43].

The safety profile of capmatinib and tepotinib was similar, and the most common adverse events (AEs) recurring in ≥10% of patients reported in clinical trials included peripheral edema, nausea, vomiting and increased blood creatinine level ^[10][18][34,40].

A better understanding of MET deregulation mechanisms contributed in recent years to the development of additional strategies: highly selective oral/intravenous MET inhibitors, combination therapies, humanized antibodies and antibody drug conjugates (ADC). Savolitinib is an oral selective MET inhibitor that showed promising results in a phase 2 trial on METex14 mutated lung sarcomatoid carcinomas with an ORR of 49% by ICR ^[22][44] and is currently under evaluation in combination with osimertinib in patients affected by MET-altered NSCLC with acquired resistance to EGFR-TKIs (NCT05015608, NCT05163249). Other molecules for MET-altered NSCLC are under investigation in early phase clinical trials (Table 12) including: glumetinib and APL-101 (oral MET inhibitors); SAR125844, a MET inhibitor administered intravenously; glesatinib (MGCD265) with activity against MET; VEGFR1/2/3; RON; TIE-2; and elzovantinib (TPX-022), a potent MET/CSF1R/SRC inhibitor. Current research is also directed to test the efficacy of biopharmaceutical drugs, such as Sym015, a mixture of two humanized IgG1 antibodies, directed against nonoverlapping epitopes of the MET ectodomain ^[23][45] and ADC telisotuzumab vedotin ^[24][46].

Rearranged during transfection (RET) gene encodes for a transmembrane receptor of the tyrosine protein kinase family that binds the glial cell-line derived neurotrophic factor ^[25][47]. In human cancer, oncogenic activation is mainly a consequence of cytogenetic rearrangements, and the derived chimeric genes determine the transcription of an aberrant receptor with dysregulation of downstream processes, such as cell proliferation, migration and survival ^[25][47]. RET fusion positive NSCLCs are an extremely rare subgroup of patients (1–2%) first described in 2012 ^[26][27][5,21]. Clinicians must be aware of the importance of the adoption of adequate molecular testing techniques and the use of NGS should be preferred over FISH or real-time PCR due to its higher sensitivity ^[28][48].

First attempts to target RET fusions included the use of several oral multikinase inhibitors with proven clinical efficacy in other solid tumors (e.g., cabozantinib, vandetanib and lenvatinib). The retrospective international registry GLORY represents the largest database of RET-rearranged lung cancers and data were collected with the aim to document outcomes of patients treated with these molecules. Globally, response rates reported ranged from 18–37%, with a median PFS of 2.3 months and a median OS of 6.8 months ^[29][49]. Outcomes are disappointing compared with the activity of targeted therapy in other genomic subsets of lung cancer, and, as consequence of the off-target side-effects, treatment is burdened by excessive toxicities.

The development of RET selective inhibitors represented an effective strategy to potentially overcome the poor results obtained with multikinase inhibitors while also reducing treatment-related adverse events (AEs). In this setting, selpercatinib is a novel agent that selectively binds to and targets various RET mutants and RET-containing fusion products. The phase 1/2 trial LIBRETTO-001, designed to test the safety and activity of selpercatinib, enrolled 105 patients with RET fusion-positive advanced NSCLC who had previously received at least a platinum-based chemotherapy. In 36% of cases, brain metastasis were present at baseline. Moreover, the study population was heavily pretreated with a median of three previous systemic lines of therapy and almost half of patients already exposed to a multitargeted kinase inhibitors with anti-RET activity. The study met its primary endpoint registering a 64% of ORR confirmed by ICR, mostly partial responses ^[30][50]. Among patients with measurable brain involvement, the objective intracranial response by ICR was 91%. Selpercatinib was effective regardless of previous therapy or specific RET fusion partner.

The ARROW trial is a multi-cohort, international, open-label, phase 1/2 study designed to define the maximum tolerated and recommended dose of the oral selective RET inhibitor, pralsetinib, and to test its clinical activity and safety. Overall, 92 patients treated with a median of two previous lines and 29 treatment naïve patients who were not candidates for standard platinum-therapies, were included in the phase 2 study. Notably, in 41% of cases, baseline central nervous system (CNS) involvement was documented. Response rates were remarkable in both the pretreated and the treatment naïve group, with an ORR of 61% and 70%, respectively. Shrinkage of intracranial metastases was seen in all patients with measurable intracranial metastases at baseline. At a median follow up of 14.7 months, in previously treated patients median PFS was 17.1 months and median OS not reached. At a median follow-up of 11.6 months, in untreated patients median PFS and median OS were 9.1 and not reached, respectively ^[31][51]. As seen for selpercatinib, pralsetinib activity was not affected by previous treatments received, including anti-PD1, anti-PDL1 or multikinase inhibitors, or by RET diverse fusion partners.

Overall, 93% of patients had treatment-related AEs, most common G ≥ 3 AEs were neutropenia (18%), anemia (10%), hypertension (11%) and pneumonia (10%). Most common G1-2 AEs (reported in ≥10% of patients) included hematological toxicity (neutropenia, anemia, leucopenia), AST/ALT increase, asthenia, constipation, hypertension, dysgeusia and increased blood creatinine ^[31][51]. Despite the limitation of cross-trial comparisons, the overall frequency of adverse events with pralsetinib was comparable to selpercatinib ^[30][50].

Following the results of LIBRETTO-001 and ARROW trials, the FDA granted approval of pralsetinib and selpercatinib for the treatment of RET fusion-positive NSCLC ^[32][33][52,53]. Selpercatinib and pralsetinib received a conditional marketing authorization from the EMA in February and November 2021, respectively ^[34][35][54,55]. Therefore, in the European Union these two targeted agents are still under additional monitoring.

Currently, the next generation of RET-TKIs is under exploration in early phase clinical trials (Table 23). Of particular interest is TPX-0046, a third generation orally bioavailable RET/SRC kinase inhibitor, with preliminary evidence of activity against a range of RET fusions and resistance mutations in tumors models ^[36][56]. BOS172738 is an investigational, potent, next generation selective oral RET kinase inhibitor with reported clinical activity and a good safety profile in a phase 1 trial ^[37][57]. As resistance is a major challenge for RET fusion-positive NSCLC ^[38][58], the development of next generation RET inhibitors with activity against acquired mutations could represent an effective treatment option after progression.

The Neurotrophic Tropomyosin Receptor Kinase (NTRK1, NTRK2, NTRK3) gene family encodes Tropomyosin Receptor Kinases (TRKA, TRKB, TRKC, respectively) ^{[39][40][41][42]}[59,60,61,62].

Physiologically expressed in neuronal cells, these three transmembrane proteins, binding neurotrophic factors, are fundamental to the development and function of the nervous system. Ligand binding causes the oligomerization of these receptors’ kinases, leading to final activation of intracytoplasmic pathways (MAPK, PI3K and PLC-γ) involved in cell proliferation, differentiation and survival ^[41][42][61,62]. Alterations in TRK pathways are involved both in nervous system diseases (depression, epilepsy, or neuropathic pain, etc.) ^[43][63] and cancers ^{[39][40][41][42]}[59,60,61,62]. The main oncogenic gene alteration in cancer is NTRK gene fusion, producing overexpressed or constitutively activated fusion receptor kinases ^[44][45][64,65]. Alternative oncogenic mechanisms include TrkA alternative splicing, implicated in neuroblastoma, and in-frame deletion of NTRK1, related to acute myeloid leukemia ^[41][61]. NTRK fusions represent a rare therapeutic target in solid neoplasms, in NSCLC their frequency is reported from 0.1% up to 1% ^[46][47][19,20] and generally are mutually exclusive with other oncogene alterations. NTRK fusions have also been described as a mechanism of acquired resistance to EGFR TKIs in patients with EGFR mutated NSCLC ^[48][66].

NTRK1/2/3 fusion gene detection is independent of tumor type and follows the European Society for Medical Oncology (ESMO) recommendations ^[49][67]. Molecular testing includes two alternative methods: screening with immunohistochemistry (IHC), followed by next generation sequency (NGS), if possible, RNA-based NGS; or NGS techniques confirmed by IHC in positive cases ^[49][67].

Several targeted drugs for NRTK rearrangements are under current development and some of these have been introduced in clinical practice thanks to different basket trials. Among TRK inhibitors, multi-kinase inhibitors also present anti-TRK activity (entrectinib, repotrectinib, selirectinib, taletrectinib, etc.), while larotrectinib is a member of selective inhibitors ^[42][49][62,67].

Larotrectinib is characterized by high selectivity for TRKA, TRKB and TRKC and it is the first oral pan-TRK inhibitor to receive tissue-agnostic FDA approval (November 2018) for advanced NTRK fusion-positive solid tumors, based on the results on 55 patients of three multicenter, open-label, single-arm clinical trials (LOXO-TRK-14001, SCOUT and NAVIGATE trials) ^[50][68]. Larotrectinib also received a conditional marketing authorization by the EMA in September 2019 with the same therapeutic indications ^[51][69].

In April 2020, Hong and colleagues ^[52][70] presented updated results of these three trials: among 159 patients, 12 patients had NSCLC with an ORR of 75%, consistent with ORR in the overall population (79%). Considering survival in all patients, a median OS of 44.4 months and a median PFS of 28.3 months were reported. In the safety analysis, larotrectinib was well tolerated with grade 4 adverse events in 1% and grade 3 in 13% of patients (grade 3–4: neutropenia in 2%, anemia in 2% and elevation of aspartate aminotransferase, AST, or alanine aminotransferase, ALT, in 3%), without treatment-related deaths. Among AEs of all grades, the most frequent ones were fatigue in 30% of patients, increase of liver aminotransferase in 28% and cough in 27%.

Entrectinib, as a member of the oral multi-kinase inhibitors, also inhibits (in addition to TRKA, TRKB, TRKC) ROS1 and ALK. Its peculiarity is its high ability to cross the blood–brain barrier, showing activity in patients with CNS disease ^[42][49][62,67]. In August 2019, Entrectinib reached FDA accelerated approval for advanced solid tumors with NTKR fusions and for metastatic ROS1-positive NSCLC based on the results of three multicenter, single-arm, clinical trials (STARTRK-1, STARTRK-2, ALKA-372-001) ^[53][71]. In July 2020, entrectinib received a conditional marketing authorization for both NTRK gene fusion solid tumors and ROS1-positive advanced NSCLC, addressing a major unmet medical need in this subset of patients ^[54][72].

Recently, an updated integrated analysis of three studies focused on ROS1 fusion-positive NSCLC showed an ORR of 67%, a median DoR of 15.7 months, a median PFS of 15.7 months and a median OS was not estimable ^[55][73]. Considering patients with CNS metastases, the intracranial ORR was 79% (95% CI, 58–93%), intracranial DoR was 12.9 months and median intracranial PFS was 12.0 months. The updated safety analysis was consistent with the primary one: most adverse events were grade 1–2 (dysgeusia 43%, dizziness 34%, constipation 31%), while grade 3 adverse events were weight increase (8%), ALT elevation (3%) and diarrhea (3%). Grade 4 adverse events were reported in 3% of patients (hyperuricemia, limbic encephalitis, anorectal disorder, hypertriglyceridemia, myocarditis, blood creatine phosphokinase myocardial band increase, anorectal disorder). Both for larotrectinib and entrectinib, dose modifications were able to control treatment-related AEs.

Another oral multi-kinase inhibitor is taletrectinib (DS-6051b/AB-106), which presents a high selectivity for ROS1/NTRK fusion genes. The efficacy of this TRK inhibitor in ROS1+ NSCLC emerged in two phase 1 studies (U101, conducted in United States, and J102, in Japan) ^[56][74]. With a median follow up of 14.9 months, in ROS1 TKI-naive patients an ORR of 67% was detected, while in crizotinib pretreated patients ORR was 33%. Considering survival, in the first group PFS was 29.1 months, and it was 14.2 months in the second group. Taletrectinib presented a manageable safety profile: most reported AEs were ALT and AST increase (both 73%), and nausea and diarrhea (both 50%), of which grade ≥3 were ALT and AST increase (18% and 9%, respectively) and diarrhea (5%). A multicenter, phase 2 clinical trial (TRUST, NCT04395677) is currently ongoing to evaluate the efficacy of taletrectinib in Chinese ROS1-positive NSCLC patients, while the TRUST-II trial is the ongoing global study (NCT04919811). At ASCO 2021, Zhou et al. ^[57][75] showed that all enrolled Chinese patients at the data cutoff presented a response to the treatment with an ORR of 100% (95% CI, 72–100%) and a safety profile consistent with phase 1 data.

Acquired resistance is still an inevitable circumstance in patients treated with TRK inhibitors, despite the durable and terrific duration of response, regardless of tumor type ^[49][58][67,76]. This acquired resistance is often due to the appearance of new NTRK mutations ^[49][67]. Repotrectinib (TPX-0005) and selirectinib (LOXO-195) are next generation TRK inhibitors designed to overcome resistance to first-generation TRK inhibitors. Repotrectinib is highly selective and active for ALK, ROS1 and NTRK, thus potentially overcoming acquired resistance ^[59][77]. The TRIDENT-1 trial is the ongoing phase 1/2 study of repotrectinib for ALK/ROS1/NTRK fusion gene-positive NSCLC (NCT03093116). Selirectinib’s chemical structure is similar to larotrectinib, apart from the more compact form. In a preclinical study, selirectinib showed resistance to secondary resistance mutations in the TRK kinase domain ^[58][76]. In 31 patients who received selirectinib after progression to a TRK inhibitor (mainly larotrectinib), the ORR was 34%, while it was 45% in patients with secondary resistance mutations ^[60][78]. An ongoing phase 1/2 study to test efficacy and safety of selirectinib is active, not recruiting (NCT03215511).

The family of rat sarcoma oncogenes (RAS) includes the isoforms Kirsten rat sarcoma (KRAS), neuroblastoma rat sarcoma (NRAS) and the Harvey rat sarcoma (HRAS). Ras proteins activate signaling pathways controlling cell proliferation, differentiation and survival ^[61][79]. KRAS accounts for 85% of RAS mutations observed in human cancer and, given that RAS is the most frequently mutated oncogene, it is the most prevalent genomic driver event in NSCLC, present in up to 35% of lung cancers ^[62][63][80,81]. Notably, KRAS mutations are more common in the adenocarcinoma histotype than in squamous NSCLC (20–40% and 5%, respectively) and most frequently found in smokers (30%) vs. non-smokers (11%) and in the Caucasian vs. Asian population (26% and 11%, respectively) ^[64][65][82,83]. The KRAS p.G12C single-nucleotide variant, with glycine replaced by cysteine at codon 12, is the most recurrent variant in NSCLC, with a prevalence of nearly 13% in adenocarcinoma histotype ^[66][24]. It represents 39% of KRAS mutations, followed by G12V (21%) and G12D (17%) ^[67][84].

Sotorasib is a small molecule that specifically and irreversibly inhibits KRAS^G12C from binding covalently to a pocket present only in the inactive GDP-bound conformation, trapping KRAS^G12C in the inactive state and hindering KRAS oncogenic signaling ^[68][85]. In the phase 1/2 CodeBreak 100 trial, sotorasib monotherapy was evaluated in patients with locally advanced or metastatic KRAS^G12C mutated NSCLC. At a median follow-up of 15.3 months, ORR (primary endpoint) was 37.1% and DCR 80.6%, with a median time to response of 1.4 months (range, 1.2–10.1), a median duration of response of 11.1 months and a PFS of 6.8 months. The most common adverse events were diarrhea, nausea, fatigue, arthralgia and increase in the transaminases ^[69][86]. In May 2021, the FDA approved sotorasib as the first targeted agent for KRAS^G12C mutated NSCLC, pretreated with at least one prior systemic therapy. Sotorasib has also been given conditional authorization by the EMA in January 2022 in the same therapeutic setting ^[70][87]. This is the first authorized targeted therapy for tumors with KRAS mutation. Currently, the phase 3 trial, CodeBreak 200 is comparing sotorasib with docetaxel in patients with KRAS^G12C mutated NSCLC in progression to a platinum-based doublet chemotherapy and a checkpoint inhibitor (NCT04303780).

Adagrasib is another highly selective, small-molecule, covalent inhibitor of KRAS^G12C, with a longer half-life than sotorasib ^[71][88]. In the phase 1 trial, KRYSTAL-1, adagrasib was well tolerated and exhibited antitumor activity with 53.3% of partial responses, a median DoR of 16.4 months and a median PFS of 11.1 months ^[72][89]. Recently published data on the registrational phase 2 cohort of the KRYSTAL-1 trial, shows that heavily pretreated patients achieved an ORR of 42.9% with adagrasib, with a median DoR of 8.5 months, median PFS and OS of 6.5 months and 12.6 months, respectively ^[73][90]. Similarly to sotorasib, most common treatment-related adverse events (of any grade) were nausea, diarrhea, vomiting and fatigue ^[72][89]. Based on the findings from the phase 2 KRYSTAL-1 trial, adagrasib received breakthrough therapy designation from the FDA for patients with advanced NSCLC harboring the KRAS^G12C mutation, and a new drug application (NDA) was filed in February 2022. A marketing authorization application has also been submitted to the EMA seeking adagrasib approval in May 2022 ^[74][91]. The ongoing confirmatory phase 3 KRYSTAL-12 trial is evaluating the use of adagrasib compared with docetaxel in patients with KRAS^G12C-mutated NSCLC in a second line setting (NCT04685135).

Various KRAS inhibitors are currently under investigation (e.g., GDC-6036/RG6330; NCT04449874; JDQ443; NCT04699188; D-1553; NCT04585035; JAB-21822; NCT05276726; RMC-6236, NCT05379985; LY3537982; NCT04956640), as well as combination therapy. (Table 34). Other strategies under evaluation for KRAS  mutant NSCLC include the inhibition of downstream signaling pathways with MEK inhibitors (NCT04967079, NCT03170206), either as monotherapy or combined with other molecules (NCT03170206, NCT04735068).