Abstract
Precision oncology is now the evidence-based standard of care for the management of many advanced non-small cell lung cancers (NSCLCs). Expert consensus has defined minimum requirements for routine testing and identification of epidermal growth factor (EGFR) mutations (15% of tumors harbor EGFR exon 19 deletions or exon 21 L858R substitutions) and anaplastic lymphoma kinase (ALK) rearrangements (5% of tumors) in advanced lung adenocarcinomas (ACs). Application of palliative targeted therapies with oral tyrosine kinase inhibitors (TKIs) in advanced/metastatic lung ACs harboring abnormalities in EGFR (gefitinib, erlotinib, afatinib) and ALK/ROS1/MET (crizotinib) has consistently led to more favorable outcomes compared with traditional cytotoxic agents. In addition, mutations leading to resistance to first-line EGFR and ALK TKIs can now be successfully inhibited by soon to be approved third-generation EGFR TKIs (osimertinib, rociletinib) and second-generation ALK TKIs (ceritinib, alectinib). Notably, increasing feasibility, accessibility, and application of molecular profiling technologies has permitted dynamic growth in the identification of actionable driver oncogenes. Emerging genomic aberrations for which TKIs have shown impressive results in clinical trials and expansion of drug labels for approved agents are awaited include ROS1 rearrangements (1–2% of tumors, drug: crizotinib) and BRAF-V600E mutations (1–3% of tumors, drugs: vemurafenib, dafrafenib + trametinib). Evolving genomic events in which TKI responses have been reported in smaller series include MET exon 14 skipping mutations (2–4% of tumors, drug: crizotinib); high-level MET amplification (1–2% of tumors, drug: crizotinib); RET rearrangements (1% of tumors, drug: cabozantinib); and ERBB2 mutations (2–3% of tumors, drug: afatinib), among others. Unfortunately, the most common genomic event in NSCLC, KRAS mutations (25–30% of tumors), is not targetable with approved or in development small molecule inhibitors. Here, we review currently approved, emerging, and evolving systemic precision therapies matched with their driver oncogenes for the management of advanced NSCLC.
Keywords
Introduction
Lung cancer remains a common and most lethal solid tumor malignancy worldwide. However, in recent years, identification of tumor-specific genomic abnormalities has enabled dramatic improvements in the efficacy and tolerability of systemic anticancer therapies for palliation of advanced non-small cell lung cancer (NSCLC). Identification of oncogenic driver mutations (nucleotide mutations, amplifications, and rearrangements) and application of precision tyrosine kinase inhibitors (TKIs) have rendered the ability to optimally match targeted systemic therapies with tumor-specific abnormalities and have changed the outlook and approach to patients with advanced disease, particularly in lung adenocarcinomas (ACs).
To date, five oral precision therapies have won approval from the United States (US) Food and Drug Administration (FDA) for use in advanced NSCLC: gefitinib, erlotinib, and afatinib for epidermal growth factor (EGFR)-mutated cancers (15–20% of all lung ACs); and crizotinib and ceritinib for anaplastic lymphoma kinase (ALK)-rearranged cancers (5% of lung ACs). In addition, mutations resulting in primary/secondary resistance to these EGFR and ALK TKIs (i.e. EGFR-T790M and ALK mutations) can now be successfully inhibited by soon to be approved third-generation EGFR TKIs (osimertinib, rociletinib) and second-generation ALK TKIs (ceritinib, alectinib), respectively. As a consequence, molecular testing for EGFR mutations and ALK rearrangements is now the evidence-based standard of care in the initial diagnostic evaluation and therapeutic planning for patients with this histology and with advanced/metastatic disease [Lindeman et al. 2013].
On the basis of the relative tolerability and objective, durable responses observed with the use of TKIs in EGFR-mutated and ALK-rearranged NSCLCs (compared with traditional cytotoxic chemotherapies), research into additional putative targets and mechanisms of resistance has exploded in recent years (Table 1). Emerging targetable oncogenic drivers for which TKIs have shown impressive results in clinical trials (and expansion of the FDA-approved indications for use of TKIs is awaited) include: c-ros oncogene 1 (ROS1) rearrangements (1–2% of tumors, drug: crizotinib) and Raf murine sarcoma viral oncogene homolog B1 (BRAF)-V600E mutations [1–3% of tumors, drugs: vemurafenib, dabrafenib plus the mitogen-activated protein kinase (MEK) inhibitor trametinib].
Molecular targets and their targeted therapies.
ALK, anaplastic lymphoma kinase; BRAF, serine/threonine-protein kinase B-raf; DDR2, discoidin death receptor 2; EGFR, epidermal growth factor; ERBB2, erb-B2 receptor tyrosine kinase 2; FGFR, fibroblast growth factor receptor; HRAS, harvey rat sarcoma viral oncogene homolog; KRAS, v-ki-ras2 kirsten rat sarcoma viral oncogene homolog; LKB1, liver kinase B1; MAP2K1, mitogen-activated protein kinase kinase 1; MET, mesenchymal epithelial transition factor proto-oncogene; NF1, neurofibromin 1; NRAS, neuroblastoma RAS viral oncogene homolog; NTRK1, neurotrophic tyrosine kinase, receptor, type 1; PI3K, phosphatidylinositol 3 kinase; RET, rearranged during transfection; RIT1, ras-like without CAAX 1; ROS1, c-ros oncogene 1; STK11, serine/threonine kinase 11; TKI, tyrosine kinase inhibitor.
Evolving genomic events for which TKI responses have been reported in smaller series (but as yet inadequate data for routine use of TKIs) include mesenchymal epithelial transition factor proto-oncogene (MET) exon 14 skipping mutations (2–4% of tumors, drug: crizotinib); high-level MET amplification (1–2% of tumors, drug: crizotinib); rearranged during transfection (RET) rearrangements (1% of tumors, drug: cabozantinib); and V-erb-b2 erythroblastic leukemia viral oncogene homolog 2 (ERBB2) mutations (2–3% of tumors, drug: afatinib), among others. Unfortunately, the most common genomic event in NSCLC, v-ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutations (25–30% of tumors), is not currently ‘druggable’ with approved or in-development small molecule inhibitors.
Additionally, the vast majority of currently known and targetable tumor-specific genomic changes occur in lung ACs; therefore, the identification of therapeutically relevant oncogenic drivers remains a persistent and significant area of unmet need with respect to the management of lung squamous cell carcinomas (SCCAs). Notably, in order to identify rarer potential targets (which may not be detected by standard/targeted tumor genotyping platforms), comprehensive testing with next generation sequencing (NGS) may well be needed [Zheng et al. 2014]. Determining optimal circumstances for use of NGS assays in day-to-day clinical practice is an area of ongoing exploration in the management of advanced lung cancers.
Here, we will review currently approved, emerging, and evolving targeted systemic palliative therapies for advanced ACs and SCCAs of the lung (Table 1, Figure 1).
Pie chart with known mutually exclusive genomic events in lung adenocarcinoma. Highlighted in different colors are the driver oncogenes that function as predictive biomarkers for approved, emerging, evolving and future precision therapies with kinase inhibitors. ALK, anaplastic lymphoma kinase; BRAF, serine/threonine-protein kinase B-raf; EGFR, epidermal growth factor; ERBB2, erb-B2 receptor tyrosine kinase 2; FGFR, fibroblast growth factor receptor; HRAS, harvey rat sarcoma viral oncogene homolog; KRAS, v-ki-ras2 kirsten rat sarcoma viral oncogene homolog; MAP2K1, mitogen-activated protein kinase kinase 1; MET, mesenchymal epithelial transition factor proto-oncogene; NF1, neurofibromin 1; NRAS, neuroblastoma RAS viral oncogene homolog; NTRK1, neurotrophic tyrosine kinase, receptor, type 1; RET, rearranged during transfection; RIT1, ras-like without CAAX 1; ROS1, c-ros oncogene 1.
Lung ACs: approved targeted therapies
EGFR mutations in advanced NSCLC
EGFR-mutated lung cancers, first identified in 2004, represent the largest subgroup of advanced NSCLCs with an approved biologically and therapeutically relevant target, accounting for some 15–20% of all lung ACs diagnosed in the USA [Keedy et al. 2011]. EGFR mutations are detected in approximately 30–40% of lung ACs from Asian patients and in 10–20% of lung ACs from white or black patients [Jorge et al. 2014]. EGFR is a member of the ErbB family of transmembrane receptor tyrosine kinases involved in signal transduction pathways that regulate proliferation and apoptosis. Mutations in exon 19 (in-frame deletions around the LREA motif; amino acids L747 to A750) and exon 21 (L858R substitution) result in a constitutively active receptor tyrosine kinase by affecting changes to the receptor’s adenosine triphosphate (ATP) binding site. Exon 19 deletions and L858R substitutions account for 85–90% of EGFR-mutated tumors [Gerber et al. 2014]. Though EGFR mutations are most frequently seen in East Asian women and in those who are never smokers with AC, multiple expert groups recommend rapid testing for EGFR mutations in all patients with advanced AC, regardless of sex, race, or smoking history [Morgensztern et al. 2015]. EGFR mutations are mutually exclusive of other oncogenic drivers in TKI-naïve tumors, including ALK, KRAS, ERRB2 and BRAF [Li et al. 2011].
Oral EGFR TKIs in advanced NSCLC: first- and second-generation EGFR TKIs
In the presence of a sensitizing EGFR mutation, the recommended first-line palliative systemic therapy is one of three approved TKIs: gefitinib 250 mg orally daily, erlotinib 150 mg orally daily, or afatinib 40 mg orally daily [National Comprehensive Cancer Network, 2015]. Both gefitinib and erlotinib are first-generation TKIs which bind reversibly to the aberrant EGFR. The second-generation TKI, afatinib, binds irreversibly to both wild-type (WT) and aberrant EGFR. EGFR exon 19 deletions and L858R proteins create a favorable therapeutic window (i.e. the mutant EGFR is inhibited at a lower threshold when exposed to TKI blockade as opposed to WT EGFR); this defines the maximum tolerated dose in humans for the first- and second-generation EGFR TKIs. Sensitizing EGFR mutations may be identified in tumor specimens via gene sequencing assays.
In contrast to their limited activity in EGFR WT NSCLC, numerous randomized trials have demonstrated that, in EGFR-mutated NSCLC, first-generation reversible EGFR TKIs (gefitinib and erlotinib) and second-generation irreversible EGFR TKIs (afatinib) result in higher response rates (RRs), better quality of life (QoL), and longer progression-free survival (PFS) and overall survival (OS) than traditional cytotoxic chemotherapies. Based on these data, the TKIs gefitinib (since 2009 in Europe/Asia and 2015 in the USA), erlotinib (since 2011), and afatinib (since 2013) have received approval labels that indicate that, in the first-line setting, their use should be restricted to the treatment of lung ACs harboring EGFR exon 19 deletions or L858R mutations.
Other EGFR mutations have also been associated with some sensitivity to gefitinib, erlotinib, and afatinib. These include exon 18 point mutations in position G719 (G719A, C or S, ~3% of mutations), inframe exon 19 insertions (<1% of mutations), and the exon 21 L861Q mutant (~2% of mutations). The third most common group of EGFR mutations in NSCLC is composed of heterogeneous in-frame insertions within exon 20 of EGFR that comprise 10% of all mutations. Most (>90%) of these mutations lie near the end of the C-helix within the N-lobe of the kinase and lead to EGFR activation without altering the TKI-binding pocket. Unlike EGFR exon 19 deletion and L858R-bearing tumors, most NSCLCs with EGFR exon 20 insertion mutations do not respond clinically to gefitinib, erlotinib or afatinib. Therefore, EGFR exon 20 insertions, outside the TKI-sensitive A763_Y764insFQEA, comprise a unique set of EGFR-activating mutations that renders no clinical benefit with currently available first- or second-generation EGFR TKIs [Yasuda et al. 2012b, 2013].
The results of key randomized trials of oral EGFR TKIs for EGFR exon 19 deletion and L858R-mutated tumors are summarized in Table 2. The randomized trials comparing the EGFR TKIs gefitinib, erlotinib, or afatinib with traditional first-line cytotoxic chemotherapy have consistently shown RRs that exceed 60–70%, PFS times that are 42–84% higher, and improved QoL parameters compared with traditional cytotoxic chemotherapies [Sequist et al. 2013]. Phase III clinical trials of these agents have demonstrated a median PFS of 9–13 months and OS times that exceed 24–30 months [Mok et al. 2009; Mitsudomi et al. 2010; Maemondo et al. 2010; Zhou et al. 2011; Rosell et al. 2012]. Additionally, application of EGFR TKIs has yielded preserved improvements in PFS, even in the second-line setting [Rosell et al. 2012]. A proven OS benefit, while not demonstrated in individual trials (likely due to >90% use of EGFR TKIs in crossover groups at progression) is demonstrable when compiling patient outcomes across studies [Zhao et al. 2015]. In addition, compared with historical controls treated with systemic cytotoxic chemotherapy, OS of patients with EGFR-mutated advanced NSCLCs has migrated substantially over time and in the face of regular application of these oral EGFR TKIs [Mok et al. 2009; Mitsudomi et al. 2010; Maemondo et al. 2010; Zhou et al. 2011; Rosell et al. 2012; Yang et al. 2013; Wu et al. 2014].
Phase III clinical trials of EGFR tyrosine kinase inhibitors (gefitinib, erlotinib, or afatinib) versus platinum-doublet chemotherapy in EGFR-mutated NSCLC.
In the trials above, starting doses were gefitinib 250 mg daily, erlotinib 150 mg daily or afatinib 40 mg daily.
EGFR epidermal growth factor receptor; EURTAC, European tarceva versus chemotherapy; IPASS, Iressa Pan-Asia Study; NSCLC, non-small cell lung cancer; NEJ, North East Japan; NR, not reported; NRea, not reached; OS, overall survival; PFS, progression-free survival; WJTOG, West Japan Thoracic Oncology Group.
First- and second-generation EGFR TKIs have generally been tolerated at their recommended starting doses (250 mg daily of gefitinib, 150 mg daily of erlotinib, and 40 mg daily of afatinib) across these various studies, with common notable toxicities/adverse events (AEs) including acneiform rash, paronychia, diarrhea, liver function abnormalities, and rare cases of pneumonitis/pulmonary fibrosis for first-generation EGFR TKIs; second-generation EGFR TKIs have higher rates of rash, diarrhea, and mucositis. It is well known that dose reductions of gefitinib, erlotinib, and afatinib can improve AEs, without significantly affecting efficacy, in most patients with EGFR-mutated ACs [Yeo et al. 2010].
Notably, in the IPASS study, patients receiving gefitinib were not selected a priori for the presence of known activating EGFR mutations, though the studied cohort was enriched with Asian never smokers in whom EGFR mutations are more commonly seen. The mutated EGFR cohort was therefore evaluated in a retrospective subgroup analysis. Outcomes in the EGFR WT population were inferior to those receiving chemotherapy, reinforcing the notion that first-line EGFR blockade should be restricted to those with known EGFR-sensitizing mutations [Mok et al. 2009].
Resistance to EGFR blockade in advanced NSCLC: third-generation EGFR TKIs
Though the initial response to EGFR blockade is often robust, resistance is inevitable, with 50–65% of patients acquiring resistance within the first 12 months of therapy with a first- or second-generation oral EGFR TKI. Mechanisms of resistance that have been implicated to date include secondary acquisition/clonal selection of tumor cells harboring a point mutation in exon 20 of the EGFR (i.e. EGFR-T790M) [Kobayashi et al. 2005; Costa et al. 2007; Nguyen et al. 2009]; activation of bypass oncogenic pathways (13%); BRAF mutation, ERBB2 or MET amplification (5%); and transformation to small cell histology (3%) [Sequist et al. 2011; Yu et al. 2013a; Chong and Janne, 2013]. Activation of alternate signal transduction pathways (i.e. ‘bypass pathways’) that have been linked to acquired resistance include signaling via the phosphatidylinositol 3 kinase (PI3K)/protein kinase B (AKT), mitogen-activated protein kinase (MAPK)/extracellular signal regulated kinase (ERK), Janus kinase (JAK)/signal transducer and activator of transcription (STAT), or ERBB2 pathways [Costa et al. 2007; Sequist et al. 2011; Yu et al. 2013a; Chong and Janne, 2013].
The continuation of EGFR blockade following progression or resistance is an evolving area of investigation. Repeat biopsy at the time of progression on the current EGFR TKI is an emerging standard. If there is asymptomatic or indolent progression, then continuation of the current EGFR TKI can be considered if the patient is believed to be deriving clinical benefit. Central nervous system (CNS) progression is a common site of progression and incidence rises with each year of continued survival on targeted therapy [Rangachari et al. 2015]. In the case of symptomatic oligoprogression, application of local palliative therapies (i.e. surgery or radiation) with continuation of the current EGFR TKI may provide durable control of disease (6–30 months in published studies), permitting delay of subsequent systemic therapies that may be more toxic [Weickhardt et al. 2012; Yu et al. 2013b]. Another non-FDA approved strategy after resistance to first-line EGFR therapy is combining cetuximab with afatinib in patients with acquired resistance to first-generation EGFR inhibitor. The phase I trial included those with or without T790M mutation and patients had an overall RR of 29% and PFS of 4.7 months, though the rate of grade 3/4 AEs was 44% in the group with combination therapy [Janjigian et al. 2014].The optimal transition from EGFR blockade to an alternate line of therapy at the time of symptomatic or significant radiographic progression or resistance is less well vetted. In one retrospective single institution study, some 25% of patients were noted to demonstrate a clinically significant flare in their disease when the current EGFR TKI was stopped [Chaft et al. 2011]. In the recently published IMPRESS study, continued gefitinib beyond progression in patients with EGFR-mutated tumors receiving second-line palliative chemotherapy with a platinum-based regimen offered no OS benefit, though there were minimal adverse effects if the TKI was continued [Soria et al. 2015]. In light of these findings, our own institutional practice has been to continue with the current EGFR TKI up until the start of subsequent therapy if there have been no significant issues with treatment-related toxicities.
For tumors in which the EGFR-T790M mutation is identified, the evolution of potent third- generation EGFR TKIs (osimertinib and rocelitinib) has permitted clinically beneficial outcomes in this subset of patients. Both drugs are being investigated in ongoing trials, with recently published data from phase I and II studies indicating objective RRs around 60% in patients with T790M-affected disease. Notably, RRs and PFS were significantly higher in patients in the T790M+ cohorts (RR 61%, PFS 9.6 months with osimertinib; RR 59% for rociletinib) compared with the T790M- cohorts (RR 21%, PFS 2.8 months with osimertinib; RR 29% for rociletinib). This likely reflects the biologically relevant activation of alternate bypass pathways in the T790M group and for which further blockade of signaling via the EGFR pathway does not afford meaningful antineoplastic effect [Janne et al. 2015; Sequist et al. 2015]. Both drugs were generally well tolerated, with notable toxicities, including diarrhea, rash, nausea, and anorexia (for osimertinib); and hyperglycemia, diarrhea, nausea, cardiac events, and fatigue (for rociletinib). Osimertinib has been granted breakthrough therapy designation, orphan drug, and fast track status by the FDA; its approval with a companion diagnostic assay for the EGFR-T790M mutation is imminent. Rociletinib has also been granted breakthrough therapy designation by the FDA.
Despite remarkable and lasting responses seen with osimertinib and rociletinib in lung ACs with acquired resistance to first- or second-generation EGFR TKIs harboring the EGFR-T790M mutation, it is increasingly evident that selective pressure eventually leads to resistance to even third-generation EGFR TKIs through the development of the EGFR-C797S mutation in exon 20 and other still to be defined bypass mechanisms. T790M + C797S generates proteins that are resistant to osimertinib, rociletinib, and all reversible/irreversible EGFR TKIs by impairing covalent binding of these drugs to the C797 amino acid residue of EGFR. Novel treatment strategies are therefore warranted to avoid or delay resistance. Likely strategies include combining EGFR inhibitors with PI3K/MAPK inhibitors, immunotherapies, or cytotoxic agents to expand the spectrum of antineoplastic effect.
ALK rearrangements in advanced NSCLC
Rearrangements at the ALK locus, first identified in 2007, account for 5% of patients with lung ACs. Similar to the demographic profile for patients with EGFR-mutated advanced NSCLCs, aberrations in ALK are noted to be more common in never or light smokers; abnormalities in ALK are also similarly mutually exclusive with other oncogenic drivers, including EGFR, KRAS, ERBB2, and BRAF [Gerber et al. 2014]. The most common genomic event involves fusion of ALK and echinoderm microtubule-like protein (EML4), which functions as a potent oncogenic driver and renders sensitivity to ALK TKIs [Camidge et al. 2012]. Although ALK is not an important signal for untransformed lung cells (hence the low to undetectable expression of ALK in normal lung tissues), aberrant ALK signaling is a main driver of ALK-rearranged lung ACs. ALK rearrangements can be identified in tumor samples via use of fluorescence in situ hybridization (FISH), immunohistochemistry (IHC), or NGS; in the USA, a FISH-based assay (break apart probe) and IHC-based assay (Ventana ALK D5F3 Companion Diagnostics assay) are the currently approved genomic tests.
Oral ALK TKIs in advanced NSCLC
Crizotinib is a multitargeted TKI of ALK, ROS1, and MET. In 2010, the initial experience of an expanded phase I trial with crizotinib in patients with ALK-rearranged advanced NSCLC showed clinical benefit (defined as tumor shrinkage or stable disease) in most patients [Kwak et al. 2010]. As demonstrated in subsequent studies of ALK-targeted therapies in this setting, RRs of 60–75% with PFS of 7.7–11 months [Camidge et al. 2012; Shaw et al. 2013; Solomon et al. 2014] have been consistently demonstrated. In retrospective analysis, patients who were enrolled in the initial phase I study were observed to have a 1- and 2-year OS of 74% and 54%, respectively [Shaw et al. 2011]. The results of key trials of ALK-directed therapies in advanced NSCLC are summarized in Table 3.
Clinical trials of crizotinib (first generation) and second-generation ALK inhibitors for advanced ALK-rearranged NSCLC.
ALK, anaplastic lymphoma kinase; NR, not reported; NRea, not reached; NSCLC, non-small cell lung cancer; OS, overall survival; PFS, progression-free survival; RR, response rate.
Combined OS of docetaxel and pemetrexed arms.
The second-line randomized phase III trial (PROFILE 1007, which compared crizotinib with docetaxel or pemetrexed), reported in 2013, confirmed that crizotinib led to improved objective outcomes and QoL compared with standard cytotoxic chemotherapies: RR of 65% versus 20%, respectively; PFS of 7.7 versus 3.0 months, respectively. The first-line randomized phase III trial (PROFILE 1014, which compared crizotinib with platinum- pemetrexed), reported in 2014, unequivocally showed that RR, PFS, and QoL favor the oral TKI even against the most effective first-line platinum doublet for lung ACs. The median OS of ALK-rearranged lung ACs has been reported to exceed 24–36 months in crizotinib-treated cases. As a consequence, first-line palliative systemic therapy with crizotinib now presents the standard of care for patients with known ALK rearrangements.
Resistance to ALK blockade in advanced NSCLC: second-generation ALK TKIs
As observed with targeted therapies in other settings, resistance to ALK blockade eventually emerges. Acquired resistance to crizotinib therapy can come about through multiple biological mechanisms, including: ALK kinase mutations (30% of cases), that is, L1196M (the gatekeeper position for ALK), F1174L, and G1202R; activation of other oncogenes (such as EGFR) resulting in bypass signaling cascades; and pharmacokinetic failings of the drug itself (i.e. inadequate systemic levels or poor CNS penetration) [Doebele et al. 2012; Yamaguchi et al. 2014; Costa et al. 2011]. In many cases, no definitive resistance mutation can be identified.
As with resistance to EGFR TKIs, asymptomatic, indolent progression on crizotinib does not mandate an immediate change in therapy if the patient is believed to be deriving clinical benefit. In cases of symptomatic oligoprogression, application of local palliative therapies (i.e. surgery or radiation) with continuation of the current ALK TKI may provide durable control (>6 months in published studies) of disease prior to the need to initiate a novel systemic therapy [Costa et al. 2015]. However, clinicians should be cautious in continuing with crizotinib when patients develop oligometastatic disease given the low CNS penetration of this TKI [Costa et al. 2011]. As with EGFR-mutated disease, CNS progression is a common complication, with upwards of 45% of patients with CNS involvement by 3 years in the setting of targeted therapy [Rangachari et al. 2015]. For further information on management and treatment implications of CNS disease, please refer to these recent reviews [Costa et al. 2015; Rangachari et al. 2015; Baik et al. 2015]. Second-generation ALK TKIs with more potent activity against ALK and ALK kinase mutants have been developed and are being studied. The most advanced second-generation ALK TKIs include ceritinib and alectinib. In 2014, the FDA approved ceritinib, a second-generation ALK inhibitor. Approval was granted on the basis of a multicenter, single-arm, open-label clinical trial enrolling a total of 163 patients with metastatic, ALK-rearranged NSCLC who had progressed on or were intolerant of crizotinib. The expanded phase I study demonstrated a 58% overall RR and median PFS of 7 months, comparable to outcomes observed in patients receiving first-line ALK blockade with crizotinib. Additionally, responses in previously untreated CNS disease were seen [Shaw et al. 2014a]. Alectinib also has similar systemic and intracranial activity, and its approval is imminent in the USA; it is already approved for use in other countries worldwide.
While these second-generation ALK TKIs (ceritinib, alectinib, brigatinib) may afford significant additional palliative benefits to patients who have previously progressed on crizotinib, they cannot achieve durable control of disease for ALK-rearranged tumors as second-line monotherapy alone. For ceritinib, emerging data show that ALK-G1202R and F1174V/C are ceritinib-resistant mutations found in rebiopsy specimens; the associated median PFS for patients harboring these abnormalities is less than 6 months. It is likely that additional ALK mutations, oncogenic bypass signaling pathways, and pharmacokinetic escape in sanctuary sites (i.e. the CNS) are responsible for acquired resistance to crizotinib, ceritinib, and alectinib when given in sequence. As with EGFR-mutated disease, it will therefore be necessary to evolve more potent ALK TKI combination therapies (i.e. combining ALK TKIs with PI3K/MAPK inhibitors, immunotherapies, or cytotoxic agents) to achieve maximal and durable benefit. There are other second- and third-generation ALK TKIs currently in clinical development with more than a 10-fold higher potency against ALK than crizotinib [Seto et al. 2013; Gadgeel et al. 2014]. X396, which is in phase I trials, showed responses in both crizotinib-naïve and crizotinib-resistant patients with ALK-positive NSCLC [Horn et al. 2014], Additionally, PF-06463922, a third-generation ALK/ROS1 inhibitor, has higher CNS penetration and activity against crizotinib-resistant tumors [Zou et al. 2015a, 2015b; Mologni et al. 2015].
Lung ACs: emerging targeted therapies
Additional known driver oncogenes that are found in smaller subsets of advanced lung ACs are explored below and with consideration of off-label use of approved targeted therapies or participation in a clinical trial.
ROS1 rearrangements
Chromosomal rearrangements of ROS1 occur in 1–2% of NSCLC cases. ROS1 fusions can be detected by FISH, polymerase chain reaction, IHC, NGS, and chromogenic in situ hybridization. Similar to EGFR- and ALK-driven NSCLCs, most patients with ROS1-rearranged NSCLC are younger, never smokers with AC histology. There is high homology in the kinase domain of ROS1 and ALK, which initially led to testing of ALK inhibitors in ROS1-positive cell lines [Chin et al. 2012, Yasuda et al. 2012a]. The expanded phase 1 study of crizotinib (a multitargeted TKI with inhibitory activity against ALK, ROS1, and MET) for patients with ROS1 rearrangements showed an overall RR of 72% with three complete responses and 33 partial responses. Median PFS was 19.2 months with 25 patients (50% of the cohort) still in follow up without progression at the time of publication. There was no correlation observed between response to crizotinib and type of ROS1 rearrangement. Similar to the experience with crizotinib in ALK-rearranged disease, the most common toxicities were transient visual impairment, nausea, diarrhea, constipation, or peripheral edema [Shaw et al. 2014b]. Preliminary data from a phase II trial of crizotinib in ROS1-rearranged NSCLC conducted by the French National Cancer institute also showed promising results, with an overall RR of 63% and few severe AEs [Moro-Sibilot et al. 2015].
Importantly, preliminary data from a retrospective analysis of patients with ROS1-rearranged NSCLC showed remarkably higher responses to chemotherapy and overall survival, regardless of the systemic therapy used, and exceeding what has been observed with other molecular subtypes that predominate in never or light smokers (including EGFR-mutated and ALK-rearranged NSCLCs), suggesting that ROS1 may serve as both a prognostic and predictive biomarker in clinical practice [Schleffer et al. 2015]. Similar to the experience with other targeted therapies in advanced NSCLC, however, resistance to TKI therapy is an inevitable reality and ROS1 mutations have already been described in clinical specimens [Awad et al. 2013]. Whether sequential therapy with other more potent ROS1 TKIs in ROS1-rearranged NSCLC is a successful strategy for crizotinib-resistant tumors remains to be determined. PF-06463922 has shown promising results in patients with ALK- or ROS1-rearranged NSCLC, who have CNS disease and those who received prior TKI therapy [Bauer et al. 2015]
BRAF mutations
BRAF is a serine-threonine kinase in the MAPK pathway. BRAF mutations are found in 1–3% of patients with NSCLC, most often in ACs and in former or current smokers. The V600E point mutation accounts for roughly half of all BRAF mutations in NSCLC, resulting in constitutive downstream phosphorylation of the MAPK signaling pathway; this has been associated with biologically more aggressive disease, with reduced disease-free survival and OS [Marchetti et al. 2011]. BRAF inhibitors (vemurafenib and dabrafenib) are FDA approved for metastatic melanoma, but not in NSCLC, though currently listed as emerging therapies as per expert guidelines for BRAF-V600E-mutated NSCLC [National Comprehensive Cancer Network, 2015].
A single institution case series of five patients with BRAF-V600E-mutated NSCLC showed an OS of 46 months in patients treated with an oral BRAF inhibitor [Goldman and Gray, 2015]. A phase II trial of BRAF-V600E-mutated NSCLC treated with single agent dabrafenib showed a 32% overall RR [Planchard et al. 2013]. Interim results from a phase II study of the BRAF inhibitor dabrafenib in combination with MEK inhibitor trametinib in BRAF-V600E-mutated metastatic NSCLC showed an overall RR of 63%, with radiographic responses observed within 6 weeks of drug initiation [Planchard et al. 2015]. The most common AEs included diarrhea, nausea, vomiting, anorexia, pyrexia, cough, rash, and peripheral edema. A low rate of severe AEs was noted, including hyponatremia, neutropenia, and dehydration; one patient had a fatal event related to pleural effusion/disease progression. Twenty-seven percent of patients required dose reduction due to AEs [Planchard et al. 2015]. Of note, a case report of a patient with BRAF-V600E-mutant NSCLC with CNS metastases achieved an objective intracranial response to therapy with vemurafenib [Robinson et al. 2014].
MET mutations and amplification
MET is a receptor tyrosine kinase activated by binding of its ligand hepatocyte growth factor (HGF), with downstream effects on the PI3K/AKT and MAPK pathways. Mechanisms of MET-associated aberrant signaling in NSCLC include overexpression of its ligand HGF or its receptor, high-level amplification, and point mutations in juxtamembrane or kinase domains. MET overexpression generally carries a poor prognosis [Park et al. 2012]. Existing evidence suggests that true high-level MET amplification as a mutually exclusive driver oncogene is a rare event, occurring in only 2% of lung ACs [Camidge et al. 2014]. Notably, MET amplification can be found in up to 15–20% of EGFR-mutated NSCLCS with acquired resistance to EGFR inhibitors [Sequist et al. 2011; Yu et al. 2013a; Chong and Janne, 2013]. Given this association, developing strategies for combining MET inhibition with EGFR blockade is an active area of development.
The ongoing phase I trial of crizotinib (which led to the initial proof of concept of TKI inhibition for ALK- or ROS1-rearranged NSCLCs) continues to enroll patients with MET amplification. Initial results have shown activity of crizotinib in this subgroup. The highest RR of 66.6% was observed in the six tumors with the highest MET copy numbers. The most common AEs included diarrhea, nausea, vomiting, peripheral edema, and visual impairment [Camidge et al. 2014]. It seems that true high-level amplification of MET as a mutually exclusive driver oncogene is a rare event occurring in only 2% of lung ACs [Camidge et al. 2014]. Defining a cutoff for MET copy number using FISH or NGS is essential to understand which tumors are or are not addicted to MET signaling.
Mutations in the MET exon 14 RNA splice acceptor and donor sites leading to exon skipping, deletion of the juxtamembrane domain, and persistence of an aberrant MET protein have been reported to be oncogenic in preclinical models [Kong-Beltran et al. 2006]. MET exon 14 skipping deletions can be found in 4% of lung ACs [Cancer Genome Atlas Research Network, 2014; Frampton et al. 2015]. Clinical response or stable disease with off-label use of either crizotinib or cabozantinib (both of which are multitargeted TKIs with inhibitory effects on MET) have been described in four patients with metastatic lung ACs harboring exon 14 skipping mutations [Paik et al. 2015]. Ongoing clinical trials will define if crizotinib or other MET TKIs, such as MGCD265, should be used in this cohort of patients with NSCLC. MGCD265 is a spectrum-selective and ATP-competitive inhibitor with MET that is generally well tolerated and the phase I clinical trial enrollment has been expanded [Kollmannsberger et al. 2015].
RET rearrangements
One to two percent of lung cancers have rearrangements in RET, which encodes the RET receptor tyrosine kinase [Ju et al. 2012]. RET rearrangements are found in lung ACs, more commonly in never smokers, and are mutually exclusive with other mutations, such as EGFR, ALK, or KRAS [Lipson et al. 2012]. There are a number of multitargeted TKIs with RET-blocking activities that are FDA approved for use in other malignancies, including vandetanib, sorafenib, sunitinib, and cabozantinib. In vitro RET inhibition with vandetanib, sunitinib, and sorafenib has shown efficacy in preventing growth of tumors [Kohno et al. 2012; Lipson et al. 2012; Takeuchi et al. 2012; Matsubara et al. 2012; Suzuki et al. 2013; Saito et al. 2014]. Lenvatinib, a multitargeted TKI [vascular endothelial growth factor receptors 1-3 (VEGFR1-3), fibroblast growth factor receptor 1–3 (FGFR1–3), RET], has been shown in a preclinical study to inhibit cell growth in a lung cancer model [Okamoto et al. 2013]. Additionally, ponatinib, an FDA-approved therapy for Philadelphia chromosome-positive leukemia, has been shown in vitro to inhibit RET [Mologni et al. 2013].
In a case series, three patients with RET-rearranged NSCLC were administered cabozantinib (60 mg daily) and remained progression free for 4–8 months with partial response or stable disease. Two of the three patients required dose reductions due to toxicity (fatigue and hypertension), though continued to have clinical benefit on the lower dose [Drilon et al. 2013]. Another case report of a RET-rearranged NSCLC refractory to chemotherapy had a response to vandetanib (300 mg daily) [Gautschi et al. 2013]. An update from a phase II trial of 20 patients with advanced RET-rearranged NSCLCs receiving cabozantinib demonstrated 33% of patients with a partial response to therapy and 72% with stable disease. Median PFS was 7 months. Toxicities included diarrhea, palmar-plantar erythrodyesthesia, fatigue, transaminitis, and thrombocytopenia [Drilon et al. 2015]. Ongoing phase II clinical trials for patients whose tumors harbor RET rearrangements will provide future direction for this population of patients.
ERBB2 mutations
Human epidermal growth factor receptor 2 (HER2, ERBB2) is a tyrosine kinase in the HER (ErbB) family. HER2 mutations occur in 2–3% of NSCLC cases and are mutually exclusive of other known oncogenic drivers. Most commonly, HER2-mutated tumors occur in female patients who are never smokers with AC histology [Buttitta et al. 2006]. HER2-mutant metastatic lung cancers have a distinct natural history, with a median OS of 1.6 years [Kris et al. 2014]. A recent single institution study reported a median OS of 2.3 years, with most patients receiving standard systemic chemotherapy compared with HER2-targeted therapy [Eng et al. 2015]. In a retrospective study of 16 patients with HER2-mutated NSCLC, the observed RR with trastuzumab-based regimens was 60% [Mazieres et al. 2013]. As previously discussed, afatinib is a potent irreversible inhibitor of both EGFR and HER2. In a phase II study of five patients with NSCLCs harboring a HER2 mutation and treated with higher than usual doses of afatinib (starting dose of 50 mg daily), three patients achieved a partial response or stable disease [De Greve et al. 2012]. However, the effects of afatinib in this cohort are limited by needing to achieve dosing schemes that may be unreachable at the 20–40 mg daily dose that is often used in EGFR-mutated tumors. Additional clinical trials of different dosing schemes of WT sparring EGFR/ERBB2 TKIs are warranted in this cohort of NSCLCs.
Importantly, HER2-mutated lung cancers are distinct from tumors in which HER2 amplification or HER2 protein overexpression occur [Li et al. 2015]. In contrast to the experience with breast or gastric cancers, HER2 amplification in NSCLC does not portend a prognostic or predictive role in NSCLC. The application of the anti-HER2 monoclonal antibody trastuzumab has not shown significant clinical benefit in unselected HER2-amplified NSCLCs [Langer et al. 2004; Gatzemeier et al. 2004].
Lung ACs: future targeted therapies
KRAS
Activating KRAS mutations account for 25–30% of all ACs and are found most commonly in current and former smokers [Rodenhuis and Slebos, 1992]. KRAS mutations in lung cancer most commonly occur at codons 12 (>80%), 13, or 61 and involve nucleotide transversions (i.e. purine for a pyrimidine) which are characteristically associated with tobacco-associated mutagenesis. As noted with other oncogenic drivers, KRAS mutations are mutually exclusive with other genomic events. Multiple oncogenic effectors lay downstream to KRAS (include the MAPK pathway) and with resultant pleiotropic effects on pathways leading to malignant cell proliferation, survival, and invasion. Unfortunately, no currently available or in development targeted therapies have demonstrated any significant benefit for patients with this molecular phenotype. The most advanced clinical developments for KRAS-mutated tumors include the use of MEK inhibitors in conjunction with cytotoxic chemotherapy, with selumetinib having a small signal of activity when combined with docetaxel [Janne et al. 2013]; or the use of immune checkpoint inhibitors.
Other possible driver oncogenes
A significant proportion of lung ACs, as recently confirmed by the massive sequencing efforts of the Cancer Genome Atlas (TCGA) and the Lung Cancer Mutation Consortium, harbor genomic aberrations that encompass future putative therapeutic targets [Cancer Genome Atlas Research Network, 2014]. Some of these may eventually translate into precision therapies if an oral inhibitor can be matched to the genomic event in preclinical models and subsequently in clinical trials (Figure 1).
Lung SCCAs: evolving and future therapies
Though potentially targetable mutations have been identified in lung SCCAs in the TCGA cohort, none as yet have progressed to the point of being paired with a consistently effective or approved systemic targeted therapy. A large cohort of patients with SCCA at a single institution underwent tumor genotyping with capability of identifying some 300 genomic targets. Interim results indicate PIK3CA mutations in 12%, CCND1–2 amplification in 16%, CDK4 amplification in 3%, FGFR1–4 amplification in 24%, FGFR1–4 mutation in 7%, and MET amplification in 3%. Of note, however, the majority of amplification events were low level and thus may be challenging to target in a biologically and clinically relevant manner [Sacher et al. 2015]. The National Cancer Institute-sponsored LungMAP study will evaluate these aforementioned targets for lung SSCAs.
DDR2 mutations
Discoidin death receptor 2 (DDR2) is a receptor tyrosine kinase activated by collagen that may signal via the SRC and STAT pathways. DDR2 mutations occur in up to 4–5% of lung SCCAs [Hammerman et al. 2011]. In preclinical models and in case reports, dasatinib, a multitargeted TKI approved for treatment of chronic myelogenous leukemia, blocks DDR2 with case report level clinical responses noted [Hammerman et al. 2011; Haura et al. 2010]. Further evaluation of this targeted therapy is needed. However, caution needs to be taken for AEs associated with dasatinib in NSCLC. One phase II trial of dasatinib for lung SCCAs had to be halted due to intolerable pleural effusions and other severe toxicities [Brunner et al. 2013]. Novel and safer DDR2 inhibitors are warranted.
FGFR genomic aberrations
The FGFR family includes four receptor tyrosine kinases (FGFR1, FGFR2, FGFR3, FGFR4). FGFR1 amplification occurs in up to 10–20% of lung SCCAs [Dutt et al. 2011], but true high-level amplification that confers oncogene addiction may only occur rarely (<1–3% of tumors). FGFR mutations and rearrangements are also seen in less than 5% of SCCAs and can confer sensitivity to FGFR TKIs in preclinical models [Kim et al. 2014]. Many currently available and in development multitargeted TKIs demonstrate inhibitory effects on FGFR, as well as VEGFR, PDGFR, FLT3, RET, and KIT. Multiple studies of FGFR inhibitors are underway (AZD4547, JNJ-42756493, BGJ398, ponatinib) and with as yet uncertain benefits [Tucker et al. 2014]. Class effects of FGFR-specific inhibitors are thought to include hyperphosphatemia and tissue calcification. Clinical trial development will help identify if any of the SCCA-associated FGFR genomic aberrations is truly oncogenic and can be targeted with FGFR TKIs.
Conclusions and future directions
The approach to treatment for advanced NSCLC has been revolutionized in the past decade with the arrival of biologically relevant and clinically effective and tolerable targeted therapies. Application of EGFR- and ALK-directed therapies in appropriately selected molecular subgroups has afforded consistently robust RRs, improvements in PFS, reduced toxicity, and improved QoL compared with conventional cytotoxic agents. Resultant clonal selection with development of acquired resistance to these agents has become an inevitable reality. However, evolving insights into mechanisms of resistance and identification of promising emerging therapies, either advanced generation TKIs or combination strategies affording multipronged antagonism of neoplastic pathways, offers hope for extending the spectrum of benefit of such agents.
Stemming from the experience with oral EGFR and ALK inhibitors, use of multitargeted TKIs which are already approved for other indications has shown promise in patients with mutations in ROS1, BRAF, MET, RET, and ERRB2. Additional investigations are underway to delineate optimal use of targeted therapies in these subgroups.
Targeting KRAS and identification and effective blockade of oncogenic drivers in lung SCCAs remain two distinct areas of continued unmet need.
Given the current emphasis on personalized medicine (i.e. optimally matching patients with the most effective, least toxic therapies) and exponential growth in the feasibility, accessibility, and application of molecular profiling technologies, identification of additional biologically and therapeutically relevant targets remains an important priority in the management of patients with this lethal disease.
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded in part by the Lung Cancer Foundation of America International Association for the Study of Lung Cancer (grant to DBC), the American Cancer Society (grant number RSG 11-186 to DBC), and the National Cancer Institute (grant number CA090578 to DBC).
Conflict of interest statement
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: DBC has received consulting fees and honoraria from Pfizer Inc. and Boehringer Ingelheim, respectively. DBC also conducts unremunerated clinical trials using osimertinib (AstraZeneca) and rociletinib (Clovis Oncology).