Patients with ALK-rearranged lung carcinomas tend to be young non-smokers presenting with advanced disease [9–11]; however, use of clinical or demographic criteria to select patients for ALK-targeted therapy or testing is not recommended [8]. FISH is generally considered the gold standard for detection of ALK rearrangements in lung ACA, as a result of its use in the selection of patients for crizotinib therapy in the original clinical trials. The FDA has approved the ALK Break Apart FISH Probe Kit (Abbott Molecular, Des Plaines, IL) as a companion diagnostic for use of this drug in lung cancers. ALK fusion events across human tumors lead to increased ALK transcription and protein expression by immunohistochemistry (IHC); however, the level of protein expression varies according to tumor type. Commercially available IHC antibodies optimized for use in lung cancer include clones 5A4 and D5F3. The FDA has approved an antibody kit (Ventana ALK (D5F3) CDx Assay, Roche Diagnostics, Indianapolis, IN) for use on an automated staining platform for selection of patients with crizotinib therapy. However, neither FISH nor IHC has demonstrated perfect sensitivity or specificity and may be used in a complementary fashion in practice, especially in the context of unexpected or discordant results. Molecular methods for rearrangement detection including anchored multiplex PCR and hybrid capture/next-generation sequencing are employed in some settings [12, 13]; DNA-based sequencing approaches in particular enable detection of ALK rearrangements from liquid-based specimens such as plasma [14].

With few exceptions, most published studies demonstrate that ALK rearrangements and other oncogenic driver mutations occur in a mutually exclusive fashion [15]. On occasions where both ALK rearrangement and another driver oncogenic mutation are detected, multimodality evaluation may help to clarify false-positive results [16, 17]. Large series suggest that ALK and EGFR alterations may occur concomitantly in ~1% of selected populations, but these dually altered tumors are differentially responsive to EGFR and/or ALK inhibition; EGFR and ALK phosphorylation status may help guide selection of an appropriate inhibitor [18].

Resistance to the ALK inhibitor crizotinib emerges almost inevitably within 1–2 years of initiating therapy. The mechanisms of resistance include secondary mutations in the ALK tyrosine kinase domain, fusion gene amplification, and upregulation of bypass signaling pathways (Table 12.2) [19]. Alternative, ALK-independent survival pathways that can hamper the effectiveness of crizotinib include the epidermal growth factor pathway, insulin-like growth factor pathway, RAS/SRC signaling, and AKT/mTOR signaling, among others [20]. Importantly, the type of resistance mechanism often dictates the efficacy of subsequent lines of therapy. Dual ALK and EGFR inhibition may be active against crizotinib-resistant tumor cells driven by EGFR pathway activation, whereas combined ALK and KIT inhibition may overcome KIT-amplification driven resistance [21]. Despite the diversity of resistance mechanisms, most crizotinib-resistant tumors continue to depend on ALK signaling and are sensitive to more potent, structurally distinct, second-generation ALK inhibitors, such as ceritinib , alectinib , brigatinib , and lorlatinib , which may be effective in the context of acquired resistance mutations arising in the ALK tyrosine kinase domain [19]. Certain mutations appear to arise preferentially following use of specific first- and second-generation inhibitors, and multiple resistance mechanisms may be detected in an individual tumor (Table 12.2). The location of an ALK kinase domain resistance mutation may have significant implications for ALK-targeted therapy. Gatekeeper mutations that influence the kinase-ATP interaction occur at codon 1196 and are analogous to the T790M mutation in EGFR-mutated lung adenocarcinomas and ABL T315I mutations in BCR-ABL in chronic myelogenous leukemia [22]. For ALK, the effects of this gatekeeper mutation may be overcome by using potent inhibitors. In contrast, mutations arising at the solvent front (codon 1202) appear to lead to steric hindrance of and resistance to most of the available ALK inhibitors, with the exception of the highly selective third-generation inhibitor, lorlatinib [19]. Given the differential patterns of resistance and unique sensitivities of individual inhibitors, some authors have argued for routine biopsies at the time of relapse on ALK-targeted therapies [19, 23, 24]. This approach has not been adopted in routine practice, however, and has not yet been endorsed in testing guidelines [in preparation]. In the context of crizotinib resistance, outcomes following treatment with second-generation ALK inhibitors are promising. Brain metastases are common in ALK-rearranged lung carcinomas [25], and some second-generation inhibitors, including alectinib , effectively penetrate the blood-brain barrier and demonstrate excellent activity in the central nervous system. In phases I and II trials of alectinib in the crizotinib-resistance setting, about half of patients responded, but interestingly tumor shrinkage did not correlate with survival outcomes, with 78% of patients still alive after 3 years on therapy [26].

ROS1 is a transmembrane receptor tyrosine kinase similar in structure to ALK, consisting of an extracellular ligand-binding domain, a short transmembrane domain, and an intracellular tyrosine kinase domain. While its extracellular ligand remains unknown, it is thought to function like other receptor tyrosine kinases by intracellular tyrosine phosphorylation, resulting in activation of downstream PI3K, STAT3, and RAS/MAPK signaling with effects on cell proliferation, survival, and cell cycling [27]. ROS1 fusion events have oncogenic activity in a variety of tumor types including glioblastoma, cholangiocarcinoma, inflammatory myofibroblastic tumor, and lung adenocarcinoma. Using FISH- and IHC-based screening programs, ROS1 rearrangements involving a variety of fusion partners including CD74, SLC34A2, SDC4, EZR, and Fig1 have been reported in approximately 1–2% of lung adenocarcinoma [28, 29]. Similar to ALK rearrangement, these fusions result in the placement of the ROS1 kinase domain downstream of a coiled-coil domain of the 5′ fusion partner, although activation of ROS1 fusion signaling does not appear to involve dimerization [27]. Crizotinib, the same multitargeted inhibitor effective in ALK-rearranged tumors, has been shown to be effective in treating ROS1-rearranged lung adenocarcinomas. In a phase 1 trial, 72% of patients with ROS1-rearranged lung carcinomas responded to crizotinib therapy [30]. A retrospective analysis in a European cohort confirmed this robust response pattern (Table 12.3) [31]. ROS1-rearrangement also appears to be a positive predictor of response to pemetrexed-based chemotherapy regimens [32].

As with ALK, ROS1 rearrangements are significantly more common in young never-smokers whose tumors lack other oncogenic driver mutations [33]. Although focused testing of cohorts containing patients fitting these characteristics will enrich for ROS1-rearranged tumors, tumors from older patients and smokers may also harbor these alterations; therefore, selection of patients for testing or treatment based on clinical or demographic features is discouraged.

In contrast to ALK, there is no companion diagnostic required for use of crizotinib therapy in patients with ROS1-rearranged tumors. FISH is often considered the gold standard for detection of ROS1 fusions given a heavy reliance on this technology in the original clinical trial of crizotinib for ROS1-rearranged lung adenocarcinomas [30]. However, in light of the rarity of ROS1 rearrangements in lung cancers, more economical and less technically demanding screening approaches may be preferable to FISH for many laboratories. ROS1 immunohistochemistry using the commercially available D4D6 clone has a sensitivity for detection of ROS1-rearranged lung adenocarcinomas approaching 100% according to most studies, with variable specificity (pooled estimate of 93%) (guidelines, in preparation). The lower specificity results from occasional low-level expression of ROS1 protein in lung tumors with other known oncogenic drivers as well as in benign reactive pneumocyte proliferations [34, 35]. As a result, use of FISH or molecular methods is recommended to confirm a positive immunohistochemical result. Targeted real-time PCR, anchored multiplex PCR, and hybrid capture/next-generation sequencing have all been reported as parallel or stand-alone testing approaches for ROS1 fusion detection, with good concordance with FISH and IHC [36].

ROS1 rearrangements generally occur in a mutually exclusive fashion with other driving molecular alterations; however, rare instances of combined ROS1 fusion and EGFR or KRAS mutations have been reported [37]. The clinical significance of these combined alterations, including outcomes from ROS1-targeted therapy, is currently unknown. In most cases, multimodality testing including specific sequencing-based methods can identify falsely positive ROS1 findings by FISH or IHC.

The RET proto-oncogene encodes a receptor tyrosine kinase with an extracellular domain, a transmembrane domain, and an intracellular tyrosine kinase domain. Binding of GDNF ligands to its GPI-linked co-receptor (GFRα) causes homodimerization of RET and autophosphorylation of its intracellular tyrosine residues and downstream RAS/MAPK, PI3K, and STAT pathway activation [38]. Several 5′ partners are involved in RET fusion events, including KIF5B, CCDC6, and NCOA4. Similar to ALK, the breakpoints in RET (at exons 11 or 12) unite its tyrosine kinase domain with the coiled-coil domain of its upstream partner, allowing for ligand-independent homodimerization and downstream signaling [29]. Vandetanib , a multi-specific tyrosine kinase inhibitor, including VEGFR-2, VEGFR-3, EGFR, and RET, was able to inhibit RET fusion-bearing tumor cell growth in vitro [29]. In phase 2 trials, patients with RET-rearranged lung adenocarcinomas show partial responses or disease stabilization with the MET and VEGFR2 inhibitor, cabozantinib [39, 40], as well as favorable responses to pemetrexed-based chemotherapies [41].

The relative rarity of RET-rearranged lung adenocarcinoma and obstacles to widespread screening have led to relatively limited literature on the clinicopathologic features of this tumor type; however, RET rearrangements appear to occur more commonly in lung cancer patients with a smoking history as compared to ALK and ROS1 cohorts [42]. As in thyroid carcinomas, RET rearrangements may occur more commonly in patients with a prior history of locoregional radiation therapy [43]. RET rearrangement is detectable by FISH; however, the most commonly described alteration, a small intrachromosomal inversion on chromosome 10, leading to the KIF5B-RET fusion, leads to a subtle split in the FISH probe signals that can be difficult to consistently detect in practice. Immunohistochemistry-based RET protein detection appears robust in other tumor types with RET alterations [44], but reports on the use of RET IHC in lung adenocarcinoma are limited, with variable sensitivity and specificity for RET rearrangements using different commercially available antibodies [45–47]. Targeted RT-PCR and next-generation sequencing methods can be used to detect RET fusions. While the current clinical outcome data is too limited to support routine stand-alone testing for RET rearrangement detection, laboratories that are implementing multiplexed RNA- or DNA-based sequencing assays should incorporate RET rearrangement testing into their testing platform [AMP/IASLC/CAP guidelines in preparation].