- RET rearrangements occur in 1% to 2% of patients with non-selected NSCLC and in approximately 16% of NSCLC tumors that lack other oncogenic drivers (e.g., EGFR, ALK, and KRAS). RET rearrangements are not found in squamous cell carcinoma or in small cell lung cancer. RET can partner with different genes in NSCLC, and KIF5B is the most common fusion partner in NSCLC.
- Early-phase clinical trials to evaluate multikinase inhibitors of RET (e.g., cabozantinib, vandetanib, and lenvatinib) have demonstrated moderate antitumor activity and a tolerable toxicity profile in patients with RET-rearranged NSCLC.
- The identification of alternate signaling pathways that may mediate resistance to RET inhibition will likely lead to the development of more effective therapeutic strategies in patients with RET-rearranged NSCLC.
The RET (rearranged during transfection) proto-oncogene was discovered in 1985 by Takahashi et al.,1 who reported a novel gene rearrangement during the transfection of NIH 3T3 cells with human lymphoma DNA. Shortly thereafter, RET was localized to human chromosome 10q11.2 and shown to encode a single-pass transmembrane receptor tyrosine kinase termed RET.2 Studies examining the spatial and temporal expression of RET determined that the protein was expressed predominantly on neural crest-derived and urogenital cells and that expression levels were greatest during development and least in normal adult tissues.3 Functional studies in transgenic mice demonstrated that RET is required for the development of the enteric nervous system, kidney morphogenesis, and for spermatogenesis.4 The RET protein also has been shown to function as a key regulator of Peyer’s patch organogenesis.5
The mature form of the RET receptor tyrosine kinase has a mass of approximately 170 kDa and is composed of a large extracellular domain, a transmembrane region, and an intracellular kinase domain.4 RET is activated when the receptor binds to a multimeric protein complex consisting of growth factors from the glial cell line–derived neurotrophic factor family and their ligand-binding co-receptor, the glial cell line-derived neurotrophic factor (GDNF) family receptor alpha (GFRa).6 Activation of RET leads to autophosphorylation on intracellular tyrosine residues and initiation of Ras/MAP kinase, PI3K/AKT, and phospholipase C pathways that signal cell proliferation, migration, and differentiation.7 Germline gain-of-function mutations of RET lead to multiple endocrine neoplasia type 2, whereas somatic gain-of-function mutations lead to sporadic medullary thyroid carcinoma. Somatic RET rearrangements induce the formation of RET fusion protein kinases that localize in the cytosol and have transforming and oncogenic properties.7 This editorial discusses the role of RET rearrangements in non–small cell lung cancer (NSCLC) and the latest advances in RET fusion–targeting approaches.
RET Rearrangements in NSCLC
Fusion proteins resulting from the chromosomal rearrangement of RET were identified first in papillary thyroid carcinoma (PTC).8,9 Thirteen different oncogenic RET fusions have been identified in PTC, and each chimera is produced by a distinct chromosomal translocation event in which the promoter and the 5-prime region of a heterologous gene that encodes a thyrocyte-expressed protein are fused, in frame, to the kinase-encoding 3-prime end of the RET proto-oncogene.10 Between 2011 and 2012, four independent groups of investigators identified RET fusions in NSCLC.11-14 Collectively, those studies concluded that RET fusions occurs in approximately 1% to 2% of NSCLCs and that RET rearrangements are mutually exclusive of mutations in EGFR, KRAS, ALK, HER2, and BRAF. The 3T3 cells that were infected with a retrovirus encoding a rearranged RET fusion protein produced tumors in mice, which confirmed the role of RET as an oncogenic kinase.11
To date, RET has been shown to partner with eight different genes in NSCLC: KIF5B, CCDC6, NCOA, TRIMM33, CUX1, KIAA1468, KIAA1217, and FRMD4A.15-17KIF5B is the most common fusion partner in NSCLC, and a total of seven KIF5B-RET variants have been identified in NSCLC.18 Reports investigating the clinical parameters and the molecular and genetic alterations in NSCLC indicate that patients with RET-positive tumors have identifiable clinico-pathologic characteristics. Wang et al.19 examined 936 occurrences of surgically resected NSCLC and detected RET fusions in 13 tumors (1.4%), nine of which had a KIF5B-RET fusion, three of which had a CCDC6-RET fusion, and one of which had a NCOA4-RET fusion. Patients with RET-rearranged lung adenocarcinomas (LUADs) had more poorly differentiated tumors (64%), were overall younger (≤ 60 years; 73%), were never-smokers (82%), and had a solid subtype (64%) and a smaller tumor (≤ 3 cm) with N2 disease (54%). Other independent investigators have shown that RET-rearranged LUADs possess histologic features of a solid growth pattern that contains signet ring cells and a mucinous cribriform pattern that is similar to ALK- and ROS1-mutant LUADs.19,20RET fusions do not occur in squamous cell carcinomas21 or in small cell lung cancers.13 However, the prevalence of RET rearrangement in LUADs increases to approximately 16% in patients whose tumors lack other oncogenic drivers (EGFR negative, ALK negative, or KRAS negative).22
Detection of RET Fusions in NSCLC: Recent Advances and Challenges
The identification of RET rearrangements has been incorporated into diagnostic assays to inform and guide the clinical decision-making process. The studies that led to the discovery of RET rearrangements in NSCLC, as well as subsequent studies to evaluate RET status in NSCLC,23-28 have used a variety of detection and validation approaches (Table 1). At present, there is no gold-standard method for the identification of RET rearrangements.18 Reverse transcriptase polymerase chain reaction (RT-PCR) is both sensitive and specific for the detection of known fusions, but it is not reliable for the detection of new fusion partners or isoforms. The utility of immunohistochemistry (IHC) for the detection of RET fusions has been limited because of variable staining patterns and weak reactivity.18 Although FISH is an effective technique for the detection of RET fusions, its high cost and need for technical expertise limit its utility.29
Currently, our group is conducting a phase I escalation study to evaluate the maximum-tolerated dose, the recommended phase II dose, the safety, and the antitumor activity of vandetanib in combination with everolimus, an mTOR inhibitor, in advanced solid tumors, including stage IV NSCLC.30RET rearrangement was tested using FISH and/or next-generation sequencing (NGS) on tumor tissue and/or on plasma-extracted cell-free DNA (cfDNA), also known as circulating tumor DNA. To date, our preliminary results indicate that the concordance between FISH and NGS in detecting RET rearrangements is 40% (unpublished results). Because RET rearrangements occur in only 1% to 2% of non-selected NSCLCs, it is important to consider the costs and the reliability of detection tests, especially when tumor tissue samples are used for analysis. NGS-based methods offer a simultaneous assessment of multiple genomic aberrations and may be regarded as the preferred approach for the determination of tumor molecular status.29
Targeting RET Fusion in NSCLC: Clinical Trial Results
RET fusion kinases are oncogenic and represent an actionable target. Not surprisingly, much effort is directed currently toward the development of therapies that can inhibit aberrant RET signaling. A summary of the results of the studies investigating the safety and activity of single-agent RET inhibitors in patients with RET-rearranged NSCLC is listed in Table 2.
Drilon et al.31 recently reported the results of a prospective, single-arm phase II study to evaluate the activity of cabozantinib, an oral multikinase inhibitor of RET, in 26 patients with RET-positive LUADs, as determined by FISH and/or NGS analysis. The objective response rate (ORR) in the 25 patients whose results were assessable for response was 28% (95% CI [12%, 49%]) and included seven patients with partial responses. The most common grade 3 treatment-related adverse events (AEs) were lipase elevation (15%), increased alanine aminotransferase (8%), increased aspartate aminotransferase (8%), thrombocytopenia (8%), and hypophosphatemia (8%). The median progression-free survival (PFS) was 5.5 months (95% CI [3.8, 8.4]), and the median overall survival (OS) was 9.9 months (95% CI [8.1, not reached]). Of note, six patients who had not received prior chemotherapy had longer PFS.
Lee et al.32 evaluated the efficacy and safety of vandetanib, an oral RET, VEGFR-2, and EGFR kinase inhibitor, in patients with advanced/refractory RET-rearranged NSCLC. Among the 17 patients with evaluable results, the ORR was 18% (three patients with partial responses), and the disease control rate (DCR) was 65% (eight patients with stable disease). Patients treated with vandetanib had a PFS of 4.5 months and an OS of 11.6 months. The 1-year OS rate was 33%, and 10 of 18 patients (56%) had died at the data cutoff. Overall, the treatment was well tolerated.
Yoh et al.33 conducted the phase II LURET study, in which 1,536 patients with EGFR-negative NSCLC were screened by multiplex transcriptase PCR and FISH break-apart assay. Among the patients who were screened, 34 (2%) were RET positive, and 19 were enrolled in the study and treated with 300 mg of vandetanib daily. Among 17 patients with evaluable data included in primary analysis, the ORR was 53% (95% CI [28, 77]), and the median PFS was 4.7 months (95% CI [2.8, 8.5]). The OS rate at 12 months was 47% (95% CI [20, 69]), and the median OS was 11.1 months. The most common grades 3 to 4 AEs were hypertension (58%), diarrhea (11%), skin rash (16%), and QTc prolongation (11%).
Alectinib, a tyrosine-kinase inhibitor of the anaplastic lymphoma kinase (ALK) that is also active against RET in vitro35 has demonstrated antitumor activity in patients with advanced RET-rearranged NSCLC.36 In this report, the investigators administered alectinib, as part of single-patient compassionate or off-label use, to four patients at a dose of 600 mg twice daily or 900 mg twice daily. Three of the patients had previously been treated with other RET inhibitors. Alectinib produced an objective radiographic response in two patients, including clinical improvement of brain metastases in one patient. In another patient, alectinib-derived clinical benefit was observed with a stable disease lasting approximately 6 weeks. The fourth patient progressed on treatment.36
A phase II multicenter trial to evaluate the safety and efficacy of the multikinase inhibitor ponatinib in patients with advanced RET–rearranged NSCLC is underway (NCT01813734).
The results of the above mentioned studies suggest that different multikinase inhibitors may produce differential responses depending on the type of RET fusion. For example, Drilon et al.31 observed responses only in KIF5B-RET tumors and observed no responses in tumors with other fusion variants; Yoh et al.33 noted that among the six patients whose tumors harbored a CCDC6-RET fusion, five (83%) achieved an objective response to vandetanib compared with two responses (20%) among the 10 patients whose tumors harbored a KIF5B-RET fusion.
Mechanisms of RET-Inhibitor Resistance and Novel Therapeutic Approaches
The multikinase inhibitors of RET are active in patients with RET-rearranged NSCLC; however, tumors eventually become refractory to therapy. Resistant cancer cells may overexpress drug efflux transporters or may harbor drug-resistance mutations. Huang and colleagues37 recently identified the cabozantinib-resistant KIF5B-RETV804L and the vandetanib-resistant KIF5B-RETG810A mutations in lung adenocarcinoma cells. These investigators determined that, among cabozantinib, lenvatinib, ponatinib, and vandetanib, ponatinib was the most potent inhibitor against KIF5B-RET and its drug-resistant mutants. Furthermore, the authors showed that the vandetanib-resistant KIF5B-RETG810A mutant cells displayed gain-of-sensitivity to ponatinib and lenvatinib, suggesting that the different RET inhibitors can overcome vandetanib-induced mechanisms of resistance.37 Cancer cells may also activate alternative signaling pathways to support their growth. Vitagliano et al.38 demonstrated that activation of the EGFR pathway in RET-mutant thyroid cell lines treated with vandetanib rescued cell proliferation and signaling through MAPK, indicating that dual RET/EGFR blockade may overcome RET-inhibitor resistance induced by compensatory activation of EGFR pathway in these cancer cell lines. A recent study of lung cancer cells that harbored CCDC6-RET genes and that grew in culture suggested that activation of EGFR signaling may allow the cells to become resistant to RET inhibition via bypass survival signaling through ERK and AKT.39 In preclinical models of RET-mutant thyroid cancer, the combination of vandetanib with the mTOR inhibitor everolimus demonstrated greater antitumor activity than either single agent alone.40
We are conducting a phase I escalation study to evaluate the maximum-tolerated dose, the recommended phase II dose, the safety, and the antitumor activity of vandetanib in combination with everolimus in patients who have refractory solid tumors, including those with RET-rearranged NSCLC (NCT01582191). Preliminary results of the study indicate that the combination of vandetanib and everolimus is well tolerated. The most common treatment-related AEs are diarrhea, fatigue, mucositis, and rash.41 The combination produced significant antitumor activity in patients with RET-rearranged NSCLC, as determined by NGS (five patients with partial responses; six patients who had RET-positive disease and had evaluable results).41 However, in patients with NSCLC whose results were RET positive by FISH, but RET negative by NGS, no objective responses were observed. Antitumor activity of the combination was observed in patients with RET-positive NSCLC brain metastases and in cabozantinib-refractory disease.41 This study is active and recruiting patients; updated results about a larger cohort of patients with RET-rearranged NSCLC are pending.
Improving Patient Selection for Targeting RET-Rearranged NSCLC
Patients with RET-rearranged NSCLC are a small but targetable subgroup of patients in whom the administration of RET multikinase inhibitors has shown antitumor activity and an overall acceptable toxicity profile. The identification of biomarkers to select patients with NSCLC who will derive clinical benefit from therapeutic approaches that target RET is an active area of investigation. Similarly, the identification of potential mechanisms that mediate primary and acquired resistance to RET inhibitors will lead to the development of more effective therapeutic strategies. It remains to be seen whether more selective RET inhibitors and combination strategies will have greater antitumor activity and efficacy than multikinase inhibitors.
About the Authors: Dr. Cascone is a third-year hematology/oncology fellow in the Division of Cancer Medicine, with The University of Texas MD Anderson Cancer Center. Dr. Subbiah is an assistant professor in the Department of Investigational Cancer Therapeutics, Division of Cancer Medicine and Division of Pediatrics, with The University of Texas MD Anderson Cancer Center. Dr. Heymach is a professor and chair of the Department of Thoracic, Head and Neck Medical Oncology with The University of Texas MD Anderson Cancer Center.