ALK, ROS1 and RET rearrangements in lung squamous cell carcinoma are very rare

Highlights

• Revised the manuscript with help of two native English writers.

• Explained the design and the location of 3′/5′ expression probe sets based on known fusion junction sites.

• Clarified the methodology of NanoString fusion panel on detecting novel fusion variants.

• Clarified data analysis and threshold determination of NanoString assay.

• Modified the formatting of throughout manuscript.

Abstract

Objectives

Chromosomal rearrangements of ALK and ROS1 genes in non-small cell lung carcinoma (NSCLC) define a molecular subgroup of lung adenocarcinoma (ADC) that is amenable to targeted therapy with tyrosine kinase inhibitors (TKIs) crizotinib. Emerging clinical studies have demonstrated that patients with RET-rearranged NSCLC may also benefit from existing RET TKIs, including cabozantinib and vandetanib. However, the reported cases of lung squamous cell carcinomas (SCC) harboring gene rearrangements have been detected via fluorescence in situ hybridization (FISH) or immunohistochemistry (IHC) from materials such as biopsy or resection. Fusion events identified in lung SCC raise the question of whether this histologic subtype should also be evaluated for merit molecular testing. This work was undertaken to study the prevalence of lung SCC harboring ALK, ROS1, and RET translocations.

Materials and methods

Squamous cell carcinomas were confirmed using both histological examination by pathologists and immunohistochemistry analysis with positive staining of P63 and CK5/6 combined with negative CK7 and TTF-1 staining. 214 samples from surgically resected patient tissues were used to search for ALK, ROS1, and RET rearrangements by a NanoString analysis method. Fusion events were detected in a single-tube, multiplex assay system that relied on a complementary strategy of interrogation of 3′ gene overexpression and detection of specific fusion transcript variants.

Results and conclusion

ALK, ROS1 or RET gene rearrangements appeared 0 times out of 214 cases of lung SCC. Our data revealed that these fusions may be very rare in lung squamous cancer. The molecular screening strategy should therefore be focused on lung adenocarcinoma as the current National Comprehensive Cancer Network (NCCN) guideline recommends.

Introduction

Lung cancer remains the leading cause of cancer-related mortalities worldwide. Non-small cell lung cancer accounts for over 85% of all lung cancer cases and is associated with 5-year survival rates of 15% [1], [2]. Adenocarcinoma and squamous cell carcinomas are the most frequent histological subtypes, occurring in 50% and 30% of NSCLC cases, respectively [3]. In the last decades, encouraging new treatments [i.e., the tyrosine kinase inhibitors (TKIs) of EGFR and ALK] have afforded benefits to patients with adenocarcinoma, but few advances have been made in the treatment of SCC [3]. Since many genomic abnormalities have been identified in SCC, there is a growing biological significance to deciphering the molecular characterization of patients with SCC in modern profiling platforms. Therefore, it is essential to characterize the biological relevance and the true frequency of each alteration in SCC.

ALK rearrangements in NSCLC define a molecular subgroup of lung adenocarcinoma that is highly sensitive to targeted crizotinib treatment [1], which is the only agent approved for ALK-rearranged NSCLC by China Food and Drug Administration (CFDA). Activity against ALK-rearranged NSCLC has also been described in second-generation ALK inhibitors such as ceritinib and alectinib, which can potentially overcome the resistance to the treatment of crizotinib in patients with ALK positive, metastatic NSCLC. In NSCLC, ALK fusions have been reported to occur at a frequency of 2–5% in the unselected NSCLC population [4], [5], and are more prevalent in younger patients with adenocarcinoma who have never smoked or are light smokers. ROS1 (c-ros oncogene 1) and RET (ret proto-oncogene) fusions were identified at a small frequency (∼1–2%) of lung cancer cases, using a variety of genotyping techniques [6], [7], [8], [9]. Similar to the ALK fusions, the ROS1 fusions can lead to constitutive kinase activity and are associated with sensitivity to TKIs such as crizotinib [10], which has been shown as an effective ROS1 inhibitor, superior to chemotherapy. RET fusions are potential therapeutic targets of existing multi-targeted kinase inhibitors, including cabozantinib, vandetanib, sunitinib and sorafenib [8], [9], [11]. Notably, ROS1 and RET rearrangements are mutually exclusive with other genetic alterations, such as EGFR, KRAS or ALK.

It is a common belief that true ALK, ROS1, and RET translocations in SCC are very rare [12], [13]. The estimated prevalence of ALK gene fusion is ∼1% in lung SCC [3], [14], [15]. Recent molecular testing guidelines conducted a meta-analysis among lung cancer patients in SCC in which only 1 of 523 (0.2%) was found to carry an ALK rearrangement [16], [17], [18], [19], [20], [21], [22]. The data are also supported by other studies specifically reporting the frequency of 0% ALK translocation in SCC patients [23], [24], [25]. Since the current practice guidelines recommend a molecular testing for lung adenocarcinoma or mixed lung cancer with an ADC component to select patients for targeted TKI therapy, reported cases of lung squamous cancer harboring gene rearrangements challenge the molecular diagnosis based on histologic subtypes [12], [15], [26], [27], [28], [29], [30]. However, doubts have been cast on reports showing ALK and/or ROS1 rearrangements in squamous lung cancer, as the small biopsy specimens were not representative of the whole tumor characteristics, and an immunohistochemistry panel was not available to validate the squamous differentiation. Moreover, Davis et al. recently reported two ROS1-positive cases in SCC and claimed that the proportion of SCC found may not reflect that of the general population of patients in NSCLC [12].

Molecular diagnostic algorithms have undergone a significant evolution over time, moving from a “one-gene, one-test” paradigm to the inclusion of multiplex assays for common hotspot mutations, gene rearrangements, and copy number changes in relevant lung cancer genes. However, current clinically validated methods for detection of ALK, ROS1, and RET fusions are characterized by one single gene detection, such as fluorescence in situ hybridization and immunohistochemistry. Due to the low frequency of these fusions and the high incidence of NSCLC, current methods also have their own limitations in terms of cost, throughput, and required expertise necessary for interpreting these tests. We previously designed a single-tube multiplexed assay to simultaneously detect fusion transcripts of ALK, ROS1, and RET genes and to measure the imbalanced 3′/5′ expression levels without prior knowledge of the fusion partners [31]. For the gene expression assay, 4 probes, which were designed to measure the expression levels of target genes, are placed upstream and 4 probes are placed downstream of the fusion junction. The fusion events are complemented by the fusion-specific probe sets to detect the common and reported fusion variants. Advances of the NanoString fusion gene panel have provided a means of interrogating a variety of therapeutically relevant fusions in a single tube, and have also been validated in 100% concordance with FISH, the only FDA-approved “gold standard” method for fusion gene detection.

Using NanoString fusion panel, Fang et al. identified two ALK-rearranged SCC xenograft models, one carrying the well-known EML4-ALK variants 3a/b and the other harboring a novel huntingtin interacting protein 1 (HIP1)–ALK fusion gene [32]. Both were diagnosed as moderately differentiated SCC by a pathologist, and then further validated by an immunohistochemistry panel positive for CK5/6, p63, 34βE12 and negative for MOC31 and BerEP4. Given that these rare fusions predominantly exist in lung adenocarcinoma, the reported cases of these gene fusions identified in SCC raise the question of whether this histologic subtype should be evaluated for the rare fusions’ molecular tests. In order to determine the prevalence of the ALK, ROS1, and RET fusions in lung cancer resected specimens with squamous cell carcinoma, we screened a total of 214 cases of SCC samples by NanoString fusion assay, combining two cohorts of surgically resected samples collected from Korean and Chinese patients.