Abstract
Introduction
Gene rearrangements are frequent oncologic drivers in NSCLC, and many are suitable for treatment with Food and Drug Administration–approved or experimental targeted therapies. We evaluated the accuracy, specimen acceptance profile, and limits of detection of a rapid fusion assay (Idylla GeneFusion Assay), a commercially available ultrarapid molecular assay, for its clinical utility.
Methods
A collection of 97 specimens which had previously undergone next-generation sequencing testing were analyzed using the rapid fusion assay. Accuracy was evaluated by sensitivity and specificity compared with the next-generation sequencing results. The performance characteristics were tested by using a variety of different clinically relevant specimen types. Limits of detection were assessed by evaluating different input of tumor percentage and material amount.
Results
The rapid fusion assay was found to have 100% sensitivity in detecting fusions of ALK, ROS1, RET, NTRK1, and MET exon 14 skipping and 83% sensitivity for NTRK2/3 fusions. There were 100% specificity in detecting fusions of ROS1, RET, NTRK2/3, and MET exon 14 skipping and 98% specificity for ALK. Testing was successful with formalin-fixed paraffin-embedded biopsy and surgical tissues, cell blocks from fine-needle aspiration and pleural fluid (down to 5% tumor content, 18 mm2 tissue scraped), cytology smears (≥300 cells), and previously extracted RNA (minimal 20 ng).
Conclusions
The rapid fusion assay is quick, accurate, and versatile, allowing reliable detection of ALK, ROS1, RET fusions, and MET exon 14 skipping in NSCLC, and NTRK fusions. Rapid molecular testing may expedite treatment with appropriate targeted therapies.
Keywords: Gene fusions, Rapid testing, Non–small cell lung cancer, Targeted therapy, Idylla
Introduction
NSCLC represents the largest category of lung cancers, with adenocarcinoma being the most common.1,2 The discovery of oncologic drivers and development of targeted therapies have substantially improved patient outcomes.1,3 Several gene rearrangements and splicing alterations involving tyrosine kinase receptors are suitable for treatment with targeted therapies.4, 5, 6, 7 Indeed, multiple small molecules have been approved by the Food and Drug Administration targeting ALK (approximately 4% of NSCLC), ROS1 (approximately 2%), RET (approximately 2%), and NTRK (approximately 0.1%) rearrangements, and MET exon 14 skipping (approximately 3%).1,8 Overall, these molecular alterations account for approximately 11% of potentially targetable NSCLC. As targeted therapies are primarily approved for the treatment of patients with metastatic disease, the rapid detection of these alterations can potentially expedite treatment decision making for patients who have symptoms and are in need of the most effective therapy.
Accuracy and feasibility of detecting such alterations has been challenging. Routine testing by multiple standalone single-gene assays, such as immunohistochemistry and fluorescence in situ hybridization (FISH), is not recommended owing to concern for tissue exhaustion. Next-generation sequencing (NGS) panels, ideally including RNA-based component, are generally preferred because they may assess the presence of all relevant mutations and are more cost effective than multiple single-gene tests.9, 10, 11, 12 Nevertheless, the long turnaround time (2–4 wk) dampens their clinical value in scenarios where timely decisions for patient management must be made. In these instances, a rapid and accurate assay with minimal tissue requirement is strongly desired.
Beta testing reports of an ultrarapid gene fusion assay, using mainly prototype cartridges, revealed encouraging accuracy results.13, 14, 15 In the current study, we evaluated the clinical utility of this recently commercially available assay, using the manufactured kits. We tested its performance of accuracy, tissue acceptance profile, and limit of detection for capturing clinically relevant gene fusions involving ALK, ROS1, RET, NTRK1/2/3 genes, and MET exon 14 skipping alterations.
Materials and Methods
This study was deemed exempt by the Mayo Clinic Institutional Review Board, and patient’s informed consent was obtained.
Assay
The Idylla GeneFusion Assay (Biocartis, Mechelen, Belgium) is a rapid and automated cartridge-based system optimized for formalin-fixed, paraffin-embedded (FFPE) tumor tissue.13,14 RNA extraction, amplification, multiplex quantitative reverse-transcriptase polymerase chain reaction (RT-qPCR), and data analysis are performed within a single cartridge-based workflow. It detects fusions by the following two methods: (1) fusion-specific (FS) detection and (2) expression imbalance (EI) analysis. FS detection includes RT-qPCR primers designed specifically for 17 ALK, 13 ROS1, and seven RET rearrangements, and primers designed for the MET exon 13-exon 15 junction. The EI method detects fusion events by analyzing expression difference between the 3′ (kinase domain) and 5′ ends of the mRNA of ALK, ROS1, RET, and NTRK1/2/3.
Samples
A total of 97 tumor samples from 87 unique patients with prior clinical NGS profiling were included. There were 64 cases with NSCLC, including 44 samples with either ALK, ROS1, or RET fusion, or MET exon 14 skipping alterations. There were 20 NSCLC samples negative for gene rearrangements. An additional 33 samples had a fusion involving NTRK1/2/3, predominantly representing sarcomas and central nervous system tumors. Accuracy was evaluated for each individual gene tested, compared with the NGS results. The reference NGS testing included five NGS panels currently used at the Mayo Clinic, which are as follows: Lung Panel with Tumor Rearrangement, MayoComplete Solid Tumor Panel, Sarcoma Targeted Gene Fusion Panel, NTRK Gene Fusion Panel, and Neuro-Oncology Expanded Panel. Detailed information can be found in Supplementary Table 1.
Specimen-Type Acceptance Studies
For the NSCLC samples, a wide range of material types were included, which are as follows: 33 FFPE surgical specimens, 10 fine-needle aspiration (FNA) or pleural fluid in cell blocks (also FFPE), 10 FNA smears stained with Diff-Quik or Pap, and 11 cases of previously extracted RNA. Samples positive for NTRK1/2/3 fusions were all FFPE specimens.
Limit of Detection Studies, Variable Parameters
To assess minimum tumor percentage input requirement, a pathologist used a hematoxylin and eosin slide to mark areas with approximately 10% or 5% tumor cell content for three different samples. Unstained FFPE sections were scraped accordingly and tested.
For minimal size of tumor tissue required, four FFPE samples were each scraped at areas of 72 mm2, 36 mm2, and 18 mm2 and tested. To evaluate the cellularity requirement for smears, two samples with high cellularity (>5000 cells on a slide), four samples with medium cellularity (3000–5000 cells), and four samples with low cellularity (300–3000 cells) were run, all containing at least 5% tumor cell content. To determine the lowest acceptance limit of nucleic acid, three different samples were each tested at RNA amounts of 20, 10, and 5 ng.
Results
Accuracy
All 12 samples (100%) with known ALK fusion were correctly detected by this assay, including 11 by FS detection and one by EI analysis (Table 1). Ten samples with ROS1 fusion (100%) were correctly detected, all by the FS method. All 12 RET-rearranged cases (100%) were successfully detected (11 FS, one EI). Ten samples with MET exon 14 skipping were all (100%) detected. The sensitivities of detecting ALK, ROS1, and RET fusions and MET exon 14 skipping in the tested NSCLC samples are all 100%. All results were available in approximately 3 hours.
Table 1.
Accuracy Summary of Fusion Detection of Each Gene
| Gene | Number of Positive Samples | Number of Samples Detected by FS Method | Number of Samples Detected by EI Only | Sensitivity | Specificitya |
|---|---|---|---|---|---|
| ALK | 12 | 11 | 1 | 12/12 (100%) | 51/52 (98%) |
| ROS1 | 10 | 10 | 0 | 10/10 (100%) | 54/54 (100%) |
| RET | 12 | 11 | 1 | 12/12 (100%) | 52/52 (100%) |
| MET exon 14 skipping | 10 | 10 | N/A | 10/10 (100%) | 54/54 (100%) |
| 20 samples negative for fusion or MET exon 14 skipping | |||||
| NTRK1 | 10 | N/A | 10 | 10/10 (100%) | 87/87 (100%)b |
| NTRK2 | 10 | N/A | 1 | 1/10 (10%) | 36/36 (100%)b |
| NTRK3 | 13 | N/A | 13 | 13/13 (100%) | 32/37 (86%)b |
| NTRK2/3 | 23 | N/A | 19 | 19/23 (83%) | 26/26 (100%)b |
Note: Samples positive for ALK, ROS1, RET fusions and MET exon 14 skipping and negative controls are NSCLC specimens
Samples positive for NTRK1/2/3 are non-NSCLC specimens
EFS, fusion specific; I, expression imbalance; N/A, not applicable.
Different NGS panels (served as reference method) have different gene coverage.
Including NSCLC samples.
Of the 10 NTRK1 fusion-positive samples, all (100%) were successfully detected by this assay. Among the 10 samples positive for NTRK2 fusion, only one (10%) produced a NTRK2 fusion–detected result. Surprisingly, five samples were classified as detection of NTRK3 fusion, whereas four samples had no fusion detected. All 13 (100%) NTRK3 fusion-positive samples were accurately detected.
This assay did not detect fusions in the 20 fusion-negative NSCLC samples. Because each sample run contained multiplex RT-qPCR reactions evaluating multiple gene rearrangements, a sole detection of one gene fusion implied negative results in other genes. One false-positive detection of ALK fusion was called by EI analysis only in a RET-rearranged sample. For this sample, break apart ALK FISH found no ALK gene rearrangement but 80% of the nuclei had three to five copies of the ALK gene. In the tested NSCLC samples, the overall specificities for ALK, ROS1, and RET fusions and MET exon 14 skipping alteration detection were 98% (51 of 52), 100% (54 of 54), 100% (52 of 52), and 100% (54 of 54), respectively.
No false-positive calls of NTRK1 or NTRK2 fusion were noted in the tested samples. The individual specificities of NTRK1, NTRK2, and NTRK3 fusion detection were 100% (87 of 87), 100% (36 of 36), and 86% (32 of 37), respectively. If reported in combination as NTRK2/3 fusion, the detection specificity was 100% (26 of 26) with the sensitivity of 83% (19 of 23).
Specimen-Type Acceptance
Various specimen types were validated. For the 64 NSCLC samples, the concordances were 100% across every material type evaluated (Table 2), except for a false-positive ALK fusion call on a RET-rearranged case from an FFPE tissue block. Most samples were less than 4 years old. Rearrangement events were concordantly detected in two samples older than 5 years and in four samples collected between 4 and 5 years ago.
Table 2.
Summary of Testing Results of Samples With Different Specimen Types
| Specimen Type | Number of Samples | Positive Cases by NGS | Positive Cases by Idylla | Negative Cases by NGS | Negative Cases by Idylla | Concordance of Idylla vs. NGS |
|---|---|---|---|---|---|---|
| FFPE tissue block | 33 | 23 | 23a | 10 | 10 | 33/33 (100%) |
| FNA or pleural fluid in cell block (FFPE) | 10 | 7 | 7 | 3 | 3 | 10/10 (100%) |
| Cytology/FNA smear | 10 | 7 | 7 | 3 | 3 | 10/10 (100%) |
| Pre-extracted RNA | 11 | 7 | 7 | 4 | 4 | 11/11 (100%) |
FFPE, formalin-fixed, paraffin-embedded; FNA, fine-needle aspiration; NGS, next-generation sequencing.
A RET-rearranged sample revealed both RET and ALK fusions detected by Idylla.
Limits of Detection
All three cases produced concordant results when areas of either 10% or 5% tumor content were scraped and tested. Tumor size limit of detection studies revealed that all four samples were concordant when 72, 36, or 18 mm2 of the tissue was scraped. All cytology smears yielded concordant results, including one ROS1-rearranged case detected at 5% tumor content and low cellularity (300–3000 cells). Finally, three of three RNA samples were concordant at 20 ng, one of three was concordant at 10 ng, but all three samples were invalid when 5 ng was used. These findings indicated that 20 ng of RNA input is required when using pre-extracted RNA (Table 3).
Table 3.
Limit of Detection of Idylla GeneFusion Assay and Comparison With Large Panel NGS Assays
| Characteristic | Suggested Requirement for NGSa | Minimum Requirement for NGSa | Suggestion Requirement by Manufacturer for Idylla | Minimum Requirement for Idylla After Verification |
|---|---|---|---|---|
| Tissue size | 72 mm2 | 36 mm2 | 20 mm2 | 18 mm2 |
| Tumor % | 40% | 20% | 10% | 5% |
| Total cells on smear | At least 5000 cells | 3000 cells | N/A | 3000 cells, possible >300 cells |
| RNA, total | 80 ng | 80 ng | N/A | 20 ng |
N/A, not available; NGS, next-generation sequencing.
Using clinically validated MayoComplete Solid Tumor Panel for reference.
Discussion
In clinical settings where a timely therapeutic decision must be made, rapid molecular testing is desired. The ultraquick turnaround time of the rapid fusion assay, yielding results in approximately 3 hours, can have impactful clinical utility. For patients with NSCLC who have rapid disease progression, quick molecular testing for gene fusions, EGFR mutation status, and programmed death-ligand 1 immunohistochemistry can provide valuable information for immediate treatment options. Furthermore, surgical candidates considered for neoadjuvant immunotherapy16 would benefit from timely exclusion of oncogenic mutations and fusions.
We found that the rapid fusion assay has robust sensitivities and specificities in detecting ALK, ROS1, and RET fusions, and MET exon 14 skipping in the NSCLC samples. The detection of ALK, ROS1, and RET fusions includes dual FS and EI methods in this assay. The FS detection covers common fusion partners (e.g., EML4::ALK and CD74::ROS1). Using the FS method only, regardless of the findings by EI, 94% (32 of 34) of the positive cases were successfully detected by this method with 100% (52 of 52) specificity. The EI analysis provided additional detection covering fusions with uncommon partners or exon combinations. Determining fusion detection by either FS or EI, as suggested by the manufacturer, increases the sensitivity to 100% (34 of 34) but slightly decreases the specificity to 98% (51 of 52). According to the manufacturer, the EI algorithm was designed with a higher threshold; indeed, among 32 samples with fusion detection by the FS method, only 22 were called positive by the EI method. Nevertheless, using the EI method to complement the FS method successfully increases the sensitivity, as it was designed to identify novel fusions. However, one false-positive case was revealed in our study. Thus, in clinical practice, we recommend confirmatory testing (i.e., FISH or NGS) when fusions are detected by the EI method only.
This assay detects NTRK1/2/3 fusions by the EI method only without the FS method. It revealed an acceptable accuracy in detecting NTRK fusions. Using non-NSCLC samples, it revealed excellent detection rate for NTRK1 and NTRK3 fusions but not for detection of NTRK2 fusions. Among the 10 samples known to have NTRK2 fusion, only one produced a positive result by this assay, whereas five were misidentified as NTRK3 fusion detected. Importantly, the assay was developed with a limited number of NTRK2-rearranged specimens.13, 14, 15 Furthermore, it has been recently reported that baseline tyrosine kinase gene expression differs among cancer types: (1) NTRK3 displayed higher expression in glioblastoma and colorectal cancer compared with other kinase receptors, such as ALK, ROS1, RET, and NTRK1; and (2) NTRK2 had strong difference in baseline expression between adenocarcinoma and squamous cell carcinoma of the lung.15 These factors could have affected the NTRK2 and NTRK3 EI method-calling algorithm. Because NTRK fusions are rare in NSCLC (approximately 0.1%–0.3%)7 and treatments among NTRK1/2/3 fusions are the same, we suggest reporting NTRK1 and combined NTRK2/3 findings, with confirmatory NGS or FISH testing.
This automated system revealed wide specimen acceptance profile and a simple workflow. It is common for thoracic oncology molecular testing to encounter low tumor percentage or limited-size tissues from bronchoscopy biopsy and FNA specimens. Although NGS RNA panel assay generally requires at least 20% of tumor content and 10 to 15 unstained slides, this assay can reliably detect gene fusions on samples with as little as 5% tumor content with only one to three unstained slides (Table 3). We successfully validated a wide range of tissue types, including tissue FFPE, FNA, and pleural fluid cell block FFPE, cytology smears, and extracted RNA, expanding its practical clinical utility. In the current practice, a significant number of non–small cell carcinoma diagnoses are made by bronchoscopy procedures. However, cytology specimens have not been thoroughly validated in molecular tests nor in immunohistochemistry studies. Here, we provided validation data to support the reliable clinical utility of this assay to accept various cytology specimens, including FNA, smears, and fluid cell block preparation. Finally, the automated workflow of this assay requires minimal hands-on time, saving labor cost for molecular laboratories.
In conclusion, our study reveals the rapid fusion assay as a fast and reliable alternative assay for detecting targetable ALK, ROS1, and RET fusions, and MET exon 14 skipping in the setting of NSCLC and detection of the less common NTRK1/2/3 fusions. In the clinical setting, this assay offers timely and impactful molecular information for managing patients with NSCLC with targeted therapy and selecting patients for neoadjuvant therapy or clinical trials. Rapid molecular testing could be considered before larger comprehensive gene panel testing to expedite patient care.
CRediT Authorship Contribution Statement
Alessia Buglioni: Data curation, Writing—original draft, Writing—review and editing.
Patricia L. Caffes: Data curation, Methodology, Validation, Writing—review and editing.
Mark G. Hessler: Data curation, Methodology, Validation, Writing—review and editing.
Aaron S. Mansfield; Writing—review and editing.
Ying-Chun Lo: Conceptualization, Data curation, Funding acquisition, Project administration, Supervision, Writing—review and editing.
Footnotes
Dr. Buglioni and Ms. Caffes contributed equally to the work.
Disclosure: The Mayo Clinic received sponsorship of Idylla GeneFusion Assay cartridges from Biocartis.
Cite this article as: Buglioni A, Caffes PL, Hessler MG, Mansfield AS, Lo YC. Clinical utility validation of an automated ultrarapid gene fusion assay for NSCLC. JTO Clin Res Rep. 2022;3:100434.
Note: To access the supplementary material accompanying this article, visit the online version of the JTO Clinical and Research Reports at www.jtocrr.org and at https://doi.org/10.1016/j.jtocrr.2022.100434.
Supplementary Data
References
- 1.Addeo A., Passaro A., Malapelle U., Luigi Banna G., Subbiah V., Friedlaender A. Immunotherapy in non-small cell lung cancer harbouring driver mutations. Cancer Treat Rev. 2021;96 doi: 10.1016/j.ctrv.2021.102179. [DOI] [PubMed] [Google Scholar]
- 2.Molina J.R., Yang P., Cassivi S.D., Schild S.E., Adjei A.A. Non-small cell lung cancer: epidemiology, risk factors, treatment, and survivorship. Mayo Clin Proc. 2008;83:584–594. doi: 10.4065/83.5.584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dolly S.O., Collins D.C., Sundar R., Popat S., Yap T.A. Advances in the development of molecularly targeted agents in non-small-cell lung cancer. Drugs. 2017;77:813–827. doi: 10.1007/s40265-017-0732-2. [DOI] [PubMed] [Google Scholar]
- 4.Peters S., Camidge D.R., Shaw A.T., et al. Alectinib versus crizotinib in untreated ALK-positive non-small-cell lung cancer. N Engl J Med. 2017;377:829–838. doi: 10.1056/NEJMoa1704795. [DOI] [PubMed] [Google Scholar]
- 5.Gainor J.F., Curigliano G., Kim D.-W., et al. Pralsetinib for RET fusion-positive non-small-cell lung cancer (ARROW): a multi-cohort, open-label, phase 1/2 study. Lancet Oncol. 2021;22:959–969. doi: 10.1016/S1470-2045(21)00247-3. [DOI] [PubMed] [Google Scholar]
- 6.Shaw A.T., Ou S.H., Bang Y.J., et al. Crizotinib in ROS1-rearranged non-small-cell lung cancer. N Engl J Med. 2014;371:1963–1971. doi: 10.1056/NEJMoa1406766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cocco E., Scaltriti M., Drilon A. NTRK fusion-positive cancers and TRK inhibitor therapy. Nat Rev Clin Oncol. 2018;15:731–747. doi: 10.1038/s41571-018-0113-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Paik P.K., Felip E., Veillon R., et al. Tepotinib in non-small-cell lung cancer with MET exon 14 skipping mutations. N Engl J Med. 2020;383:931–943. doi: 10.1056/NEJMoa2004407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lim S.M., Kim E.Y., Kim H.R., et al. Genomic profiling of lung adenocarcinoma patients reveals therapeutic targets and confers clinical benefit when standard molecular testing is negative. Oncotarget. 2016;7:24172–24178. doi: 10.18632/oncotarget.8138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hagemann I.S., Devarakonda S., Lockwood C.M., et al. Clinical next-generation sequencing in patients with non-small cell lung cancer. Cancer. 2015;121:631–639. doi: 10.1002/cncr.29089. [DOI] [PubMed] [Google Scholar]
- 11.Takeda M., Sakai K., Terashima M., et al. Clinical application of amplicon-based next-generation sequencing to therapeutic decision making in lung cancer. Ann Oncol. 2015;26:2477–2482. doi: 10.1093/annonc/mdv475. [DOI] [PubMed] [Google Scholar]
- 12.Desmeules P., Boudreau D.K., Bastien N., et al. Performance of an RNA-based next-generation sequencing assay for combined detection of clinically actionable fusions and hotspot mutations in NSCLC. JTO Clin Res Rep. 2022;3 doi: 10.1016/j.jtocrr.2022.100276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chu Y.H., Barbee J., Yang S.R., et al. Clinical utility and performance of an ultrarapid multiplex RNA-based assay for detection of ALK, ROS1, RET, and NTRK1/2/3 rearrangements and MET exon 14 skipping alterations. J Mol Diagn. 2022;24:642–654. doi: 10.1016/j.jmoldx.2022.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Depoilly T., Garinet S., van Kempen L.C., et al. Multicenter evaluation of the Idylla GeneFusion in non-small-cell lung cancer. J Mol Diagn. 2022;24:1021–1030. doi: 10.1016/j.jmoldx.2022.05.004. [DOI] [PubMed] [Google Scholar]
- 15.Sorber L., Van Dorst B., Bellon E., et al. NTRK gene fusion detection in a pan-cancer setting using the Idylla GeneFusion assay. J Mol Diagn. 2022;24:750–759. doi: 10.1016/j.jmoldx.2022.03.014. [DOI] [PubMed] [Google Scholar]
- 16.Forde P.M., Spicer J., Lu S., et al. Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer. N Engl J Med. 2022;386:1973–1985. doi: 10.1056/NEJMoa2202170. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.