Genomic Testing to Identify Targetable Fusions

In December 2023, Hanna Tukachinsky, PhD, and Jessica Lee, MS, were authors on a published paper titled, “Circulating Tumor DNA Enables Sensitive Detection of Actionable Gene Fusions and Rearrangements Across Cancer Types.” They performed this research while serving as translational and data scientists for Foundation Medicine. Together they focus on diagnostics, precision oncology, and the clinical applications of genomic data. Dr. Tukachinsky and Ms. Lee sat down with us to explain how FoundationOne®Liquid CDx’s large assay panel can detect, in liquid biopsy samples, not only targetable mutations, but also actionable fusions.

Unstable Genomes Can Generate Fusions

Cancer cells have unstable genomes that contain chromosomal gains and losses as well as DNA rearrangements. When these large chromosomal pieces shift around, genes can be disrupted and fused to other genes. In some cases, this rearrangement occurs in a tumor suppressor gene and the result is a loss of gene function. In other cases, gene fusions can result in a gain of function, or activation. In these cases, for example, a kinase can become constitutively active resulting in an oncogenic driver. While targetable fusions occur in 1-2% of cancer patients overall, certain cancer types are more likely to have fusions.

“It’s crucial to detect these fusions,” said Dr. Tukachinsky, “because you are ﬁguring out what’s driving the cancer, and then, in cancer types such as non-small cell lung cancer, you can potentially therapeutically target this driver. It’s standard-of-care at this point to test for a variety of drivers to select the right therapy for the patient.”

Circulating Tumor DNA (ctDNA) Can Reveal Actionable Fusions

FoundationOne Liquid CDx analyzes multiple classes of genomic alterations in DNA to identify genetic drivers of cancer, including point mutations, short insertions and deletions, rearrangements, gene deletions and amplifications. The real advantage of DNA as an analyte, said Dr. Tukachinsky, is that DNA is highly stable and can survive in tissue even if the tissue is archival and had been biopsied many years previously. Likewise, DNA does not degrade quickly in the blood. While RNA analysis can add value to tissue biopsies, robust DNA sequencing still dominates in both tissue and liquid biopsies.

Dr. Tukachinsky explained that FoundationOne Liquid CDx uses DNA to detect fusions such as ALK, RET, and ROS1. Fusions can be difficult to detect because fusion breakpoints can occur not only anywhere along a gene’s exons, but also, in many cases, within a gene’s introns. This means that even though introns do not code for the protein product of the gene, they are important for genomic testing. Unfortunately, introns are very large and there are time and cost challenges associated with sequencing every intron. The challenges are magnified in liquid biopsies because the sample typically contains only a small amount of ctDNA, requiring greater sequencing depth. A well-designed assay must thus maximize the number of detectable fusions of interest while simultaneously sequencing with enough depth to sensitively detect drivers.

FoundationOne Liquid CDx Platform

FoundationOne Liquid CDx extracts circulating cell-free DNA from peripheral whole blood and sequences it using hybridization-captured, adaptor ligation-based libraries. The assay sequences a total of 324 cancer-related genes for base substitutions, short insertions and deletions, copy number ampliﬁcations, homozygous deletions, and large genomic rearrangements. In addition to reporting this information, the assay also uses de novo assembly to identify the presence of complex biomarkers such microsatellite instability, blood tumor mutational burden, and ctDNA tumor fraction.

Perhaps more significantly, the assay also detects fusions and it does this without requiring any prior knowledge of the gene fusion partner. The assay can thus detect any rearrangement of a gene on the panel, wherever that gene happens to land in the genome, thereby giving the assays both ﬂexibility and detection power. This approach is a major advantage compared to the ﬂuorescence in situ hybridization (FISH) approach which used probes on either side of the breakpoint to detect a fusion event between known gene partners. Dr. Tukachinsky explained that these advantages persist even when compared to whole exome sequencing (WES). This is because while WES sounds thorough in that it sequences every known gene, it is limited by the fact that WES prioritizes sequencing of exons and may not include additional intronic coverage; therefore, it is restricted in its ability to detect fusion events, most of which occur in introns.

Another challenge particular to using next-generation DNA sequencing to detect fusions is that, unlike mutations which have only a few changed base pairs, fusions usually involve messy non-homologous end joining DNA repair. Dr. Tukachinsky emphasized that such repairs do not use a template and can result in the introduction of many new base pairs that, in turn, create challenges in aligning next generation sequence reads to each other. Foundation Medicine solves these technical challenges by using hybrid-capture to enrich for select introns that result in oncogenic fusions and using an analytical pipeline that can assemble the DNA even if some of the sequence is novel and has been introduced by sloppy DNA repair.

Foundation Medicine has 48 companion diagnostic indications across its comprehensive genomic profiling (CGP) portfolio with the number growing as trial results become available. Thus, clinicians who see that their patients have a fusion listed on their Foundation Medicine report may then determine whether the patient is eligible to receive an associated targeted therapy. Dr. Tukachinsky explained that, while many of these indications are cancer-type speciﬁc, ﬁve, including the one for RET fusions, are pan-tumor companion diagnostics. FoundationOne CDx is a companion diagnostic for RET alterations, for example, and so when a RET fusion appears on a patient’s report, the clinician may choose to treat the patient with a RET inhibitor. Since RET fusions are a pan-tumor indication, the ﬁnding is relevant not just for patients who have lung cancer or thyroid cancer where RET fusions are more common, but for any solid tumor where a RET fusion is detected.

Latest Research

Traditionally, clinicians have used tissue samples to identify genomic rearrangements and have been skeptical about the ability to reliably detect the rearrangements in liquid biopsies. The skepticism is largely based on published literature that has indicated a lower detection of rearrangements in liquid biopsy compared to tissue. “Our study examined the landscape of fusions and rearrangements detected by liquid biopsy of ctDNA across a range of cancer types,” said Ms. Lee. “We analyzed samples that were proﬁled using our validated next-generation sequencing panel assay FoundationOne Liquid CDx.” For their paper, Ms. Lee and Dr. Tukachinsky analyzed liquid biopsy results from 53,842 patients across 66 solid tumor types and found that 14% had one or more pathogenic rearrangement.

Ms. Lee explained that the study identiﬁed a wide variety of rearrangements in the ctDNA of liquid biopsy samples with approximately half of the samples with pathogenic rearrangement having at least one gain-of-function oncogenic rearrangement and half having at least one loss-of-function rearrangement. In some cases, the rearrangements were clonal events, while in other cases they appeared to be the result of acquired resistance. The researchers also found that the cancer types with the higher prevalence of gain-of-function rearrangements (including FGFR2, BRAF, ALK, and RET) included cancer types associated with canonical fusion drivers such as cholangiocarcinoma, non-small cell lung cancer, bladder cancer, and prostate cancer. Clinicians who treat patients with cholangiocarcinoma may ﬁnd their paper especially meaningful since these patients often lack sufficient tissue for genomic testing.

Thus, while many clinicians may still believe that it is diﬃcult to identify genetic rearrangements by next generation sequencing of liquid biopsy samples, the work by Dr. Tukachinsky and Ms. Lee indicates that FoundationOne Liquid CDx can robustly detect fusions in liquid biopsy samples. Ms. Lee acknowledged that assay performance characteristics may vary from manufacturer to manufacturer due to diﬀerent design and validation strategies and thus it may not be possible to extrapolate their ﬁndings to other assays. Speaking speciﬁcally to FoundationOne Liquid CDx, Ms. Lee explained that the results suggest that a well-designed DNA-based comprehensive genomic proﬁling platform can detect fusions and rearrangement events even in intronic regions.

The new data highlight the capabilities of FoundationOne Liquid CDx to identify actionable fusions in liquid biopsy. The data also underscore the potential value of FoundationOne Liquid CDx for clinicians who treat diseases for which the use of genomic testing to identify a fusion driver is standard-of-care. An example of this would be the identiﬁcation of ALK or ROS1 fusions in patients with non-small cell lung cancer. Robust fusion detection in liquid biopsies will be especially meaningful in situations when tissue is not available, or the fusion is an acquired resistance alteration.

FoundationOne®CDx and FoundationOne®Liquid CDx are qualitative next-generation sequencing based in vitro diagnostic tests for advanced cancer patients with solid tumors and are for prescription use only. FoundationOne CDx utilizes FFPE tissue and analyzes 324 genes as well as genomic signatures. FoundationOne Liquid CDx analyzes 324 genes utilizing circulating cell-free DNA and is FDA-approved to report short variants in 311 genes. The tests are companion diagnostics to identify patients who may benefit from treatment with specific therapies in accordance with the therapeutic product labeling. Additional genomic findings may be reported and are not prescriptive or conclusive for labeled use of any specific therapeutic product. Use of the tests does not guarantee a patient will be matched to a treatment. A negative result does not rule out the presence of an alteration.

Some patients may require a biopsy for testing with FoundationOne CDx when archival tissue is not available which may pose a risk. Patients being considered for eligibility for therapy based on detection of NTRK1/2/3 and ROS1 fusions should only be tested if tissue is unavailable. Patients who are tested with FoundationOne Liquid CDx and are negative for other companion diagnostic mutations should be reflexed to tumor tissue testing and mutation status confirmed using an FDA-approved tumor tissue test, if feasible.

For the complete label, including companion diagnostic indications and important risk information, please visit www.F1CDxLabel.com and www.F1LCDxLabel.com.