Skip to main content
Premium Trial:

Request an Annual Quote

Caris Prepares FDA Submission of RNA Sequencing Test; Will Include Companion Dx Claims for Fusions


SAN FRANCISCO (GenomeWeb) – Tumor profiling company Caris Life Sciences aims to position its cancer transcriptome sequencing test as a companion diagnostic for drugs targeting gene fusions.

The company, which launched the assay, called MI Transcriptome, this week, is currently writing up its validation study and preparing its submission filing for approval from the US Food and Drug Administration. In addition, the firm will apply for FDA Breakthrough Device Designation, according to David Spetzler, Caris' president and chief scientific officer.

Spetzler said that the firm has been running a targeted RNA-seq panel that uses amplicon sequencing to analyze 53 genes. "In the [pharmaceutical] pipeline, we see a whole host of fusion-based targeted therapeutics," Spetzler said, so rather than expanding on its targeted RNA-seq assay, the company decided to analyze the whole transcriptome, ensuring that all potential drug targets are covered and to glean additional insights from the data in regards to patients' tumor profiles and their outcomes. Already, he said, the machine-learning techniques the company uses have enabled "fascinating insights into how patients respond to drugs."

In addition, the targeted amplicon-based methods tend to be less sensitive than methods using hybrid capture for the entire transcriptome, Spetzler said.

A handful of gene fusions have targeted therapies associated with them, including ALK and ROS1 fusions in lung cancer and NTRK fusions in various cancer types. For example, the FDA approved Loxo Oncology's Vitrakvi (larotrectonib) last November to treat adult and pediatric patients with solid tumors characterized by NTRK fusions, regardless of the tumor's body site of origin, making it the second tumor-agnostic targeted therapy approved to date.

Loxo Oncology has until 2021 to file a report with the FDA on a companion diagnostic for the therapy. Previously, the pharmaceutical company said it is collaborating with Illumina to develop such an assay.

Spetzler said that Caris' RNA-seq assay could detect all three types of fusions and that the company would aim to seek approval for the assay as a companion diagnostic for all three indications. For the ALK and ROS1 fusions, because there are already approved tests, Spetzler said that Caris would have to do a non-inferiority study, showing that its RNA-seq test performs equivalently to the already approved assays. For the NTRK assay, the firm could use the same approach once there is an approved assay, or it could collect enough outcomes data on patients to show independently that its assay can be used as a CDx. He noted that the company is not collaborating with Loxo and is seeking approval independently of the pharmaceutical company.

Spetzler said that an RNA-seq assay would differentiate Caris' offering from other tumor profiling tests on the market, particularly because of its ability to detect gene fusions. DNA-based tests can often identify genomic rearrangements but are unable to specify whether they have resulted in a fusion, deletion, inversion, or some other change. "For drugs like larotrectinib, it's really important to know the type, because the drug is approved for fusions; it's not approved for deletions or duplications or inversions," Spetzler said. As a result, a patient cannot receive the therapy unless the specific NTRK fusion is confirmed.

In addition, aside from gene fusions, the RNA-seq assays will be able to detect relevant splice variants and gene expression. Drugs like Herceptin, for example, are indicated for patients with overexpressed HER2 genes. Therapeutically relevant splice variants include the androgen receptor v7 variant in prostate cancer and EGFR variant 3 in glioblastoma. Spetzler said that while there are not yet approved glioblastoma drugs for EGFRv3, there are drugs, including immunotherapies, in development. In addition, knowing ARV7 status in prostate cancer can be helpful in predicting efficacy of treatment, he said.

While splice variants can often be detected in DNA-based assays, gene fusions cannot, Spetzler said. The RNA-seq test will not be a standalone clinical test, however, and will be deployed as part of Caris' MI Tumor Seek and MI Profile assays. MI Profile is a comprehensive assay that tests a variety of biomarkers depending on the tumor type submitted, using technologies such as NGS, immunohistochemistry, and in situ hybridization. MI Tumor Seek involves NGS of both DNA and RNA and ordering clinicians have the option of adding selected immuno-oncology markers. Previously, the RNA sequencing portion of Caris' assays was its targeted amplicon-based 53-gene RNA panel, but now will be the entire transcriptome.

For the DNA portion of its tests, Caris had previously launched a 592-gene panel and said in 2017 that it planned to take that panel through FDA approval. However last year, following the validation of the panel for also detecting microsatellite instability, the firm said it would submit an expanded panel of 650 to 700 genes that would also include MSI, tumor mutational burden, and genome-wide loss of heterozygosity testing. Instead, the firm has now decided to take an exome sequencing assay through FDA clearance instead of the gene panel, a plan it first revealed at the JP Morgan Healthcare conference last month.

Spetzler said that submitting a gene panel to the FDA would potentially be more burdensome in the future, as new targets are discovered and new drugs are developed, since it would require updating the test to include those new targets. As a result, the company decided to switch to the exome test.

He did not say when the company would launch and submit the exome test to the FDA, noting that there are a number of hurdles to overcome that are not present for a transcriptome test.

"One challenge is how to handle pseudogenes and homopolymer regions," Spetzler said, adding that with RNA sequencing, pseudogenes are not a problem because they are not transcribed and so not sequenced.

Despite switching to more comprehensive testing, Spetzler said, the out-of-pocket cost to patients, which varies depending on patients' insurance plan, would not change. The firm has contracts with insurance companies with 170 million lives covered.