In recent years, tyrosine kinase inhibitors that specifically target RET have entered the pipeline. Although none have yet received FDA approval, the promise of these agents has spurred interest in diagnostic assays.
Lynette M. Sholl, MD
Testing for RET aberrations is starting to gain importance in clinical practice along with the awareness that RET fusions and mutations are drivers in multiple tumor types. In recent years, tyrosine kinase inhibitors that specifically target RET have entered the pipeline. Although none have yet received FDA approval, the promise of these agents has spurred interest in diagnostic assays that can accurately and efficiently detect RET abnormalities.
The RET receptor tyrosine kinase plays a crucial role in cell growth and differentiation. RET rearrangements are observed in 1% to 2% of non—small cell lung cancers (NSCLCs). RET mutations are found in 60% of sporadic cases of medullary thyroid cancer (MTC) and 90% of hereditary MTC (Figure 1). They are also observed in papillary renal cell carcinoma, breast, colon, and pancreatic cancers (Figure 2).1
Testing for RET aberrations is challenging because multiple different assays may be required to detect the range of possible alterations in different tumor types.
RET aberrations in NSCLC are usually the result of a gene fusion, most commonly between RET and KIF5B. Additionally, a host of gatekeeper mutations typically denote resistance to therapy, particularly V804L and V804M.2
“RET [in NSCLC] is like other driver oncogenes, such as EGFR, ROS1, and ALK, in that something essentially leads to an abnormal event—for instance, when a fusion occurs—and that allows the tumor to propagate irrespective of external signals,” said Lynette M. Sholl, MD, an associate pathologist at Brigham and Women’s Hospital and associate professor at Harvard Medical School in Boston, Massachusetts.
In patients with NSCLC, RET fusions are more common in younger individuals, both smokers and never-smokers, and regardless of gender. In this setting, they occur just in patients with adenocarcinoma with a predominantly lepidic or papillary growth pattern in their tumor.2 On the other hand, MTC often involves RET-activating point mutations, according to Jacob S. Van Naarden, chief business officer of Loxo Oncology, which is developing the RET-targeting agent LOXO-292.
In NSCLC in particular, it is usually more important in a clinical setting to determine whether a patient has an EGFR, ALK, RET, NTRK, BRAF, or ROS1 mutation, each of which has a cognate targeted therapy either in clinical development or approved by the FDA. RET is seldom at the top of the list.
The molecular testing guidelines for lung cancer from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology recommend that RET testing not be performed as a routine standalone assay outside the context of a clinical trial. In the expert consensus opinion, the guidelines further state that it is appropriate to include RET as part of larger testing panels, either initially or when routine tests for EGFR, ALK, and ROS1 mutations are negative.3
When testing for RET rearrangements, 3 main options exist: fluorescence in situ hybridization (FISH), immunohistochemical (IHC) testing, and next-generation sequencing (NGS).
Traditionally, FISH has been used to detect RET rearrangements, but as a single-gene assay, it is not practical when seeking to test for multiple aberrations. “FISH is a very robust assay, but it is limited, like other single-gene target assays, in that if you are interested in a range of different potential genomic alterations that could be amenable to targeted therapies, you are generally reluctant to serially test every single possibility in a small biopsy specimen,” Sholl said. She explained that multiple driver mutations are of interest in lung cancer, and it can be cumbersome to do sequential testing for them all.
FISH is also relatively expensive, requires technical expertise for interpretation, and is usually available just in larger centers and reference laboratories. It also cannot determine the RET fusion partner gene, and no standard cutoffs define positivity—estimates range from 10% to 20% of cells with RET split signals.4
Given the rarity of RET aberrations in NSCLC, patients and their advocates wouldn’t want serial testing done, Van Naarden said. NGS offers a better option. “Tumor biopsies are scarce; they often require some degree of potential harm to the patient, so increasingly, physicians want to get as much information out of that scarce material as possible, and using one of the NGS assays has the potential to answer multiple nonoverlapping questions,” he said.
IHC presents a similar problem in that screening for each oncogene would have to be done serially, but there is interest in performing IHC testing for RET. “There have been efforts made to identify good antibodies and protocols for detection of RET rearrangements,” Sholl said. IHC also has frequent rates of false positives and false negatives, making its results unreliable, even though the test requires relatively little expertise to interpret.4 IHC tends to be a cheaper option than FISH, and it is easier for pathologists to perform in a laboratory setting, she said, but there are not enough data to support IHC as an effective testing method for RET.
Therefore, it is generally more efficient to test for RET with NGS, but that method has its own shortcomings. Sholl explained that although a lot of the DNA-based hybrid capture NGS assays are effective at detecting fusions, they are not perfect. This is partly because when using a DNA-based assay, fusion breakpoints typically localize to the introns of a gene, and introns are “notoriously difficult” to sequence, Sholl said: “They tend to contain of repetitive regions that are common to many areas of the genome, so it may be difficult to get specific information about that 1 particular intronic area of the gene.”
As a result of this challenge, many laboratories have paired assays that rely on RNA-based sequencing. One of the more popular types of approaches, according to Sholl, involves anchored multiplex polymerase chain reaction (PCR) testing, where the pathologist is essentially sequencing from the RET side of the gene transcript and searching in the RNA to find a fusion between RET and a partner gene.
“That’s an approach that is increasingly being used in practice; you will see laboratories that maybe have a focused NGS panel that can detect mutations and maybe some copy number alterations by using the DNA, and another panel that’s going to be good for detecting fusions by using the RNA,” Sholl said.
Loxo Oncology recently partnered with Illumina to develop and commercialize a multigene panel for broad tumor profiling. They plan to seek FDA approval for Illumina’s TruSight Tumor 170 NGS assay as a companion diagnostic for Loxo Oncology’s larotrectinib, which targets NTRK gene fusions, and LOXO-292, which targets RET alterations.5
TruSight Tumor 170, as the name suggests, is designed to cover 170 genes in an enrichment- based targeted panel that simultaneously analyzes DNA and RNA, making it helpful when searching for RET rearrangements. The assay can assess fusions, splice variants, insertions and deletions, single-nucleotide variants, and amplifications of the genes in the panel, meaning it can detect both RET fusions and point mutations.6
“The RNA component of the [TruSight Tumor 170] assay is particularly unique and useful for fusion detection,” Van Naarden said. “DNA-based fusion detection methods are inherently more challenging because of the way fusion events happen in the genome and the way the sequencing assays are designed.” For patients with NSCLC, Illumina’s NGS approach would catch any RET fusions without passing over other gene mutations that are actionable targets for therapy.
In the setting of MTC, where RET point mutations are much more common, some endocrinology and head and neck specialists immediately suspect RET. “Physicians will [often] use a first-pass test that is more RET-directed. It may still be sequencing based, from a technology perspective,” Van Naarden said, noting that RET is more of a concern for a patient with MTC than for a patient with lung cancer, given the frequency.Although no FDA-approved agents specifically for RET currently exist, some drugs target RET incidentally. “There are a few drugs that are promiscuous kinase inhibitors, with marginal RET activity in humans and used off-label with limited success,” Van Naarden said. Vandetanib (Caprelsa) and cabozantinib (Cometriq) are both approved for MTC, and alectinib (Alecensa), which is FDA-approved for ALK-positive NSCLC, has also shown activity against RET alterations. Ponatinib (Iclusig) and lenvatinib (Lenvima) are also multikinase inhibitors with activity against RET.2
There have been successes with targeted RET inhibitors in clinical trials. LOXO-292 has generated great interest, given a mild toxicity profile and a high level of response across RET fusion partners, regardless of gatekeeper mutations.
In comparison, multikinase inhibitors that are approved for NSCLC or MTC are sometimes referred to as “dirty” drugs because they attack multiple targets along with RET signaling, leading to a greater incidence of adverse events (AEs).1
The phase I/II open-label LIBRETTO-001 clinical trial (NCT03157128) is testing LOXO-292 in patients with RET fusion—positive NSCLC, MTC, and other solid tumors with RET activity. Interim results of LIBRETTO-001 demonstrated an objective response rate (ORR) of 77% (95% CI, 61%-89%) per RECIST 1.1 criteria among 49 patients with RET fusion-positive tumors, including 38 with NSCLC, 9 with thyroid cancer, and 2 with pancreatic cancer. In 29 patients with RET-mutated MTC, the ORR was 45% (95% CI, 24%-68%), with 1 complete response (CR) and 1 additional CR awaiting confirmation. In 4 participants with no known activating RET alteration, there was no response with LOXO-292.1
There were just 2 treatment-related grade 3 AEs—tumor lysis syndrome and increased alanine aminotransferase levels, both of which were resolved—and no grade 4 AEs. The most common AEs in the study were fatigue (20%), diarrhea (16%), constipation (15%), dry mouth (12%), nausea (12%), and dyspnea (9%). The most common AEs directly related to treatment were fatigue (13%), dry mouth (6%), nausea (5%), diarrhea (2%), and constipation (2%).
Two other RET inhibitors are also in development: BLU-667, tested in the phase I ARROW trial (NCT03037385), and RXDX-105, tested in a phase I/Ib trial (NCT01877811). LOXO-292 is currently the furthest along in its development.As these agents’ development advances, it follows naturally that testing for RET mutations and fusions will become more commonplace. “I suspect if the LOXO-292 data [come] out supporting its first-line use as part of routine therapy in patients with RET rearrangements, then there will be more of a push to see if we can optimize an IHC screening approach,” Sholl said.
Assays such as TruSight Tumor 170 are also projected to become more popular for RET. “The technology is evolving to increasingly include RNA for better fusion detection, so you don’t miss patients. If you’re going to go through the whole process of a biopsy and testing, you don’t want to use a test that by its very design can’t find what you’re looking for,” Van Naarden said.