The last decade has seen a complete transformation of the diagnostic and treatment landscape in the management of non–small cell lung cancer (NSCLC). Following the initial discovery of recurrent epidermal growth factor receptor (EGFR) mutations and then anaplastic lymphoma kinase (ALK) translocations marking key subsets of NSCLCs highly responsive to targeted therapy, up-front reflex molecular testing has become requisite, impacting the workflow of pathology departments around the world in major ways and calling for careful tissue stewardship and close attention to proper triaging/processing/turnaround times. With recent data demonstrating excellent activity of the ALK/ROS/MET inhibitor crizotinib for ROS translocation–positive patients and robust activity of Raf kinase inhibitors with/without MEK inhibitors for BRAF V600E–positive NSCLCs, yielding subsequent US Food and Drug Administration (FDA) approvals, ROS/BRAF testing was quickly added appropriately by the guideline to this short list.1,2 Although high-throughput genomics studies complemented by large data sets from tumors tested by multigene test panels and episodic key clinical observations have led to further discoveries on actionable mutational subsets, the field's attention has suddenly turned away to a new wave of major discoveries changing the landscape with the successful introduction of checkpoint inhibitors for the management of advanced NSCLC calling for tissue testing for programmed death ligand-1 (PD-L1) and possibly an expanding spectrum of alternative immune biomarkers to optimize patient management.3,4
It is in this context that we need to consider the updated College of American Pathologists (CAP)/International Association for the Study of Lung Cancer (IASLC)/Association for Molecular Pathology (AMP) guideline5 on the incorporation of testing for RET/ErbB2/MET aberrations in the diagnostic paradigm of optimal lung cancer management. Given the precedent on the above-mentioned molecularly defined tumor subsets, what is now the gold standard as to moving a biomarker up from investigational to clinically validated and deserving full partnership in the lineup of time-tested biomarkers? Do we need phase 3 studies demonstrating survival or quality-of-life benefits? Are earlier-phase studies demonstrating high levels of activity sufficient? Or is the gold standard the availability of an FDA-approved drug for a biomarker-based lung indication or adoption and endorsement of the marker by key stakeholder/expert groups, such as, for example, the National Comprehensive Cancer Network? Looking from another angle, let us consider the downside of the incorporation of a stand-alone molecular test for any particular marker with some but limited utility. Besides added cost and ill-advised utilization of limited resources, there are certainly concerns of misguided tissue stewardship, as, given the range of tissue needs for patient management—basic histology, core molecular markers, immune biomarkers, and potential tissue needs for present and future experimental studies and further interrogations based on unanticipated new discoveries—any biomarker added competes on a certain level for the same minute tissue piece, and wise rationing is indeed called for.
So let us consider the available and anticipated data for the 3 markers in question. RET translocations occurring in about 1% of lung adenocarcinomas were discovered contemporaneously with ROS translocations, and during the last 5 years, multiple cohort and phase 2–level studies have been reported in print and in abstract form demonstrating somewhat disappointingly limited activity of a range of available RET kinase inhibitors.6 Admittedly, most of the studied compounds so far have been multitargeted kinase inhibitors with significant toxicities because of their cross-inhibiting properties, limiting the potential therapeutic window of these agents. In the last few months, promising data have started to emerge about more specific and potent RET inhibitors, including RXDX-105 and LOXO-292.7,8 Although there is certainly hope that these studies ultimately will yield drugs with high enough activity to warrant further development and FDA approval, currently a patient with a RET-translocated tumor likely is still better served up front getting chemotherapy/immunotherapy rather than one of the available “dirty” agents, and targeted therapy certainly is still more experimental than proven therapy.
The next target, ErbB2, is a very complex issue, partly because of the variety of molecular abnormalities, such as ErbB2 gene amplification, exon 20 YVMA insertions, extracellular domain mutations such as S310F, and other rare aberrations likely identifying different subsets with regard to disease biology and treatment responsiveness.9 In addition, the available literature on the most commonly used drug, trastuzumab, is nearly impossible to interpret, as most reported experiences used trastuzumab along with chemotherapy and the actual contribution of the targeted agent is very difficult to ascertain.10 Recent data on the use of T-DM1 have generated conflicting information,11 and the use of the first class of dual EGFR/ErbB2 inhibitors such as dacomitinib/neratinib/afatinib has generally led to disappointment, although anecdotal experiences have demonstrated occasional responses. Similar to RET, 2017 suddenly brought some bright light to this field with the development of more potent and specific EGFR/ErbB2 inhibitors that at least biochemically seem to overcome the innate resistance of EGFR and ErbB2 exon 20 insertion mutations, such as poziotinib and AP32788.12,13 So all in all, although interest continues, the data have been sparse and confusing enough that lifting the floodgates to call for mandated general testing for the wide range of ErbB2 alterations would seem in fact premature.
The cleanup hitter in this lineup, MET, provides probably the most compelling story. Despite decades of belief with regard to the involvement of the MET kinase in lung cancer oncogenesis, until 3 years or so ago the actual clinical development of MET inhibitors in lung cancer management had been a recurrent story of tremendous failures with multiple exciting drugs miserably failing in the phase 3 setting, such as the MET Mab antibody as well as the small-molecule MET inhibitor tivantinib.14 The last few years now suddenly seem to have brought clarity and a clear vision for this field with the discovery of recurrent MET kinase aberrations affecting exon 14 of MET leading to exon 14 skipping. These MET exon 14–skipped variants occur in 2% to 3% of all NSCLC with enrichment in the sarcomatoid subtype and mark a tumor subset with quite exquisite sensitivity to several MET inhibitors, most notably the already available compounds crizotinib and cabozantinib, with multiple other more potent MET inhibitors showing high-level activity in addition in phase 2–level studies.15–19 Phase 3 studies are not anticipated given rarity of the subset, high reported activity of agents, and similar precedents suggestive of likely FDA approval with phase 2–level data of possibly several of these agents. Therefore, although testing for such aberrations at diagnosis might not be a necessity, testing through the treatment continuum represents good practice, offering patients an extra line of likely active therapy in the face of a deadly illness. Of course, the complex and numerous molecular alterations leading to MET exon 14 skipping create formidable technical challenges. In addition to MET exon 14 skipping, MET amplification also is starting to rise in the ranks of biomarker validation. Although de novo MET amplification without concurrent MET exon 14 aberrations is rare, if high level it still seems to indicate actionability, and, even more importantly, MET amplification arising in the acquired resistance setting in EGFR-mutated and other molecularly targeted tumors appears to be an uncommon but actionable acquired resistance mechanism.20–22 In particular, recent data suggest that combined EGFR/MET inhibition in this acquired resistance setting will have a clinical impact and MET amplification–mediated acquired resistance might be arising more commonly with even more potent EGFR T790M–targeting inhibitors, such as osimertinib.23
So having considered these 3 key molecular subsets, we should also take into account other up-and-coming biomarkers calling for attention, such as (1) NTRK translocations marking a rare but important subset where TRK inhibitors have great activity,24 (2) tumor mutation burden, and (3) other predictive markers of immunotherapy benefit or lack thereof, such as STK11.25 Although the queue for incorporation into tissue-based biomarker testing continues to grow and the pressure might seem to be mounting on guards of this domain, such as CAP/IASLC/AMP, to strike the right balance between astute resource utilization and best patient management, the good news is that rapid technological developments promise to make some of these issues moot in the very near future, as all biomarker roads seem to be leading towards next-generation–based platforms. Indeed, the large majority of emerging biomarkers are captured very well in such already quite readily available next-generation sequencing (NGS) assays, which can be done at affordable prices on small amounts of tissue. In addition, NGS-based circulating tumor DNA testing provides an excellent complement to tissue-based testing and similarly is very quickly penetrating the market for more or less validated uses.26 Therefore, the recommendation by the CAP/IASLC/AMP expert panel not to endorse single-gene testing for RET/ErbB2 and MET but to recommend ensuring that these 3 emerging markers be included in such test panels if pursued is indeed the right way to approach this Gordian knot. Still, the onus remains on all of us clinicians interacting with patients to ascertain that we appropriately inform our patients about the actual value of biomarker-based treatment interventions, and this calls for continued investment in carefully conducted biomarker-driven clinical studies.
Dr Halmos receives research funding from Merck, Mirati, Astra-Zeneca, Boehringer-Ingelheim, Pfizer, Takeda, and Eli Lilly. Additionally, he consults with Foundation Medicine, Guardant Health360, Pfizer, Eli Lilly, Genentech, Astra-Zeneca, Novartis, Boehringer-Ingelheim, and Takeda.