Comprehensive Molecular Testing Powers Biomarker-Driven Care in Advanced NSCLC

The biomarker landscape in advanced non–small cell lung cancer (NSCLC) keeps expanding, but the workflows built to streamline and test for these biomarkers remain a work in progress, according to Sanja Dacic, MD, PhD.

“We have a big elephant in the room: biomarkers for a disease with an ever-growing number of biomarkers,” Dacic said during a presentation at the 27th Annual International Lung Cancer Congress.¹ “The question is: How, where, and when [do we] fit [these biomarkers] into an already complicated workflow?”

In the presentation, Dacic shed light on the importance of molecular profiling and testing in all patients with advanced NSCLC to identify targetable driver mutations and gene fusions, testing for immunotherapy-related biomarkers like PD-L1, and figuring out where biomarkers for antibody-drug conjugates (ADCs) fit into the molecular testing workflow.

Dacic is a professor of pathology and the vice chair and director of Anatomic Pathology at the Yale School of Medicine in New Haven, Connecticut.

What do current guidelines recommend regarding molecular testing in NSCLC? How does the advanced NSCLC molecular pathology workflow play out in practice?

Dacic pointed out how the traditional view of advanced NSCLC pathology has evolved over time. NSCLC was initially considered one unified disease but is now categorized by molecular characterizations and PD-L1 expression levels. Moreover, she noted how future molecular pathology discoveries in advanced NSCLC may push the boundaries even further via protein expressions. Using comprehensive testing panels for mutations and fusions is now standard, and PD-L1 testing has been in place for years, but the guideline updates keep raising the floor.

Per the current National Comprehensive Cancer Network guidelines for molecular testing in advanced NSCLC, the bare minimum is testing for targets with an available therapy, spanning EGFR, ALK, ROS1, BRAF, HER2, MET, RET, NTRK, KRAS, and NRG1, each now matched to at least 1 targeted agent, from osimertinib (Tagrisso) and alectinib (Alecensa) to newer options like zenocutuzumab (Bizengri) for NRG1 fusions.²

Testing is still driven largely by histology, Dacic noted. Adenocarcinomas and tumors with an adenocarcinoma component typically go through the workflow a given institution’s multidisciplinary team has agreed on, whether reflex testing or testing on request.

In terms of non-adenocarcinoma histologies, most labs still test squamous, large cell neuroendocrine, and small cell carcinomas only on request, though large cell neuroendocrine tumors are increasingly folded into reflex protocols, and many centers apply clinical criteria, such as smoking status and age, to decide which patients should be tested.

Liquid biopsies, collected alongside the tissue specimen, help with quicker turnaround times but require thorough examination.

“The bottom line is, if [a patient has] positive results [from liquid biopsy, the] doctor can talk about it. If [they] get a negative result, you have to go for the tissue to make sure it is the right result,” she added.

How are NGS platforms evolving to meet turnaround time demands in advanced NSCLC molecular pathology workflows?

Broad next-generation sequencing (NGS) panels are recommended across guidelines, but the commercial landscape is dense enough that Dacic argued the importance of understanding the technical differences between assays.

“The broad NGS panels are highly recommended…. Be aware of the technical aspects of all the assays in terms of the nucleic acid input and the enrichment method,” she explained.

Nucleic acid input, enrichment method (amplicon-based vs hybrid capture), and gene coverage all vary meaningfully across platforms such as the Oncomine Comprehensive Assay Plus, the QIAseq Pan-Cancer Multimodal Panel, and TruSight Oncology 500.³ However, turnaround time has been a chief complaint about broad NGS panels, according to Dacic. Guidelines cite that these turnaround times can be up to 10 business days, though real-world results often take longer, she said.

Fortunately, rapid NGS platforms now sequence more than 500 genes and return tumor mutational burden and microsatellite instability data in approximately 72 hours, with a full tissue-based workflow completed in under 5 days.

“[Rapid NGS platforms are] exciting; this is the way to go,” she said, adding that turnaround times should keep shrinking as the technology matures.

Why do EGFR, HER2, and MET alterations require careful interpretation?

Dacic used EGFR mutations to illustrate why raw variant identification is not enough. Classical EGFR alterations (exon 19 deletions and exon 21 L858R mutations) account for approximately 80% of EGFR-mutant NSCLC, although atypical mutations (5%-10%) and exon 20 insertions (5%-10%) represent distinct diseases with different frontline regimens.⁴

“You have to read [EGFR mutations] like a fine print in the report,” she said, particularly when pathology flags an unusual mutation and its associated treatment implications.

HER2 alterations are less common but similarly nuanced, she underscored. Mutation hotspots and frequencies differ meaningfully between NSCLC and other tumor types, and knowing whether a variant falls inside or outside the tyrosine kinase domain can inform expected response to therapy.⁵

When testing for MET exon 14 skipping mutations, Dacic highlighted how negative DNA-based NGS results are not the end of the workflow.

“If you get a completely negative DNA result on NGS, you have to go to RNA,” she said, since RNA-based testing captures the diversity of MET splice-site variants.⁶

What challenges do growing predictive IHC assays pose in NSCLC characterization and management?

The proliferation of predictive immunohistochemistry (IHC) assays for PD-L1, HER2, MET, and DLL3 has created its own interpretive burden, Dacic said. It is important to know which antibody clone the assay uses, whether the assay is an approved companion diagnostic or a laboratory-developed test, and what cutoff and interpretation criteria apply.

“PD-L1 is an example of how it took us a few years to sort out what clones to implement and how to interpret them,” she said.

Reflex testing protocols for equivocal results also vary by biomarker and tumor type. However, equivocal ALK or ROS1 IHC results should be reflexed to a molecular method given the assays’ high false-positive rates, she noted.

Ultimately, advanced-stage NSCLC management warrants broad DNA and RNA testing in all patients, immunotherapy-related biomarkers remain part of the standard workflow, and biomarkers for ADCs increasingly need a place in that same workflow, Dacic concluded.

References

1. Dacic S. Molecular profiling in advanced NSCLC: How many tests? How to interpret them? Presented at: 27th Annual International Lung Cancer Congress; July 24-26, 2026; Huntington Beach, CA.
2. NCCN. Clinical Practice Guidelines in Oncology. Non–small cell lung cancer, version 6.2026. June 12, 2026. Accessed July 28, 2026. https://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf
3. Ionescu DN, Stockley TL, Banerji S, et al. Consensus recommendations to optimize testing for new targetable alterations in non-small cell lung cancer. Curr Oncol. 2022;29(7):4981-4997. doi:10.3390/curroncol29070396
4. Ji J, Aredo JV, Piper-Vallillo A, et al. Osimertinib in NSCLC with atypical EGFR-activating mutations: a retrospective multicenter study. JTO Clin Res Rep. 2023;4(3):100459. doi:10.1016/j.jtocrr.2022.100459
5. Robichaux JP, Elamin YY, Vijayan RSK, et al. Pan-cancer landscape and analysis of ERBB2 mutations identifies poziotinib as a clinically active inhibitor and enhancer of T-DM1 activity. Cancer Cell. 2019;36(4):444-457.e7. doi:10.1016/j.ccell.2019.09.001
6. Poirot B, Doucet L, Benhenda S, et al. MET exon 14 alterations and new resistance mutations to tyrosine kinase inhibitors: risk of inadequate detection with current amplicon-based NGS panels. J Thorac Oncol. 2017;12(10):1582-1587. doi:10.1016/j.jtho.2017.07.026