Detecting HER2 Mutations in ctDNA: What a Blood Test Can (and Can't) Catch

The clinical question

When a patient presents with metastatic non-small cell lung cancer (NSCLC), the modern workup asks a battery of molecular questions in parallel: is there a targetable driver, and if so, which drug class does it open the door to? For a subset of these patients, roughly 2, 4% of NSCLC, the answer is an activating mutation in ERBB2, the gene encoding HER2 [5]. That's a specific and important distinction: we're talking about activating mutations here, most commonly small insertions in exon 20, not HER2 amplification or protein overexpression, which are separate biological phenomena scored by different assays [6][7].

The clinical stakes are concrete. An activating HER2 mutation gates eligibility for a HER2-directed antibody, drug conjugate (ADC), trastuzumab deruxtecan (T-DXd). The liquid-biopsy question, then, is narrow but consequential: can a blood draw reliably identify these patients so treatment isn't delayed, or missed entirely, when tumor tissue is scarce, exhausted, or hard to reach?

What's measured, and how

The analyte is circulating tumor DNA (ctDNA), fragmented tumor-derived DNA shed into the bloodstream, isolated from a plasma sample. The assay is targeted next-generation sequencing (NGS), the same core technology used on formalin-fixed tissue, but tuned to detect low-frequency variants against a background of normal cell-free DNA.

Two FDA-approved plasma platforms are relevant here. In August 2024, the FDA approved Guardant360 CDx as a companion diagnostic to identify HER2 (ERBB2) activating mutations in NSCLC and select patients for first-line T-DXd. It was the first liquid biopsy companion diagnostic approved specifically for this indication, and it opens a plasma-first selection pathway that doesn't mandate a tissue biopsy [1]. Separately, FoundationOne Liquid CDx has held FDA approval since November 2022 as a pan-solid-tumor plasma NGS companion diagnostic whose coverage includes ERBB2 mutations, a second FDA-cleared ctDNA route for HER2 mutation detection in NSCLC [2].

A word of caution on sensitivity, because this is where plasma testing is most often overstated. ctDNA assays detect what's shed into circulation, and shedding varies enormously with tumor burden, site, and biology. A negative plasma result does not rule out a HER2 mutation; it means none was detected in that sample. This is the central asymmetry of liquid biopsy: a positive result is actionable, but a negative one is not exculpatory, plasma genotyping is specific yet only moderately sensitive, bounded by how much tumor DNA is shed [9][10][11], and the positive-percent-agreement gap seen directionally across drivers (plasma recovers most of what tissue finds only when shedding is adequate) shows the point [13][14]. Where plasma comes back wild-type in a patient with strong clinical suspicion, reflex to tissue NGS remains the appropriate move [12]. The evidence suggests plasma works best as a fast, accessible complement to, not a wholesale replacement for, tissue genotyping.

The evidence

The clinical foundation for T-DXd in HER2-mutant NSCLC comes from the DESTINY-Lung program: DESTINY-Lung01 established the response signal (objective response rate 55%, 50/91) [6], and the DESTINY-Lung02 phase II trial, published in 2023, then established the tolerable dose of the ADC in this population [3][8]. Importantly for the liquid-biopsy story, HER2 mutation status in enrolled patients was determined by NGS [3][8]. The specific question of whether patients selected on plasma-detected ERBB2 mutations derive equivalent benefit, a tissue-versus-plasma subgroup split, is not broken out in the evidence we hold, so it is best read as a reasonable expectation the companion-diagnostic framework is built on, rather than a demonstrated subgroup result. That predictive linkage, plasma-detected mutation to drug-class benefit, is precisely what a companion diagnostic approval requires.

Concordance between tissue and plasma NGS for actionable NSCLC mutations continues to accumulate, though a primary HER2-mutation plasma-versus-tissue concordance study is not yet established in our evidence base. Data summarized in a 2025 ASCO conference-highlights review point to emerging plasma-first HER2 testing, including exon 20 insertions, and the turnaround-time advantage of a blood draw where tissue is limited, framed, appropriately, as maturing rather than definitive [4]. These are encouraging signals for upfront liquid use, but this concordance literature is still maturing and remains, in the strict regulatory sense, investigational for many use-cases beyond the specific approved companion Dx indications. Concordance is not symmetric: positive percent agreement (does plasma find what tissue found?) is more clinically fraught than negative percent agreement, and it's the metric most sensitive to low ctDNA shedding.

Validated versus investigational

The picture divides cleanly into validated and investigational domains.

On the validated side, plasma ctDNA NGS for ERBB2 activating mutations in NSCLC now has a firm regulatory footing. With Guardant360 CDx approved in August 2024 as a dedicated liquid-biopsy companion diagnostic [1], and FoundationOne Liquid CDx available as an FDA-approved pan-tumor plasma option covering ERBB2 [2], clinicians have two cleared pathways to identify HER2-mutant NSCLC from blood and thereby establish eligibility for the HER2-directed ADC class [3]. For a patient whose tissue is insufficient for molecular testing, a common and frustrating scenario in advanced lung cancer, this is a genuine advance. It can shorten the interval to an actionable result and spare a repeat invasive procedure.

What remains investigational, or at least unsettled, is everything beyond this front-line selection question. Using ctDNA to monitor molecular residual disease (MRD), to track response dynamics, or to guide treatment switches in HER2-mutant NSCLC is not part of the approved companion diagnostic framework. The concordance and turnaround-time work surfacing at ASCO 2025 supports upfront liquid use where tissue is constrained [4], but MRD-guided decision-making in this setting stays in the research column for now. It's a fast-moving area, and readers should treat claims about plasma-guided monitoring skeptically until prospective outcome data catch up.

The practical synthesis for a mixed audience: if you or your patient has advanced NSCLC and tissue genotyping is delayed or impossible, an FDA-approved plasma NGS test can now identify an activating HER2 mutation and open the door to a specific ADC drug class. A positive plasma result is actionable. A negative one warrants tissue confirmation, not reassurance. And using these same assays to make longitudinal, MRD-based treatment decisions is a promise the evidence hasn't yet delivered on, an important boundary to hold as the field moves quickly.

---

References

1. U.S. Food and Drug Administration. FDA approves fam-trastuzumab deruxtecan-nxki for unresectable or metastatic HER2-mutant non-small cell lung cancer and companion diagnostic. FDA Oncology (Hematology/Oncology) Approval; PMA supplement / De Novo for Guardant360 CDx. 2024. (FDA-CDx-Guardant360-HER2-NSCLC-2024)

2. Foundation Medicine, Inc. FoundationOne Liquid CDx, FDA-approved companion diagnostic and complete genomic profiling test. Foundation Medicine product page / FDA PMA P200010. 2022. (FDA-PMA-P200010-FoundationOneLiquidCDx)

3. Li BT, Smit EF, Goto Y, et al. Trastuzumab Deruxtecan in HER2-Mutant Non-Small-Cell Lung Cancer (DESTINY-Lung02). New England Journal of Medicine. 2023;389(6):514-526. (NEJM-DESTINY-Lung02-2023)

4. Steendam CMJ et al. (summarized in ASCO 2025 ctDNA highlights review). Circulating tumour DNA assays in focus: highlights from ASCO 2025 for clinicians. PMC / ASCO 2025 proceedings summary, PMC12449440. 2025. (PMC12449440)

5. Pillai RN, et al. HER2 mutations in lung adenocarcinomas: A report from the Lung Cancer Mutation Consortium. Cancer. 2017. PMID: 28743157. (HER2 mutation prevalence ~3% (24/920); vault S602.)

6. Li BT, Smit EF, Goto Y, et al. Trastuzumab Deruxtecan in HER2-Mutant Non-Small-Cell Lung Cancer (DESTINY-Lung01). N Engl J Med. 2022. PMID: 34534430. DOI: 10.1056/NEJMoa2112431. (ORR 55%, 50/91; vault S515.)

7. Li BT, et al. Ado-Trastuzumab Emtansine for Patients With HER2-Mutant Lung Cancers. J Clin Oncol. 2018. PMID: 29989854. DOI: 10.1200/JCO.2018.77.9777. (HER2 mutation distinct from amplification; vault S516.)

8. Goto K, et al. Trastuzumab Deruxtecan in HER2-Mutant Non-Small-Cell Lung Cancer (DESTINY-Lung02). J Clin Oncol. 2023. PMID: 37694347. DOI: 10.1200/JCO.23.01361. (DESTINY-Lung02, vault's trial record; distinct publication from ref 3; vault S630.)

9. Leighl NB, et al. Clinical Utility of Complete Cell-free DNA Analysis to Identify Genomic Biomarkers in Newly Diagnosed Metastatic Non-small Cell Lung Cancer (NILE). Clin Cancer Res. 2019. PMID: 30988079. DOI: 10.1158/1078-0432.CCR-19-0624. (vault S638.)

10. Douillard JY, et al. Gefitinib treatment in EGFR mutated Caucasian NSCLC: circulating-free tumor DNA as a surrogate for determination of EGFR status (IFUM). Br J Cancer. 2014. PMID: 24263064. DOI: 10.1038/bjc.2013.721. (plasma high specificity / moderate sensitivity; vault S466.)

11. Mok T, et al. Detection and Dynamic Changes of EGFR Mutations from Circulating Tumor DNA as a Predictor of Survival Outcomes in NSCLC Patients (FASTACT-2). Clin Cancer Res. 2015. PMID: 25829397. DOI: 10.1158/1078-0432.CCR-14-2594. (concordance 88%, blood sens 75%, spec 96%; vault S640.)

12. National Complete Cancer Network (NCCN). Clinical Practice Guidelines in Oncology, Non-Small Cell Lung Cancer (notes). NCCN. (Plasma-negative reflex-to-tissue posture, attributed; vault S626.)

13. Mondaca S, et al. Lung adenocarcinoma ALK/ROS1/RET fusion concordance between plasma ctDNA NGS and tissue. Lung Cancer. 2021. PMID: 34311346. DOI: 10.1016/j.lungcan.2021.06.018. (directional concordance 93% plasma→tissue vs 54% tissue→plasma; vault S643.)

14. Dagogo-Jack I, et al. Tracking the Evolution of Resistance to ALK/ROS1 Inhibitors, ROS1 plasma ctDNA NGS versus tissue. J Thorac Oncol. 2019. PMID: 30664990. DOI: 10.1016/j.jtho.2019.01.009. (ROS1 plasma-vs-tissue concordance asymmetry; vault S656.)