The hidden population: How many NRG1 fusion-positive patients are we missing?
Every oncologist has encountered the patient who defies expectations—the KRAS wild-type with advanced pancreatic cancer, the young man with invasive mucinous adenocarcinoma of the lung—cases where something feels biologically distinct, yet the molecular report comes back unremarkable.
For a subset of these patients, the answer is hiding in plain sight: an NRG1 fusion never tested for, tested for with the wrong assay, or missed entirely because there was not enough tissue.
An epidemiology problem hiding in plain sight
NRG1 fusions occur in an estimated 0.2% of solid tumors overall, with enrichment in invasive mucinous adenocarcinoma of the lung, KRAS wild-type pancreatic cancer, and certain biliary and breast cancers. [] []
Applied across millions of new solid tumor diagnoses each year, even 0.2% translates to thousands of patients annually harboring a potentially actionable alteration.
When RNA-based sequencing is used systematically, detection climbs further—one real-world cohort using integrated DNA and RNA profiling identified NRG1 fusions in 0.8% of non-squamous NSCLC, well above historical estimates. []
Related: IMA histology should trigger RNA sequencing in NSCLC—here’s whyA therapy in search of its patients
In December 2024, zenocutuzumab (Bizengri) became the first FDA-approved targeted therapy for adults with advanced NRG1 fusion-positive NSCLC and pancreatic adenocarcinoma. []
But therapy is only as useful as the diagnostic pathway that leads to it.
A therapy in search of its patients
The problem begins with tissue. In pancreatic cancer, 21% to 32% of tissue samples are inadequate for NGS testing, and the NGS success rate at one academic center was just 63.6% for pancreatic adenocarcinoma—the lowest of any tumor type studied—compared with 87.5% for lung cancer. [] []
These are precisely the two cancers for which zenocutuzumab is approved, and for which tissue acquisition is most difficult.
Even when tissue is adequate, testing may still fall short. A recent study found that 67% of NRG1 fusion–positive NSCLC cases were identified only by RNA-based sequencing—yet RNA panels remain underutilized in routine practice. []
A real-world claims analysis of over 50,000 patients found that only 37% of NSCLC patients and 35% of pancreatic cancer patients received any biomarker testing at all, far below guideline recommendations. []
This represents a systems-level failure: testing never ordered, tissue insufficient, wrong platform applied, workflows built around common mutations that silently exclude rare but actionable fusions.
Each breakdown is individually understandable. Collectively, they create a hidden population whose biology could inform treatment but whose diagnosis is never made.
Related: HER2–HER3 blockade in NRG1 fusions: The mechanism changing NSCLC careA collective failure with individual consequences
Resource constraints are real, but they should not become permission to accept the status quo. If even a fraction of projected NRG1 fusion-positive patients are going undetected, the consequence is measured in missed opportunities for patients who now have an approved treatment option.
Closing that gap demands a deliberate rethinking of testing strategy—thoughtful biopsy planning to maximize sample adequacy, tumor-enrichment techniques where feasible, and upfront comprehensive profiling that incorporates RNA sequencing alongside DNA from the outset.
The question is no longer whether NRG1 fusions matter. It is whether our systems are built to find them.