Agilent Technologies has received U.S. Food and Drug Administration approval for PD-L1 IHC 22C3 pharmDx as a companion diagnostic to identify patients with esophageal or gastroesophageal junction carcinoma who may be eligible for treatment with pembrolizumab, Merck’s anti-PD-1 checkpoint inhibitor. The diagnostic, which uses a combined positive score threshold of one or higher to determine PD-L1 expression, is now the only FDA-cleared companion test for this indication and marks the eighth approved use of the platform alongside pembrolizumab across distinct cancer types.

What this approval changes for esophageal cancer treatment selection and biomarker practice

The practical effect of this regulatory action is to formalise the biomarker gatekeeping role that PD-L1 IHC 22C3 pharmDx already played during the clinical trial programme supporting pembrolizumab in this indication. The KEYNOTE-590 phase three study, which established pembrolizumab plus platinum and fluoropyrimidine-based chemotherapy as a first-line standard in locally advanced or metastatic esophageal cancer, used the Agilent assay to centrally assess PD-L1 expression via combined positive score across all enrolled patients. That the FDA’s companion diagnostic approval now locks the CPS-one threshold as the minimum qualifying score for pembrolizumab eligibility reflects how tightly the regulatory agency has linked trial-validated biomarker methodology to label language. For oncologists and pathologists, this means the assay is no longer simply a tool used in research; it becomes a required step in the treatment decision pathway for eligible patients with unresectable or inoperable esophageal or GEJ carcinoma.

Why the CPS threshold matters more than headline survival data in this disease context

The choice of CPS one as the operative threshold for this companion diagnostic approval warrants scrutiny. The KEYNOTE-590 trial demonstrated overall survival benefit across all enrolled patients regardless of PD-L1 status, with a hazard ratio of 0.73 and a median overall survival improvement from 9.8 months to 12.4 months in the total study population. However, the magnitude of benefit was substantially larger in patients with squamous cell histology and higher CPS scores. Five-year follow-up data confirmed that patients with tumours scoring below CPS one did not derive meaningful survival benefit from the addition of pembrolizumab to chemotherapy, providing the biological rationale for using a minimum threshold. This creates a diagnostically important boundary: patients whose tumours express PD-L1 at even minimal levels are included in the eligible population, while true PD-L1-negative cases are excluded from pembrolizumab access under the labelled indication. Industry observers note that the CPS-one floor represents a relatively permissive cutoff compared with the CPS-ten threshold applied in some European markets, meaning the diagnostic approval may open access to pembrolizumab for a broader patient group in the United States than in jurisdictions where regulatory guidance has taken a more conservative stance on biomarker stringency.

How this new indication differs from Agilent’s earlier approvals in adjacent GI tumour types

The esophageal and GEJ carcinoma approval is distinct from the platform’s prior clearances in gastric or GEJ adenocarcinoma and esophageal squamous cell carcinoma, both of which were earlier indications for the same diagnostic. The current approval covers a broader combined carcinoma category that includes adenocarcinoma histology alongside squamous disease in the esophagus and the GEJ region. This is not a minor distinction. Clinicians tracking the field note that adenocarcinoma and squamous cell subtypes of esophageal cancer differ substantially in epidemiology, molecular profile, and immunological behaviour, and that the benefit signal from pembrolizumab has been less uniform across histologies than the headline trial results suggest. The KEYNOTE-590 five-year data showed a five-year overall survival rate of 11.8 percent for squamous cell carcinoma compared with 7.1 percent for adenocarcinoma, reinforcing that while a survival benefit exists across the full population, it is squamous histology and higher PD-L1 expression that drives the most durable outcomes. The companion diagnostic’s role in this setting therefore extends beyond a binary eligible or ineligible determination; it provides the initial data point in what is increasingly understood to be a continuous biomarker gradient that influences the degree of benefit rather than simply its presence or absence.

What the expanding platform footprint reveals about Agilent’s commercial diagnostics strategy

Eight companion diagnostic approvals for a single assay platform across distinct oncology indications represents an unusual level of regulatory entrenchment. For Agilent, the accumulated approvals for PD-L1 IHC 22C3 pharmDx create a commercial position that is difficult for competing platforms to displace, not because the assay is necessarily superior in analytical performance, but because the clinical trial evidence base was built specifically using this test. Each new pembrolizumab indication that references the Agilent platform in its pivotal study design becomes another anchor for the diagnostic in routine clinical use. The breadth now spans non-small cell lung cancer, esophageal squamous cell carcinoma, cervical cancer, head and neck squamous cell carcinoma, triple-negative breast cancer, gastric or GEJ adenocarcinoma, epithelial ovarian and peritoneal carcinoma, and now esophageal or GEJ carcinoma. Regulatory watchers note that this expanding footprint has occurred across a period of rapid market development for companion diagnostics more broadly, and that Agilent’s repeated use as the designated test in Merck-sponsored pembrolizumab trials reflects the depth of an established co-development relationship rather than episodic collaboration.

Risks and unresolved questions that the companion diagnostic approval does not address

Despite the regulatory clarity this approval provides, significant questions remain in how PD-L1 testing will translate from controlled trial conditions into routine pathology practice for esophageal and GEJ carcinoma. Tumour heterogeneity in esophageal cancer is a known challenge; PD-L1 expression can vary across tumour regions and between primary and metastatic sites, which introduces variability that a single biopsy specimen may not fully capture. The CPS metric itself, which counts PD-L1-staining immune cells in addition to tumour cells, adds interpretive complexity that depends on the skill and training of the reviewing pathologist. Industry observers tracking companion diagnostics implementation note that inter-laboratory concordance for CPS scoring has been an area of ongoing concern in other tumour types where the metric is used, and esophageal cancer is unlikely to be exempt from these practical constraints. The five-year survival rate for esophageal cancer of 21.9 percent underscores how aggressive the disease remains even with immunotherapy access, and the companion diagnostic approval does not alter the underlying prognosis or resolve the broader question of what additional biomarkers may better predict durable benefit from PD-1 blockade in this population. Microsatellite instability status, tumour mutational burden, and other emerging molecular markers are being evaluated as potential refinements to patient selection, but none has achieved the regulatory traction that PD-L1 scoring currently holds.

What pathologists and oncologists are likely to watch in the near term

For pathology laboratories already running PD-L1 IHC 22C3 pharmDx for other tumour types, the esophageal and GEJ carcinoma indication adds volume without requiring new platform investment, which is operationally straightforward. The more consequential question for clinical teams is how patient selection will function in the context of a disease where the line between esophageal carcinoma and GEJ adenocarcinoma is defined anatomically and can be clinically ambiguous. Siewert classification of junctional tumours, tumour epicentre localisation, and the requirement that the carcinoma not be amenable to surgical resection or definitive chemoradiation all introduce eligibility criteria that sit upstream of biomarker testing and must be resolved before the companion diagnostic’s CPS result becomes the operative variable. Clinicians managing these patients will also be watching how reimbursement policies evolve, given that pembrolizumab-based regimens in this setting add meaningful cost to a treatment course that already includes platinum and fluoropyrimidine chemotherapy. The companion diagnostic’s formal approval provides the billing and coverage framework within which that cost calculation will be made, but payer decisions on access and step therapy in this indication remain a factor that oncology practices will navigate case by case.

  Agilent Technologies, Biomarkers, checkpoint inhibitors, companion diagnostics, esophageal cancer, FDA, FDA approval, immunotherapy, KEYNOTE-590, Merck, oncology, PD-L1, PD-L1 IHC 22C3 pharmDx, pembrolizumab