PD-L1 Testing Defines New First-Line Option in TNBC

The approval turns a biomarker result into a treatment gateway

In June 2026, the FDA approved sacituzumab govitecan-hziy (Trodelvy) with pembrolizumab for adults with unresectable locally advanced or metastatic triple-negative breast cancer whose tumors express PD-L1 at a combined positive score of at least 10. The indication covers first-line treatment of advanced disease, and patients must be selected with an FDA-approved test.

That makes the pathology report more than a document filed after a biopsy. Clinicians need to see a CPS of 10 or higher before they can say the patient fits the biomarker portion of the label. A result below 10 falls outside the approved population.

PD-L1 testing therefore has to happen early enough to inform first-line treatment whenever the patient’s condition allows the wait. The dividing line is straightforward. Reading the result is not always so simple.

CPS is not the percentage of tumor cells that stain for PD-L1. The calculation counts PD-L1-staining tumor cells, lymphocytes and macrophages, divides that total by the number of viable tumor cells, then multiplies by 100, which means a tumor proportion score or an immune-cell score cannot be quietly substituted for CPS. A report that says only “PD-L1 positive” leaves an important question unanswered.

The approval also does not make the combination an option for every patient with metastatic TNBC. The cancer must meet the label’s triple-negative definition, must be unresectable locally advanced or metastatic, and must have CPS ≥10 on the companion diagnostic. The pivotal evidence came from patients receiving first-line therapy for advanced disease, so it does not directly establish benefit after previous systemic treatment in that setting.

What the randomized trial showed

The FDA decision was supported by ASCENT-04/KEYNOTE-D19, an open-label phase 3 trial that enrolled 443 patients with previously untreated, inoperable locally advanced or metastatic TNBC and PD-L1 CPS ≥10. Participants were randomly assigned to sacituzumab govitecan plus pembrolizumab or to pembrolizumab with chemotherapy selected by the investigator.

Control-group chemotherapy reflected established first-line regimens. It could be paclitaxel or nab-paclitaxel, with gemcitabine plus carboplatin as another option. The primary endpoint was progression-free survival assessed through blinded independent central review under RECIST criteria. Overall survival and tumor-response measures were additional outcomes.

After a median follow-up of about 14 months, median progression-free survival was 11.2 months with sacituzumab govitecan plus pembrolizumab and 7.8 months with pembrolizumab plus chemotherapy. The hazard ratio for progression or death was 0.65, with a 95% confidence interval of 0.51 to 0.84 and a reported p value of 0.0009.

Overall survival data were not mature at the primary analysis. The progression-free survival result supports the approval, but it cannot yet be described as proof that the regimen helps patients live longer.

Both trial groups received pembrolizumab, an important point that can get lost in a quick reading of the table. The study asked whether using sacituzumab govitecan in place of conventional chemotherapy improved outcomes within a PD-L1-selected immunotherapy approach. It did not compare the new combination with chemotherapy alone.

Testing should precede first-line pathway selection

For clinicians and health systems, implementation begins before a regimen reaches the formulary. PD-L1 testing needs to be ordered soon enough, performed with the FDA-approved companion diagnostic and reported as a numeric CPS. The pathology report is where those parts meet.

Archived tissue from the primary tumor or a metastatic site may be usable if it was preserved well and contains enough viable tumor and immune cells for scoring. Sometimes the block has already been depleted by earlier testing. Sometimes the remaining material is technically inadequate or may no longer represent the current disease, and in those cases a new biopsy may be considered if obtaining one is clinically feasible.

The label does not support inferring CPS from a different PD-L1 scoring system. It also does not support replacing the validated companion diagnostic with an unvalidated assay, even when an older report contains language that sounds close enough.

A report limited to “positive” or “negative” can hide whether the CPS 10 threshold was reached. The useful report identifies how the tissue was tested, makes clear that CPS was used and gives the number; if the result sits at the boundary, specimen adequacy and the laboratory’s quality controls deserve attention rather than informal rounding after the fact.

Other biomarkers still matter, but they answer other treatment questions. Germline BRCA1 or BRCA2 status can inform the use of PARP inhibitors. HER2 expression may shape later-line antibody-drug conjugate choices. Neither a BRCA result nor a HER2-low finding establishes eligibility for this first-line PD-L1-guided combination.

Limited tissue adds pressure to the sequence of testing. If several clinically consequential analyses are likely to be needed, teams have to coordinate them before the specimen is exhausted, rather than discovering after treatment planning has begun that there is no viable material left for CPS.

Coverage and laboratory access are part of the picture too. Delays may be more visible when a patient is diagnosed in a community setting or when the original biopsy yielded little tissue, but the underlying issue is the same: CPS testing determines whether the FDA-labeled combination applies. It is not an optional test to add after the first-line choice has already been made.

Limits and unanswered issues

ASCENT-04 was open label, meaning patients and investigators knew which treatment had been assigned. Blinded independent central review reduced the chance that this knowledge determined the primary progression endpoint, though it does not erase every limitation of an open-label design.

The findings apply most directly to trial-eligible patients with CPS ≥10 who had not received systemic therapy for advanced TNBC. They should not be carried automatically into CPS-low disease or later treatment lines. Patients whose other health conditions would have kept them out of the trial may also face a different balance of benefit and risk.

Fourteen months of median follow-up was enough for the reported progression-free survival analysis, but it was still early for a full view of overall survival and longer-term toxicity. Mature survival results, information about treatments received after progression and patient-reported outcomes could change how clinicians understand the regimen’s value. Trial sponsorship and investigators’ financial relationships also belong in the reading of the evidence alongside the FDA’s regulatory review.

Sacituzumab govitecan has clinically important risks, including neutropenia and diarrhea. Adding pembrolizumab brings the possibility of immune-mediated adverse reactions. The approval creates another choice for eligible patients, while performance status, organ function, prior treatment given with curative intent, vulnerability to toxicity and patient preferences remain part of the conversation.

The CPS number on the pathology report opens the door. It does not settle everything that follows.

Questions clinicians ask

Does any PD-L1-positive result establish eligibility?

No. The first-line combination is approved for unresectable locally advanced or metastatic TNBC with PD-L1 CPS ≥10 as determined by an FDA-approved test. A tumor proportion score or immune-cell percentage cannot be assumed to meet that requirement, and neither can a report that leaves out the scoring method.

Should treatment wait for the CPS result?

When the clinical situation allows it, CPS should be available before first-line therapy is selected, since the result determines whether the labeled pembrolizumab combination applies. Rapidly progressing disease can make the timing harder. Promptly sending the tissue and avoiding incomplete or nonvalidated testing can reduce delays that do not need to happen.

Can an archival specimen be used for PD-L1 testing?

It may be suitable if it meets the companion diagnostic’s technical requirements and contains enough viable tumor for reliable scoring. A new biopsy may be considered when older tissue is inadequate, depleted or no longer clinically representative, with feasibility and procedural risk taken into account.

Has the combination been shown to improve overall survival?

Not at the reported primary analysis. The phase 3 trial showed a statistically significant progression-free survival benefit, while overall survival data remained immature. For now, the evidence supports better disease control in the selected population, and the unresolved survival question remains beside the numeric CPS on the pathology report.

Questions people ask

What PD-L1 result is required for sacituzumab govitecan plus pembrolizumab?

The approved first-line population has unresectable locally advanced or metastatic triple-negative breast cancer with PD-L1 CPS ≥10 on an FDA-approved test. The story found that a tumor proportion score, immune-cell percentage or report saying only “positive” does not establish eligibility.

How effective was sacituzumab govitecan plus pembrolizumab in the trial?

In the phase 3 trial, median progression-free survival was 11.2 months with the new combination and 7.8 months with pembrolizumab plus chemotherapy. Overall survival data were still immature, so the findings did not yet show that patients lived longer.

Can archived tumor tissue be used for PD-L1 CPS testing?

Archived tissue may be usable when it is well preserved, technically adequate and contains enough viable tumor and immune cells for scoring. The story noted that a new biopsy may be considered when older material is depleted, inadequate or no longer representative, if obtaining one is clinically feasible.

References

1. FDA approves sacituzumab govitecan-hziy monotherapy and combination with pembrolizumab for first-line treatment of TNBC — U.S. Food and Drug Administration, 2026
2. Study of Sacituzumab Govitecan-hziy and Pembrolizumab Versus Treatment of Physician's Choice and Pembrolizumab in Previously Untreated Advanced Triple-Negative Breast Cancer — ClinicalTrials.gov, 2022
3. Pembrolizumab plus Chemotherapy in Advanced Triple-Negative Breast Cancer — The New England Journal of Medicine, 2022