PD-L1 tumour proportion score (TPS)

What it is

PD-L1 tumour proportion score (TPS) is the percentage of viable tumour cells showing PD-L1 staining, often grouped as <1%, 1-49%, or ≥50%. It is a pathology test result used mainly in non-small cell lung cancer (NSCLC) to help estimate whether a patient may benefit from certain immune checkpoint inhibitors, especially drugs that target the PD-1/PD-L1 pathway such as pembrolizumab. In practical terms, a pathologist examines a tumour sample stained by immunohistochemistry and calculates what proportion of living tumour cells have partial or complete membrane staining for PD-L1.

TPS is different from other PD-L1 scoring systems. It counts tumour cells only, not immune cells in the background. That matters because different cancers and different drugs may use different PD-L1 assays and scoring methods.

A simple way to read TPS is:

TPS category	What it means
<1%	Very low or no detectable PD-L1 staining on tumour cells
1-49%	Some PD-L1 expression
≥50%	High PD-L1 expression

These cutoffs are widely used in lung cancer trials and treatment decisions, but they are not a guarantee of response or non-response.

How it works

PD-L1 is a protein that can be expressed on tumour cells. When PD-L1 binds to PD-1 receptors on T cells, it can reduce the immune system's ability to attack the cancer. This is one way tumours may evade immune surveillance. Drugs such as pembrolizumab block this pathway and can restore anti-tumour immune activity in some patients.

TPS is measured by immunohistochemistry (IHC) on tumour tissue. The pathologist identifies viable tumour cells and estimates the percentage with membrane staining for PD-L1. Necrotic cells, crushed cells, and non-tumour cells are not meant to be counted in the TPS numerator. Because the score depends on tissue quality, staining platform, antibody clone, and reader interpretation, the result can vary between laboratories and assays.

In NSCLC, TPS has become clinically important because higher PD-L1 expression, especially TPS ≥50%, has been associated in many studies with a greater chance of benefit from single-agent PD-1 blockade. However, PD-L1 is an imperfect biomarker. Some patients with high TPS do not respond, and some with low or negative TPS still respond, particularly when immunotherapy is combined with chemotherapy.

Evidence and uses

TPS is used most often in advanced NSCLC to guide first-line treatment choices. A high TPS may support use of single-agent pembrolizumab in appropriately selected patients, while lower TPS results may still be relevant when considering chemo-immunotherapy combinations. Reviews of the KEYNOTE studies and related NSCLC literature commonly use the categories <1%, 1-49%, and ≥50% because treatment effects differed across these groups.

A retrospective study of first-line pembrolizumab in NSCLC reported better overall survival in patients with high PD-L1 TPS (≥50%) than in those with lower TPS, although the study also showed that outcomes vary by tumour histology and clinical context. Other studies have used TPS to examine prognosis in resected NSCLC, but prognostic meaning is less consistent than predictive use for immunotherapy selection.

Important points about clinical use:

1. Most established in NSCLC: TPS is best known in lung cancer, especially for pembrolizumab-related decisions.
2. Not the only factor: Oncologists also consider stage, symptoms, performance status, driver mutations such as EGFR or ALK, smoking history, and whether rapid tumour shrinkage is needed.
3. Assay matters: Different PD-L1 tests are not fully interchangeable. A report should ideally state the assay used.
4. Cancer type matters: Some cancers use combined positive score (CPS) rather than TPS. A TPS result from lung cancer should not be assumed to apply the same way in head and neck, gastric, cervical, or other cancers.

In India, PD-L1 testing is available in many tertiary cancer centres and private pathology laboratories, but access, turnaround time, tissue handling, and assay standardisation may vary. For patients, it is reasonable to ask which assay was used, whether the sample was adequate, and whether the result is being interpreted for the specific cancer type and treatment plan.

Diagnosis / how it's measured

TPS is not a blood test. It is measured on a tumour tissue sample obtained from biopsy or surgery. The basic steps are:

1. Sample collection: Tissue is taken from the tumour.
2. IHC staining: The lab applies a PD-L1 antibody stain.
3. Pathologist review: Viable tumour cells are assessed for membrane staining.
4. Score reporting: The percentage of stained tumour cells is reported as TPS.

A pathology report may include the exact percentage or a category such as TPS <1%, TPS 1-49%, or TPS ≥50%. Some reports also mention the antibody clone or platform. If the sample is small, old, poorly preserved, or contains few viable tumour cells, the result may be less reliable.

Common reasons TPS can be difficult to interpret include:

• Tumour heterogeneity: One part of the tumour may express more PD-L1 than another.
• Small biopsies: Tiny samples may not reflect the whole cancer.
• Timing: PD-L1 expression can change over time or after treatment.
• Inter-observer variation: Different pathologists may score borderline cases differently.

Because of these issues, a TPS result should be read together with the full pathology and oncology picture, not in isolation.

When to see a clinician

You should discuss a PD-L1 TPS result with your oncologist if it is being used to make a treatment decision, especially in advanced lung cancer. Ask whether the result changes eligibility for immunotherapy alone, immunotherapy plus chemotherapy, or another approach.

It is also worth asking for clarification if:

• the report does not mention the assay or method
• the sample was very small or old
• the cancer type usually uses CPS rather than TPS
• there are known driver mutations such as EGFR, ALK, or ROS1, which can affect treatment sequencing
• the report says the sample was inadequate or borderline

Patients should not interpret a high TPS as a promise that immunotherapy will work, or a low TPS as proof that it will not. Treatment decisions depend on several clinical factors beyond PD-L1 alone.

Limitations and open questions

TPS is useful, but it is an imperfect biomarker. It does not capture the full complexity of tumour-immune interactions. Response to immunotherapy depends on many factors beyond PD-L1, including tumour mutation burden, immune-cell infiltration, genomic alterations, prior treatment, and the patient's overall condition.

Evidence is strongest for TPS in NSCLC, but even there, limitations remain. Different assays and cutoffs can produce different classifications. Studies comparing TPS with other scoring systems, such as CPS, suggest they may correlate in some settings, but they are not interchangeable across all cancers. Evidence in humans also shows that PD-L1 expression can be dynamic rather than fixed.

Another open question is how best to use TPS alongside other biomarkers. In some patients, a high TPS clearly supports immunotherapy-based treatment, but in others the best strategy is less certain. Ongoing research is trying to improve prediction by combining PD-L1 with molecular testing and clinical features.

For patients and families, the key point is that TPS is a helpful decision tool, not a standalone verdict. If the result seems unclear or does not fit the treatment discussion, asking for a review by the treating oncologist or pathology team is appropriate.