KEYNOTE-024: What the Lung Cancer Trial Found

A test result decided who could enter

KEYNOTE-024 did not enrol people by stage or by tumour type alone. It enrolled them by a laboratory measure: PD-L1 protein present on at least half of the cancer cells, scored with a specific assay. People with EGFR or ALK changes were excluded.

That design is the reason PD-L1 testing became routine in lung cancer. The trial only makes a claim about the group the test picks out.

Who took part

Field	Detail
Trial	KEYNOTE-024
Identifier	NCT02142738
Phase	Phase 3
Design	Randomised, open-label, crossover permitted
Cancer type	Untreated advanced non-small-cell lung cancer, PD-L1 on at least 50% of tumour cells
Comparator	Pembrolizumab given as an infusion every three weeks, against platinum-based chemotherapy
Primary endpoint	Progression-free survival

305 people were randomised: 154 to pembrolizumab and 151 to chemotherapy. If the cancer grew on chemotherapy, people could cross over to pembrolizumab.

What the first report showed in 2016

Median progression-free survival was 10.3 months with pembrolizumab and 6.0 months with chemotherapy (hazard ratio 0.50, 95% CI 0.37 to 0.68, p<0.001). At six months, 80.2% of the pembrolizumab group were alive against 72.4% (hazard ratio 0.60, 95% CI 0.41 to 0.89, p=0.005). Tumours shrank in 44.8% against 27.8%.

Side effects ran the other way from what people expect of a cancer drug. Treatment-related events of any grade affected 73.4% of the pembrolizumab group and 90.0% of the chemotherapy group. Severe ones — grade 3 to 5 — affected 26.6% against 53.3%.

What the five-year report added

Follow-up reached a median of 59.9 months. Median overall survival was 26.3 months (95% CI 18.3 to 40.4) with pembrolizumab and 13.4 months (95% CI 9.4 to 18.3) with chemotherapy, a hazard ratio for death of 0.62 (95% CI 0.48 to 0.81).

At five years, an estimated 31.9% of the pembrolizumab group were alive, against 16.3% of the chemotherapy group.

The 39 people who finished two years of treatment

Pembrolizumab was capped at 35 cycles, roughly two years. 39 people completed all of them. At the five-year cutoff, 82.1% of those 39 were still alive.

That is a strikingly good figure, and it needs care. These are people who tolerated two full years of treatment without their cancer growing — a group already selected for doing well. It says something about how durable a response can be. It does not predict anything for someone starting treatment today.

Crossover complicates the comparison

99 of the 151 people assigned chemotherapy went on to receive an anti-PD-1 or PD-L1 drug, an effective crossover rate of 66.0%.

That does not undermine the result. If anything it works against pembrolizumab, since the control group got the drug too, just later. But it does change what is being measured. The comparison is between starting with immunotherapy and starting with chemotherapy — not between having the drug and never having it.

What this report cannot tell you

• Anything about lower PD-L1 levels. Everyone here had PD-L1 on at least half of tumour cells.
• Whether it applies with an EGFR or ALK change. Those people were excluded.
• How pembrolizumab alone compares with pembrolizumab plus chemotherapy. That comparison is not in this trial.
• What a five-year figure means for one person. 31.9% describes a group of 154.

Questions about PD-L1 testing

• What was my PD-L1 score, and which assay was used?
• Does my score put me in the group this trial studied?
• If we start with immunotherapy alone, what is the plan if it does not work?

Sources

This article was written from the sources below, which were checked on the source-check date shown above.

• NCI: Pembrolizumab for Non-Small Cell Lung Cancer Improves Survival (KEYNOTE-024) (official)
• NEJM: Pembrolizumab versus Chemotherapy for PD-L1-Positive NSCLC (primary)
• JCO: Five-Year Outcomes With Pembrolizumab vs Chemotherapy (KEYNOTE-024) (primary)

How this article was prepared

Prepared by Cancer Explained's AI-assisted editorial system and checked against the sources listed below. This article has not been reviewed by a healthcare professional unless a named reviewer is specifically shown.

Cancer Explained is published by the National Cancer Information Foundation as a nonprofit-oriented public-interest education project. It is not a diagnostic service, does not recommend treatments, and is not for emergencies.

Found an error, a broken source link, outdated information, or wording that feels insensitive? Report it here — we log and act on material corrections.