Clinical trialPhase 3Oesophageal cancer

Pembrolizumab plus chemotherapy versus chemotherapy alone for first-line treatment of advanced oesophageal cancer (KEYNOTE-590): a randomised, placebo-controlled, phase 3 study

Lancet · 2021749 patients

Sun JM, Shen L, Shah MA, et al. Lancet. 2021. 398:759-771

A newer report of this study has been published: Pembrolizumab plus chemotherapy versus chemotherapy for advanced esophageal cancer: 5-year extended follow-up for the randomized phase III KEYNOTE-590 study (2025)

Study type: Clinical trial
Study name: KEYNOTE-590
Phase: Phase 3
Design: Randomised (1:1), double-blind, placebo-controlled trial at 168 centres in 26 countries
Sample size: 749 patients
Source: Lancet · 2021
Trial registration: NCT03189719

In plain words

In this worldwide study, adding the immunotherapy drug pembrolizumab to standard chemotherapy in advanced oesophageal (food pipe) cancer (both the squamous type and adenocarcinoma) helped patients live longer; the benefit was greater in tumours that make a lot of the protein PD-L1. At 5-year follow-up, about 11 of every 100 patients in the pembrolizumab group were alive, compared with 3 in the chemotherapy-alone group.

Key results

13.9 vs 8.8 months

Median overall survival, squamous with CPS ≥10

HR 0.57 (0.43–0.75); p<0.0001

12.4 vs 9.8 months

Median overall survival, all patients

HR 0.73 (0.62–0.86); p<0.0001

13.5 vs 9.4 months

Median overall survival, CPS ≥10

HR 0.62

6.3 vs 5.8 months

Median progression-free survival, all patients

HR 0.65

10.6% vs 3.0%

5-year overall survival (extended follow-up)

Metges et al., ESMO Open 2025; HR 0.72

Drugs in this study:Capecitabine (Xeloda)Cisplatin (Platinol)Fluorouracil (5-FU)Oxaliplatin (Eloxatin)Pembrolizumab (Keytruda)

Dose and schedule cards

Pembrolizumab + cisplatin and fluorouracil (KEYNOTE-590 trial)

Oesophageal or gastroesophageal junction cancer (epicenter 1 to 5 cm above the junction), squamous cell carcinoma or adenocarcinoma, not amenable to surgical resection or definitive chemoradiotherapy; FDA label: PD-L1 CPS ≥1 (greater benefit with CPS ≥10)Cycle length: Every 21 daysOn the page “Oesophageal cancer” · Advanced or metastatic disease, first-line: squamous cell carcinoma (KEYNOTE-590 in both types)

Drug	Dose	Route	Schedule
Pembrolizumab	200 mg (or 400 mg every 6 weeks)	IV (30 minutes), before chemotherapy	Day 1
Cisplatin	80 mg/m² (maximum 6 cycles)	IV	Day 1
Fluorouracil	800 mg/m²/day	IV, continuous infusion	Days 1–5

Cycle length: Every 21 days
Duration: Pembrolizumab until progression, unacceptable toxicity or up to 24 months (35 cycles)

Doses from the FDA label (description of KEYNOTE-590) and the 5-year follow-up publication.

On the page “Oesophageal cancer” · Advanced or metastatic disease, first-line: squamous cell carcinoma (KEYNOTE-590 in both types)

First-line HER2-negative gastric/GEJ and oesophageal cancer + chemotherapy

CPS 1 or more; KEYNOTE-859 and KEYNOTE-590Cycle length: Every 3 weeksOn the page “Pembrolizumab (Keytruda)” · Standard doses and combination regimens (FDA label)

CPS 1 or more; KEYNOTE-859 and KEYNOTE-590

Drug	Dose	Route	Schedule
Pembrolizumab	200 mg every 3 weeks or 400 mg every 6 weeks	IV infusion over 30 minutes	Day 1
Cisplatin + fluorouracil	80 mg/m² (day 1) + 800 mg/m²/day (days 1 to 5)	IV	Every 3 weeks (cisplatin up to 6 cycles in KEYNOTE-590)
Gastric option
CAPOX: oxaliplatin + capecitabine	130 mg/m² + 1000 mg/m² twice a day	IV / oral	Day 1 / days 1 to 14, every 3 weeks

Cycle length: Every 3 weeks
Duration: Until progression, unacceptable toxicity or up to 24 months

In HER2-positive gastric cancer: pembrolizumab before trastuzumab and chemotherapy.

On the page “Pembrolizumab (Keytruda)” · Standard doses and combination regimens (FDA label)

Doses follow the trial protocol, guideline or official drug label (source given in the notes of each regimen). Before prescribing, check blood counts, renal and hepatic function, drug interactions and the patient’s condition, and adjust the dose according to the drug label.

Treatment arms and regimens

1Pembrolizumab + chemotherapy (373 patients)

Pembrolizumab 200 mg IV every 3 weeks (maximum 35 cycles), with cisplatin 80 mg/m² IV every 3 weeks (maximum 6 cycles) and fluorouracil 800 mg/m² per day as a continuous infusion on days 1 to 5 every 3 weeks.

2Placebo + chemotherapy (376 patients)

Placebo (saline) every 3 weeks with the same cisplatin and fluorouracil regimen.

Source of the regimen details: Pembrolizumab dose from the abstract; cisplatin and fluorouracil doses and maximum number of cycles from the trial registration on ClinicalTrials.gov (NCT03189719, arms and interventions section).

Doses are given for the information of specialists, as in this study’s protocol, and are not a prescription. Every drug must be prescribed by the treating physician for the individual patient.

Study population

Adults with previously untreated locally advanced unresectable or metastatic oesophageal cancer (squamous cell carcinoma or adenocarcinoma) or Siewert type 1 gastro-oesophageal junction cancer, regardless of PD-L1 status, with ECOG performance status 0 or 1.

Endpoints

Primary endpoints: overall survival in squamous cell carcinoma with PD-L1 CPS ≥10; and overall survival and progression-free survival in squamous cell carcinoma, in CPS ≥10 and in all randomised patients.

Safety and side effects

Treatment-related grade 3 or higher adverse events occurred in 266 patients (72%) in the pembrolizumab + chemotherapy group and 250 patients (68%) in the placebo + chemotherapy group.

Clinical significance and place in guidelines

KEYNOTE-590 was the first trial to show a survival benefit from adding immunotherapy to first-line chemotherapy in oesophageal cancer and became the basis for approval of pembrolizumab in this setting. In the current FDA label, use of pembrolizumab with chemotherapy in advanced oesophageal cancer is restricted to tumours with PD-L1 CPS ≥1. The 5-year follow-up (Metges et al., ESMO Open 2025) confirmed that the benefit persists.

Limitations

Benefit is limited in tumours with low PD-L1 expression; the improvement in median progression-free survival in the overall population was small (6.3 versus 5.8 months); the chemotherapy regimen (cisplatin and fluorouracil) differs from regimens more commonly used today, such as FOLFOX in adenocarcinoma.

Original article (DOI)PubMed ClinicalTrials.gov

This summary was prepared by the editorial team from the published paper; the original paper is the final reference. Read the original abstract on PubMed.