ProtecT: What the Prostate Cancer Trial Found

The question ProtecT asked

Most trials ask whether a new treatment beats an old one. ProtecT asked something harder: whether treating a cancer at all was better than watching it.

Between 1999 and 2009, 82,429 men aged 50 to 69 in the United Kingdom had a PSA test. A raised prostate-specific antigen led to further testing, and 2,664 men were diagnosed with localized prostate cancer. Of those, 1,643 agreed to be randomly assigned to one of three paths:

• Active monitoring, with PSA-based follow-up: 545 men.
• Radical prostatectomy, surgery to remove the prostate: 553 men.
• External-beam radiotherapy: 545 men.

The main measure was death from prostate cancer at a median of 10 years. Results appeared in the New England Journal of Medicine in 2016.

Why the question needed asking

Prostate cancer is common and usually slow. The American Cancer Society projects 333,830 new United States cases in 2026, a figure SEER carries alongside its own median age at diagnosis of 68. About 69% are found while still confined to the prostate, and five-year relative survival for that group is 100.0%.

Many of these cancers would never cause symptoms in a man's lifetime. But surgery and radiotherapy carry lasting costs — urinary leakage, erectile dysfunction, bowel problems. Treating a cancer that was never going to hurt someone means paying those costs for nothing. Nobody knew how often that was happening. Our prostate cancer page covers how the disease is graded and staged.

What happened at 10 years

Deaths from prostate cancer were rare in every group, and the differences were not statistically significant:

• Active monitoring: 8 deaths, 1.5 per 1,000 person-years.
• Surgery: 5 deaths, 0.9 per 1,000 person-years.
• Radiotherapy: 4 deaths, 0.7 per 1,000 person-years.

That is 17 prostate cancer deaths across 1,643 men, with P=0.48 for the comparison. Deaths from any cause were also similar: 169 in total, P=0.87.

Two things did differ. The cancer spread to distant sites in 33 men on monitoring, against 13 after surgery and 16 after radiotherapy (P=0.004). And disease progressed in 112 men on monitoring, against 46 in each treatment group.

What happened at 15 years

The team published longer follow-up in 2023, with data on 98% of participants.

By then 45 men (2.7%) had died of prostate cancer: 17 in the monitoring group, 12 after surgery, 16 after radiotherapy. P=0.53. Deaths from any cause reached 356 men, spread similarly across the three groups.

The pattern held. Metastases were more common with monitoring (9.4%, against 4.7% and 5.0%), as was clinical progression (25.9%, against 10.5% and 11.0%).

One number captures why this trial mattered. At the end of 15 years, 133 men in the monitoring group — 24.4% of it — were alive and had never had any prostate cancer treatment at all.

The authors also noted that more than a third of the men had intermediate or high-risk disease at diagnosis, and that they found no differential effect on cancer-specific mortality by PSA level, tumor stage, or grade.

How to read this

The finding is not "treatment does not work." It is that treatment and monitoring produced similar death rates over 15 years, while treatment reduced spread and progression, and monitoring avoided the harms of treatment for a quarter of men entirely.

That turns the decision into a trade-off rather than an emergency. It is the evidence underneath modern active surveillance and behind the recommendation that this be a shared decision.

When to get a PSA conversation started

There is no age at which a PSA test is automatic. The U.S. Preventive Services Task Force says that for men aged 55 to 69, whether to have periodic PSA screening should be an individual decision made after discussing the benefits and harms with a clinician. It advises against screening men aged 70 and over.

The Task Force lists what should feed that discussion: family history, race and ethnicity, other medical conditions, and how a man weighs the possible harms. Those harms are specific — false positives leading to biopsy, overdiagnosis, overtreatment, and treatment complications including incontinence and erectile dysfunction.

Symptoms are a separate matter and need assessment rather than screening:

• Trouble starting or stopping urination, a weak stream, or straining.
• Needing to urinate often, especially waking repeatedly at night.
• Blood in urine or semen.
• New, persistent pain in the lower back, hips, or thighs.

Prostate symptoms are far more often caused by benign enlargement than by cancer. They still warrant an appointment.

What this trial cannot tell you

• The design centered on 10-year mortality. Prostate cancer can take longer than that to prove fatal, so even the 15-year data may be short for a man diagnosed at 50.
• Only 17 prostate cancer deaths occurred by 10 years. With numbers that small, a real but modest difference between treatments could easily hide.
• Monitoring in ProtecT was PSA-based. Modern active surveillance adds MRI and repeat biopsies, so the monitoring arm would be run differently today and might perform better.
• The higher rate of metastases with monitoring is a genuine cost. Equal death rates can obscure it.
• Side-effect differences are central to the real choice and are not in these headline results. Trial phases explains what a trial's primary endpoint does and does not capture.
• Entry required PSA-detected, clinically localized disease in men aged 50 to 69, so the findings do not extend to cancer found because of symptoms, or to more advanced disease.

Sources

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

• Hamdy FC and colleagues, "10-Year Outcomes after Monitoring, Surgery, or Radiotherapy for Localized Prostate Cancer," New England Journal of Medicine, 2016 (PMID 27626136): https://pubmed.ncbi.nlm.nih.gov/27626136/
• Hamdy FC and colleagues, "Fifteen-Year Outcomes after Monitoring, Surgery, or Radiotherapy for Prostate Cancer," New England Journal of Medicine, 2023 (PMID 36912538): https://pubmed.ncbi.nlm.nih.gov/36912538/
• SEER Cancer Stat Facts, Prostate Cancer: https://seer.cancer.gov/statfacts/html/prost.html
• U.S. Preventive Services Task Force, Prostate Cancer: Screening: https://www.uspreventiveservicestaskforce.org/uspstf/recommendation/prostate-cancer-screening
• National Cancer Institute, Prostate Cancer Treatment (PDQ), health professional version: https://www.cancer.gov/types/prostate/hp/prostate-treatment-pdq

How this page was made

An AI-assisted editorial system helped prepare this page. No named medical reviewer has reviewed it unless one is listed.

The National Cancer Information Foundation publishes Cancer Explained. It is not a diagnostic service, does not recommend treatments, and is not for emergencies.

See an error, old source, or unclear wording? Tell us.