The “ProtecT Trial” Comparing Surgery, Radiotherapy, and Active Monitoring for Men with Localized Prostate Cancer:
Limitations for Informing Current Patient Management
The UK’s “ProtecT” randomized clinical trial compares treatment outcomes from 2,640 men with PSA-based, screen-detected localized prostate cancer randomized to treatment with either “active monitoring, ” radical prostatectomy, or radiotherapy. In 2020, the trial first reported the results after 10 years, showing that men randomized to receive active monitoring had a higher risk of cancer progression – specifically, 1.85% of men on active monitoring died of the disease, compared to 0.67% who had surgery and 0.73% who had radiotherapy. The “patient summary” of this report concluded: “More than 90 out of every 100 men with localized prostate cancer do not die of prostate cancer within ten years, irrespective of whether treatment is by means of monitoring, surgery, or radiotherapy.” In 2023, they reported the 15-year outcomes, concluding that the “prostate cancer-specific mortality was low regardless of the treatment assigned.” However, metastases occurred approximately twice as frequently in the active monitoring group. Recently, they reported on the long-term incidence of metastases in patients with the cribriform-positive and cribriform-negative histological types of prostate cancer in the ProtecT trial. This analysis is “secondary” because this was not one of the trial’s planned analyses. They found that the cribriform morphology was a strong, independent predictor of 15-year metastasis and that radiotherapy with neoadjuvant hormonal therapy was associated with a reduced long- term risk of metastasis. In contrast, outcomes were favorable for most patients with cribriform-negative disease, supporting their eligibility for active surveillance (AS).
Because ProtecT is a randomized controlled trial, it is very influential for patient management guidelines; however, it has significant limitations in informing current patient care. The trial recruited its participants in a PSA-testing setting; they may differ from men diagnosed in other ways or with different screening histories. Also, what is called “lead-time bias” is always a concern (i.e., in screen- detected cancer patients, the cancer may be detected earlier by PSA screening, but this may not necessarily change itsultimate clinical course). Not surprisingly, most ProtecT participants had low- or intermediate-risk prostate cancer. High-risk disease was rare (only~3 %); therefore, the findings may not apply to men with more aggressive disease. ProtecT started in the late 1990s; since then, diagnostics andtreatments have evolved substantially. Multiparametric MRI (mpMRI), which is now a standard in prostate cancer diagnosis, was not used in ProtecT. Surgical techniques have advanced, and radiotherapy has improved. We now use MRI before deciding on surveillance. ProtecT did not use MRI. Today, MRI helps clinicians image the tumor and check if the cancer is truly a low-risk tumor and catches some cancers that random biopsies alone might miss.
The “active monitoring” approach in ProtecT was mainly based on infrequent PSA testing and periodic biopsies, and was less intensive than contemporary AS programs, which usually include more frequent biopsies, MRI, and more stringent criteria. Thus, the risk for disease progression under current AS strategies might differ, and there is always a concern about missing the “window of curability. ” ProtecT enrolled very few non-white men, so its results may not be generalizable to other populations. The published results are at 15 years of follow-up, which might not be enough to capture long-term risks for metastases and prostate cancer deaths fully. Therefore, uncertainty persists beyond 15 years, especially in men with a long life expectancy. Many ProtecT study patients received treatments other than, or in addition to, those they were initially randomized to receive, which confounds the results. Thus, ProtecT may underestimate how safe modern AS can be or overestimate the risk for cancer progression because its monitoring protocol was more conservative by today’s standards.
Medicine has advanced considerably since ProtecT. We now know more about which cancers we can safely monitor, and which ones are more aggressive. Modern guidelines are more personalized and technology-driven than those used in ProtecT. Thus, experts are not taking ProtecT’s findings as the final word. Some have interpreted ProtecT as being informative mainly in the context of men treated with delayed prostatectomy, as many of the monitoring patients ultimately crossed over to prostatectomy versus those who had immediate prostatectomy. In this context, ProtecT shows that delaying prostatectomy did not lead to substantially worse outcomes. However, the new secondary analysis of ProtecT does provide provocative data to support recommending against AS in patients with certain histological types of prostate cancer, in this case, cribriform patterns (Dr. Catalona believes that the same should hold true for the so-called “intraductal carcinoma” histological pattern).
The overarching message from ProtecT is reassuring. Most men with early PSA screen-detected prostate cancer fare very well over many years, regardless of which option they choose, and very few die from their disease, even after 15 years. Modern AS is more precise. ProtecT mainly relied on PSA tests.Today, surveillance uses PSA and prostateexam every 6 months, MRI every 1–2 years, and repeating biopsies when surveillance findings change. Delaying treatment is usually safe. Most studies show that men who start with AS and later choose surgery typically do not lose control of their cancer. Life expectancy matters. We should evaluate men expected to live 10 to 15 years or more carefully before selecting AS. Modern tools now help us select those men more safely and confidently.
EEur Urol. 2020 Mar;77(3):320-330.
N Engl J Med. 2023 Apr 27;388(17):1547-1558
JAMA Oncol. 2025 Oct 16:e254125. doi:
10.1001/jamaoncol.2025.4125. Epub ahead of print.
PMID: 41100113; PMCID: PMC12532030.
