Questions to Ask About Prostate Cancer Treatment

A great many low-risk cancers found by PSA testing can be watched safely rather than treated straight away. Intermediate and high-risk prostate cancer is a different matter: it can spread and it can kill, and it is usually treated.

NCI shows why the low-risk end exists at all. Autopsy series of men who died of unrelated causes found indolent prostate tumors in 30% to 70% of men over 60, and silent prostate cancer may be present in 50% to 60% of men aged 90 and older. Population survival is correspondingly high: among US men diagnosed between 2016 and 2022, 5-year relative survival was 100.0% for local or regional disease and 98.2% across all stages, though it was 40.1% once the cancer had spread to distant sites. The median age at diagnosis is 68.

Those are population figures, and they cannot tell you what your own untreated cancer would do. That takes your grade group, clinical stage, PSA, how much tumor the biopsy found, imaging, sometimes a genomic test, and your age and general health together.

So the first question is not which treatment. It is which risk group this cancer falls into, and what that means for treating now versus monitoring.

Three numbers do most of the work: the Gleason score, the PSA, and how many cores were involved.

Ask for the Gleason score written as two numbers, not just a total. The score adds the two most common tissue patterns. Each is rated 1 through 5. A 3+4=7 and a 4+3=7 are not the same cancer. NCI explains why. The least differentiated part of the sample may carry its own prognostic weight. So the two parts get reported on their own.

Two cautions come from NCI itself. Over 95% of prostate cancers are adenocarcinomas. So an unusual type on the report is worth asking about. Second, pathologists have drifted upward over time. They now give higher Gleason scores to the same tissue patterns. NCI calls this grade inflation. It makes older survival figures hard to compare with today's.

Stage I is defined narrowly. It requires cT1a-c, cT2a, or pT2, with N0 and M0. It also requires a PSA under 10 ng/mL and a Gleason score of 6 or lower.

NCI's language here is worth reading closely. Most men with screen-detected prostate cancer may be candidates for active surveillance. Definitive treatment is held back until there are signs of progression.

The outcomes support that. Studies have used various criteria. Across them, men on watchful waiting or active surveillance had prostate-cancer-specific mortality of about 1% to 10%. The best rates came from the more recent series.

What surveillance involves, per NCI:

• Regular visits.
• Digital rectal examinations.
• PSA testing.
• Transrectal ultrasound, in some programs.
• Transrectal needle biopsies, in some programs.

Now the honest part, and it is worth raising directly. NCI says four things here are arbitrary and not set by controlled trials. Those are patient selection, testing intervals, the specific tests used, and the criteria for stepping in. So it is fair to ask which protocol a clinic follows, and where it came from.

The ProtecT trial is the strongest direct evidence. It randomized men to three arms: active monitoring, radical prostatectomy, or radiation therapy. It is the trial to ask about by name.

For stage I disease, four options gave similar survival in selected series. Those were radical prostatectomy, external-beam radiation, radioactive seed implants, and active surveillance. Those series were not randomized. NCI says the decision should rest on age, other illnesses, and personal preference.

Robot-assisted prostatectomy is now the most common technique in developed countries. One randomized trial enrolled 308 men. Urinary, sexual, and bowel function were similar with open retropubic and robotic surgery. Median follow-up was 24 months. NCI notes the trial was too small and too short to compare cancer outcomes.

Questions worth asking:

• Will a pelvic lymph node dissection be done? For small, well-differentiated nodules, positive pelvic nodes turn up in under 20%. The dissection may then be skipped. For larger, less differentiated tumors it matters more.
• What happens if the frozen section shows node involvement? NCI says radical prostatectomy is usually not performed in that case.
• Were the margins clear on the final pathology? Recurrence rises when tumor margins are positive.

One thing NCI says is not established: preoperative hormonal therapy before surgery.

This section is where a lot of anxiety lives, and NCI is unusually candid about the limits.

NCI says the best follow-up strategy after treatment is uncertain. Guidelines differ widely, and that reflects a lack of research evidence. It also warns about surrogate endpoints. Using them to drive decisions is controversial. The evidence that changing therapy on that basis helps is weak.

After radical prostatectomy, a detectable PSA marks higher risk of local failure or spread. But NCI adds a caveat. A large share of men with a rising PSA stay free of symptoms for long stretches. Biochemical failure alone may not be enough to start more treatment.

Here is the scale. One analysis followed nearly 2,000 men after prostatectomy done with intent to cure. Mean follow-up was 5.3 years. In that group, 315 men, or 15%, reached a PSA of 0.2 ng/mL or higher.

One measure has held up better than a single value. A cohort of 8,669 men had surgery or radiation. A PSA doubling time under 3 months met some criteria as a surrogate. That held for both all-cause and prostate-cancer mortality. So the trend matters more than any one number.

If a nomogram is used to predict outcomes, ask what goes into it. Before surgery, they use clinical stage, PSA, Gleason score, and the count of positive and negative biopsy cores. After surgery, they add capsular invasion, surgical margins, seminal vesicle invasion, and node involvement.

NCI raises two limits worth knowing. The nomograms were built at academic centers. They may be less accurate at the community hospitals where most men are treated. They also predict middle outcomes, such as a PSA rise. Some even predict a doctor's sense that more therapy is needed. Neither is survival.

If the diagnosis came from a screening PSA, this context is fair to raise. NCI says randomized trials of prostate cancer screening have given conflicting results. Systematic reviews and meta-analyses found no clear evidence that screening lowers the risk of death from prostate cancer. They also found no clear evidence that the benefits outweigh the harms.

Our page on prostate cancer covers the basics, and our overview of cancer treatment explains how these modalities work in general.

• National Cancer Institute, Prostate Cancer Treatment (PDQ) - Health Professional Version, updated May 14, 2025, accessed August 6, 2026
• National Cancer Institute, Prostate Cancer Treatment (PDQ) - Patient Version, accessed August 6, 2026