Prostate Cancer

A detectable, rising PSA is usually the earliest sign of recurrence after definitive treatment, and it typically precedes clinical events by years. Management hinges on distinguishing local failure (salvageable) from distant failure (treated as metastatic castrate-sensitive disease), using PSA kinetics, the interval to recurrence, and Gleason grade.

After Radical Prostatectomy

After a successful prostatectomy the PSA should fall to undetectable (PSA half-life is 2–3 days; the fraction remaining after n half-lives is 1/2ⁿ). Biochemical recurrence (AUA/ASTRO) is a PSA ≥0.2 ng/mL confirmed by a second value; a prior period of undetectable PSA is not required.

Epidemiology and Natural History

• Roughly 25–41% of men develop PSA recurrence within 10 years of prostatectomy; of these, ~50% recur within 3 years, ~80% within 5 years, and ~99% within 10 years. Recurrence beyond 15 years is rare, so PSA follow-up beyond 10 years has limited utility, especially in older men.
• Not every detectable PSA progresses — some plateau, explained by residual benign prostatic tissue, non-prostatic PSA (urethral and salivary glands), or indolent residual cancer.
• The MSKCC postoperative nomogram predicts BCR and cancer-specific survival from PSA, Gleason score, extraprostatic extension, seminal vesicle invasion, lymph node involvement, margin status, and adjuvant radiation; positive-margin length is also predictive.
• Pound et al. (1999) — in 1,997 men followed after prostatectomy without neoadjuvant or adjuvant therapy, 304 (≈15%) developed BCR; the median time from BCR to metastasis was 8 years (metastasis-free survival 10 years in a later update), and from metastasis to death 5 years — a median of more than 13 years from BCR to death. The survival impact of BCR is limited to a subgroup: a short PSA doubling time with a high final Gleason score (after prostatectomy), or a short interval to failure with a high biopsy Gleason score (after radiotherapy), carries the worst prognosis.

EAU risk-stratification:

Risk	After prostatectomy	After radiotherapy
Low	PSA doubling time >1 yr and prostatectomy Gleason <8	Interval to BCR >18 mo and biopsy Gleason <8
High	PSA doubling time ≤1 yr or prostatectomy Gleason 8–10 (ISUP 4–5)	Interval to BCR ≤18 mo or biopsy Gleason 8–10 (ISUP 4–5)

Restaging Imaging

The goal is to separate local from distant failure. MRI identifies local recurrence with high accuracy even at low PSA (sensitivity/specificity/accuracy ≈98%/94%/93%, up to 100%/97%/91% when confirmed by biopsy). For distant disease, conventional CT and bone scan have a yield that depends heavily on PSA — the median PSA at a newly detected bone metastasis is 32 ng/mL, and ~25% occur at PSA <10 ng/mL:

Modality	Detection by PSA
CT (lymph nodes)	<20 ng/mL: 0%; >20 ng/mL: 1.1% (Abuzallouf 2004)
Bone scan	<10 ng/mL: 2.3%; 10–20: 5.3%; 20–49.9: 16.2% (Abuzallouf 2004)
PSMA / novel PET-CT	Higher sensitivity at low PSA (<2.0 ng/mL); FDA-approved after BCR to assess locoregional salvage; endorsed by the 2021 CUA Best Practice Report

Management — Local Failure (Salvage Radiation)

Of all salvage options (including ADT), radiation gives the best long-term progression-free survival, improving biochemical-recurrence-free, progression-free, cancer-specific, and overall survival.

• Timing and dose — outcomes improve at lower PSA (some treat at ≤0.5 ng/mL); at least 64 Gy is delivered to the prostatic bed, with emerging evidence favouring higher doses (cf. 76–80 Gy EBRT and ~145 Gy iodine / 125 Gy palladium brachytherapy for localized disease).
• Whole-pelvis vs prostatic-bed — no mature RCT shows added benefit of whole-pelvis radiation; RTOG 0534 is testing prostatic-bed RT alone vs + 6 months ADT vs + pelvic-node RT + 6 months ADT.
• Concurrent ADT — the 2017 AUA guideline recommends offering hormone therapy with salvage RT (Grade A). GETUG-AFU 16 (Carrie 2019) showed adding short-course ADT improved progression-free survival by 15% at 10 years (65% vs 49%), and RTOG 96-01 (Shipley 2017) showed 2 years of bicalutamide improved overall, cancer-specific, and metastasis-free survival (benefit greatest with PSA >0.60).
• Prognosis — favourable salvage responses occur with late, slowly rising PSA, low-grade tumour, and no seminal vesicle or nodal involvement; a post-salvage PSA nadir >0.05 ng/mL predicts distant metastasis and reduced cancer-specific survival.

Management — ADT After Prostatectomy

ADT is non-curative for post-prostatectomy biochemical failure and has no proven survival benefit, yet ~60% of men receive it as second-line therapy. The 2020 AUA guideline advises against routine ADT for BCR after prostatectomy; two large observational studies found no mortality difference between immediate and deferred ADT. If used without metastases, intermittent ADT may be offered. High-dose bicalutamide (150 mg daily) can delay progression with survival equivalent to orchiectomy and less sexual dysfunction and osteoporosis, but possibly more cardiovascular risk (Wirth 2001).

After Radiation Therapy

Definitions of failure — the 1996 ASTRO definition (three consecutive PSA rises 6 months apart, backdated) has been largely replaced by the 2005 Phoenix definition (PSA nadir + 2 ng/mL, not backdated), which has fewer false positives. Because the two modalities are defined differently, metastasis-free or cancer-specific survival are better measures for comparing prostatectomy and radiotherapy. Biochemical and histologic failure are best confirmed at least 2 years out to account for the PSA bounce.

Natural History

PSA-only recurrence after definitive radiation still leads to local and distant failure and cancer death. Four features predict clinical progression: time from RT to BCR <3 years, PSA doubling time <3 months (the most consistent predictor), pretreatment biopsy grade group ≥4, and pretreatment stage ≥cT3b.

Diagnosis and Evaluation

Men are at risk of both local and distant failure, so differentiate local (prostate biopsy) from distant (imaging) disease.

• Imaging — post-radiation fibrosis makes imaging difficult; MRI is the most promising for local recurrence, while PET/CT (11C-choline, PSMA) is under study. On bone scan, the chance of a lesion at PSA <10 ng/mL is <1% if PSADT >6 months and 10% if <6 months. CT is reasonable when advanced imaging is unavailable, though PSMA-PET is more sensitive for nodal disease.
• Biopsy after radiotherapy — strongly recommended to document local recurrence before salvage, but reserved for men in whom salvage local therapy is being considered (radiation changes cause a high false-positive rate in the first year). Use mpMRI-directed biopsy: standard systematic biopsy if PI-RADS 1, or a 12-core systematic biopsy plus ≥1 targeted core per MRI lesion otherwise. Seminal vesicle invasion was historically reported in up to 42% of cases.

Management — Local Failure (No Metastases)

Aggressive local therapy outside a trial is not recommended; options are active surveillance, local salvage therapy (salvage prostatectomy, cryosurgery, brachytherapy, or HIFU), or ADT.

• Active surveillance — reasonable for lower-risk men (BCR >3 years from RT, PSADT ≥16 months, pre-RT grade group 1) or life expectancy <10 years.
• Local salvage therapy — may improve metastasis-free survival, delay ADT, and eradicate disease in selected men (T1c–T2, PSA <10 ng/mL at recurrence, no metastases), although the survival benefit is unproven and biopsy-proven recurrence should be confirmed first.

Salvage radical prostatectomy is a technically demanding operation for highly selected men, performed only by experienced surgeons. The EAU restricts it to men with low comorbidity, life expectancy >10 years, initial stage T1–T2, initial biopsy ISUP grade <2/3, pre-salvage PSA <10 ng/mL, and no nodal or distant metastases.

• Outcomes — positive-margin rate ≈20%; biochemical-recurrence-free survival 47–82% (5 yr) and 28–53% (10 yr); cancer-specific survival 89–100% (5 yr) and 70–83% (10 yr); overall survival 54–90% at 10 years. Pre-salvage PSA is the strongest prognostic factor.
• Complications — significant perioperative morbidity and worse function than primary surgery: bladder-neck contracture 0–55%, erectile function sufficient for intercourse in only 0–20%, and rectal injury 0–28% (managed by two-layer closure, omentoplasty, or colostomy). Contemporary radiation techniques have reduced rectal injury (0–28% before 2000 vs 2–10% after), though incontinence and erectile dysfunction remain problematic.
• Approach — a multi-institutional series of 395 salvage prostatectomies found the robotic approach reduced blood loss, hospital stay, anastomotic stricture (8% vs 17% open), and incontinence at 12 months (32% open vs 22% robotic — see Corrections note), with no difference in overall or major complications (Gontero 2019). Adding pelvic lymphadenectomy improved overall and cancer-specific survival in SEER data, with marginal benefit beyond the 7th node.

Other salvage options — cryotherapy is a less morbid, outpatient alternative with durable progression-free and overall survival (erectile dysfunction is very common; other risks include urinary obstruction, urethrorectal fistula, urethral sloughing/stricture, and rectal pain). Salvage brachytherapy and HIFU are described but rest on weaker evidence.

Management — ADT and Systemic Therapy

ADT is the most common treatment for BCR after radiation but is generally not curative; the exact timing is unknown, and it is most reasonable to start in men at highest risk of distant failure (PSADT <12 months). Intermittent ADT is noninferior to continuous ADT in men with PSA >3 ng/mL without metastases, with better physical function, fatigue, libido, and erectile function (Crook 2012).

The EMBARK trial (Freedland 2023; 1,068 men with high-risk BCR — PSA doubling time ≤9 months and PSA ≥2 ng/mL above nadir after RT or ≥1 ng/mL after prostatectomy) randomized enzalutamide + leuprolide vs enzalutamide alone vs leuprolide alone: combination therapy improved 5-year metastasis-free survival to 87% vs 71% for leuprolide alone (absolute reduction 16%, NNT 6), with a non-significant 5-year OS difference (92% vs 87%). It was the first evidence that second-generation androgen-receptor monotherapy (enzalutamide) outperforms first-generation monotherapy for progression-free survival.

After Cryotherapy or HIFU

Definitions of failure after primary cryotherapy vary widely, and management strategies (repeat cryotherapy, salvage radiation, salvage prostatectomy) are not well established. After primary HIFU, PSA nadir + 1.2 ng/mL best predicts clinical failure; salvage radiotherapy and prostatectomy have been described.

Self-Test

1. What is the definition of biochemical failure after radical prostatectomy? PSA ≥0.2 ng/mL with a second confirmatory value.

2. What did Pound et al. show about the natural history of BCR after prostatectomy? In 1,997 men, ≈15% developed BCR; the median time from BCR to metastasis was 8 years, and from metastasis to death 5 years.

3. What is the median PSA at a newly detected bone metastasis? 32 ng/mL.

4. What percentage of men with BCR after prostatectomy and PSA <10 ng/mL have a positive bone scan? About 4%.

5. What is the best investigation for local failure after prostatectomy or radiotherapy? MRI.

6. What is the minimum recommended dose of salvage radiation? At least 64 Gy.