A patient with prostate cancer in PSA relapse is one who, having received a primary treatment with intent to cure, has a raised PSA (prostate-specific antigen) level defined as "biochemical progression".
Prostate-specific antigen (PSA) is a protein produced by prostate epithelial cells, whether benign or malignant. Measuring the PSA level is a key aspect in monitoring after treatment with intent to cure, as very low levels of PSA are indicative of a successful removal of the tumour4,17.
It is known that, if the PSA level increases after a radical treatment, clinical recurrence of the tumour will be seen within a few years174,185,192-195.
The challenge is to find out how this increase in PSA levels after radical treatment involves a significantly higher risk of morbidity or mortality, which is called progression, relapse or PSA relapse. There has been much debate about the PSA limit indicative of the greatest risk4,185,196,197.
In the study by Stephenson et al185, different definitions of PSA relapse after radical prostatectomy were evaluated. It found that the best indication of metastatic progression was an increased PSA value ≥ 0.4 ng/ml, which gives a probability of PSA progression in the following 4 years of 91 %, and secondary treatment failure or clinical failure in the following 7 years of 62%. It also concludes that if the serum PSA cut-off level is increased to above 0.4 ng/ml, the probability that the patient will still be disease-free within the following 10 years is 74% [95% CI: 70-78%], which is equivalent to an increase of false negatives for biochemical progression.
Some groups choose to adhere to values of 0.2 ng/ml or higher due to the greater sensitivity of the serum PSA measurement method. Choosing a lower cut-off level results in a higher rate of secondary interventions for patients with a high probability of remaining disease-free for 10 years (false positives).
There are several studies that examine the best definition of biochemical progression after radiation therapy (external or brachytherapy), such as the Horwitz et al retrospective cohort study194, two prospective case series published by Kuban et al197,190 and the consensus document from the American Society for Therapeutic Radiology and Oncology (ASTRO), published by Roach et al196. They conclude that for external beam radiotherapy, the 2005 ASTRO definition agreed by consensus (PSA 2 ng/ml above the nadir value) has the best values for sensitivity (72-74%) and specificity (71-83 %) for clinical failure and at a distance. For brachytherapy, it also found that the definition offered the best sensitivity and specificity. The false positive rate is 2% for external beam radiation therapy (with or without neoadjuvant and/or adjuvant hormonal treatment) and for brachytherapy with hormone therapy. For brachytherapy without hormone treatment, the false positive rate reaches 4%.
| 3 | In patients who received radical prostatectomy, obtaining an increasing PSA value ≥ 0.4 ng/ml, is the definition of PSA relapse that best correlates with metastatic progression, with a probability of PSA progression in the following 4 years of 91%, and a 62% probability of secondary treatment or clinical failure in the following 7 years185. |
| 3 | If, after defining PSA relapse in patients who received radical prostatectomy, the serum PSA cut-off level is above 0.4 ng/ml, the probability that the patient will still be disease-free after 10 years is 74% [95% CI: 70-78%], which leads to an increase of false negatives. If lower cut-off points are used, there is a higher rate of false positives185. |
| 2-/4 / 3/3 |
In patients who received radical radiotherapy, the ASTRO 2005 definition of PSA relapse (PSA greater than 2 ng/ml above the nadir value) has the best values for sensitivity (72-74%) and specificity (71-83%) for clinical failure and at a distance194,196-198. |
| 4/3 | In patients who received radical brachytherapy, the ASTRO 2005 definition of PSA relapse offers the best sensitivity and specificity196,197. |
| 2-/4 / 3/3 |
In patients who received radical radiotherapy, the rate of false positives using the ASTRO 2005 definition of PSA relapse is 2% for external beam radiation therapy (with or without neoadjuvant and/or adjuvant hormone treatment) and for brachytherapy with hormone therapy. For brachytherapy without hormone treatment, the false positive rate reaches 4%194,196-198. |
| D | In prostatectomised patients, biochemical recurrence of the disease will be considered to have occurred when serum PSA levels exceed 0.4 ng/ml. |
| D | En aquellos pacientes cuya intervención con intención curativa ha sido la radioterapia o braquiterapia, se considerará recurrencia bioquímica de la enfermedad cuando los niveles séricos de PSA se incrementen en 2 ng/ml sobre el PSA nadir. |
Radical prostatectomy is a frequently used treatment for localised prostate cancer. Local recurrence of the disease occurs in more than 33% of patients within 5 years of surgery. The existence of PSA relapse implies a 34% probability of metastatic disease within 5 years of radical prostatectomy. Following metastasis, median survival is 5 years199.
Salvage treatment is offered to patients who display PSA relapse with the intention of reducing adverse outcomes caused by advanced prostate cancer (locally advanced affectation or disseminated). Proper management depends on the treatment with intent to cure and the status of the patient200.
No studies directly comparing salvage radiotherapy with immediate hormone treatment have been found.,
Two studies examined overall survival after applying salvage radiotherapy in prostatectomised patients with PSA relapse. It found that survival at 5 years was between 87% and 95%190,199. When radiation is given as soon as the disease becomes palpable, survival is 76% (p = 0.02)190.
In terms of progression-free survival for these patients, different publications126,200-202 suggest that statistically significant improvement is found when salvage radiotherapy is applied at PSA relapse, defined as PSA levels between 0.6-2.5 ng/ml.
Analysing the Stephenson et al185 and Pazona et al200 case series, several factors were found which increase the probability for these patients not to respond to salvage radiotherapy, such as a pre-treatment PSA doubling time of less than 10 months, the existence of lymph node or seminal affectation or a Gleason > 7. In general, in patients with PSA relapse, to have a PSA doubling time less than 3 months was an adverse prognostic factor for cancer-specific and overall survival in the Freedland et al203 and D'Amico et al204 publications.
The study published by Moul et al205 analysed early salvage hormone therapy after radical prostatectomy (beginning when PSA values ≤ 5 ng/ml), compared with late salvage hormone therapy (when there are clinical signs and symptoms of progression of the disease). They note that early hormone therapy provides statistically significant improvement only for metastasis-free survival in a subgroup of patients at high risk: those with a pathological Gleason > 7 or PSA doubling of less than 1 year (HR = 2.32 [95% CI: 1.14-4.70]).
| 3 | Overall survival at 5 years in prostatectomised patients with PSA relapse and salvage radiotherapy is between 87% and 95%190,199. When radiotherapy is given as soon as the disease is palpable, survival is 76% (p = 0.02)190. |
| 3 | It seems that statistically significant improvement in progression-free survival is seen if salvage RT is applied at PSA relapse after prostatectomy with PSA levels between 0.6-2.5 ng/ml126,200-202. |
| 3 | Risk factors that do not respond to salvage RT after surgery are: a pre-treatment PSA doubling time of less than 10 months, the existence of lymphatic or seminal affectation or a Gleason > 7185,200. |
| 3 | Having a PSA doubling time of less than 3 months was an adverse prognostic factor for cancer-specific and overall survival in groups of men with PSA relapse after prostatectomy203,204. |
| 2- | Early salvage hormone therapy after radical prostatectomy (beginning when PSA values ≤ 5 ng/ml), compared with late salvage hormone therapy (when there are clinical signs and symptoms of progression of the disease) shows statistically significant improvement only for metastasis-free survival in a subgroup of patients at high risk: those with a pathological Gleason > 7 or PSA doubling of less than 1 year (HR = 2.32 [95% CI: 1.14-4.70])205. |
| D | Patients with PSA relapse of the disease after radical prostatectomy without distant metastases or other risk factors, should be given early salvage radiotherapy before the PSA exceeds 2.5 ng/ml. |
| D | Salvage hormone therapy may be indicated for those men with PSA relapse after radical prostatectomy who also have local symptomatic progression, distant metastasis or duplication of PSA levels in less than 10 months. |
The use of radiotherapy as a definitive treatment for new cases diagnosed with prostate cancer has increased significantly over the past 30 years. An estimated 76% of patients with a good prognosis (cT1-cT2a and Gleason < 6) and 51% with an unfavourable prognosis (cT2b-cT3 or Gleason > 7) remain free of PSA relapse at 5 years of curative treatment. Biopsy studies have revealed the persistence of neoplastic cells in 20-50% of patients after treatment with radiation therapy, which suggests that when adequate local control of the disease is not achieved, there is a deterioration in results with an increase in late distant metastasis206.
There are no randomised studies showing direct comparisons between different salvage alternatives in patients with PSA relapse after treatment with intent to cure. Furthermore, the retrospective comparison of existing data reveal methodological difficulties due to the different definitions of PSA relapse used in the different studies207.
Faria et al208 has results from 178 men with asymptomatic PSA relapse after external beam radiotherapy. Some received salvage treatment with hormone therapy and others chose to wait and see (watchful waiting). There were no deaths among these patients due to prostate cancer. With a median follow-up of 7 years, overall survival was 95% in the hormone therapy group and 89% in the watchful waiting group.
In a study published by Pinova et al209, with 248 male patients who had salvage treatment (hormone therapy vs watchful waiting), the metastasis-free survival rate at 5 years was 88% vs 92% (p = 0.74) in those with a PSA doubling time of ≥ 12 months. In those who had a PSA doubling time of <12 months, results were 78% vs 57% (p = 0.0026).
Diferentes series206,210,211 gave salvage treatment results with prostatectomy. Cancer-specific survival was 73% at 10 years, and 60% at 15 years. When cystoprostatectomy instead of retropubic prostatectomy was practised, cancer -specific survival at 10 years was much lower (38% vs 77%; p <0.001). The proportion of possible complications after salvage radical prostatectomy was: urinary incontinence (48%), urinary extravasation (15%), contraction of the bladder neck (22%), rectal injury (4%) and kidney damage (2%).
The European Association of Urology4 clinical practice guideline suggests that salvage prostatectomy be considered in patients with few comorbidities, a life expectancy of at least 10 years, cT1-T2, Gleason < 7 and preoperative PSA < 10 ng/ml.
In one study (n = 49), overall survival at 5 years after salvage brachytherapy therapy was 56% [95% CI: 36-71%] and the cancer-specific survival rate was 79% [95% CI: 58-91%]. Median follow-up: 2 years (range: 3 months to 6.5 years)212.
In another series (n = 116), cancer-specific mortality at 5 years was 8.3% for salvage cryotherapy and 5.4% for radical prostatectomy, with no statistically significant difference between them. Biochemical progression appeared in 66.7% of those treated with cryotherapy and 28.6% of those treated with surgery213.
In another group of patients treated with salvage cryotherapy, with an average follow-up of 13.5 months, there was biochemical progression in 58% of patients. 31% of cases had undetectable levels of PSA214.
Side effects found between 12-13.5 months after the salvage cryotherapy were common: urinary incontinence (28-73%), obstructive symptoms (67%), impotence (72-90%) and severe perineal pain (8%)214,215.
On the use of high intensity focused ultrasound (HIFU) as a salvage treatment, there is a very small case series with a short follow-up period for which no conclusions about efficacy could be drawn216.
| 2- | In a group of men with asymptomatic PSA relapse after external radiotherapy and salvage treatment (hormone therapy vs watchful waiting), no death was due to prostate cancer. With a median follow-up of 7 years, overall survival was 95% in the hormone therapy group and 89% in the watchful waiting group208. |
| 2- | In a group of men with salvage treatment (hormone therapy vs watchful waiting) after radiotherapy, the metastasis-free survival rate at 5 years was 88% vs 92% (p = 0.74) in those with a PSA doubling time of ≥ 12 months. In those who had a PSA doubling time of <12 months, results were 78% vs 57% (p = 0.0026)209. |
| 3 | In those treated with salvage prostatectomy after radiotherapy, cancer-specific survival was 73% at 10 years, and 60% at 15 years210. When cystoprostatectomy instead of retropubic prostatectomy was practised, cancer -specific survival at 10 years was much lower (38% vs 77%; p <0.001)206. The proportion of possible complications after salvage radical prostatectomy was: urinary incontinence (48%), urinary extravasation (15%), contraction of the bladder neck (22%), rectal injury (4%) and kidney damage (2%)206,211. |
| 4 | It is recommended that salvage prostatectomy be considered in patients with few comorbidities, a life expectancy of at least 10 years, cT1-T2, Gleason < 7 and preoperative PSA < 10 ng/ml4. |
| 3 | In a small series (n = 49), overall survival at 5 years after salvage brachytherapy therapy was 56% [95% CI: 36-71%] and the cancer-specific survival rate was 79% [95% CI: 58-91%]. Median follow-up: 2 years (range: 3 months to 6.5 years)212. |
| 3 | In patients with salvage treatment after radiotherapy (prostatectomy vs cryotherapy) and average follow-up: 4.6 years vs 5.1 years, the cancer-specific mortality was 5.4% vs 8.3%, with no statistically significant differences. Patients with PSA relapse: 28.6% vs 66.7%213. |
| 3 | In those treated with salvage cryotherapy after radiotherapy, with an average follow-up of 13.5 months, there was PSA relapse in 58% of patients. 31% had undetectable levels of PSA214. |
| 3 | In those treated with salvage cryotherapy after radiotherapy, adverse effects between 12-13.5 months were common: urinary incontinence (28-73%), obstructive symptoms (67%), impotence (72-90%) and severe perineal pain (8%)214,215. |
| 3 | On the use of high intensity focused ultrasound (HIFU) as a salvage treatment after radiotherapy, there is a very small case series with a short follow-up period for which no conclusions about efficacy could be drawn216. |
| D | Salvage radical prostatectomy can be offered after radiotherapy treatment for patients with local recurrence with few associated comorbidities, a life expectancy of at least 10 years, with cT1-T2, Gleason < 7 and a pre-surgical PSA < 10 ng/ml. |
| D | Hormone therapy should be considered as a salvage therapeutic option in patients treated with radiotherapy with local recurrence of the disease, who cannot be offered salvage radical prostatectomy. |
| D | The adoption of other salvage therapeutic alternatives (cryotherapy or high intensity focused ultrasound) should be considered as experimental. |
| D | RESEARCH RECOMMENDATION: Clinical trials evaluating local salvage therapies for survival and quality of life in men with biochemical recurrence after radiotherapy or brachytherapy should be started. |
The objectives of disseminated prostate cancer treatment include prolonging survival, preventing the symptoms of progression of the disease, improving the quality of life and reducing morbidity due to the treatment itself16,217.
Androgen suppression hormone therapy is one possible alternative treatment. It can be started early (when the patient is diagnosed with asymptomatic PSA relapse), or on a deferred basis, (when the signs and symptoms of disease progression appear)14,217.
There is a Cochrane review of Nair et al217 which compares immediate hormonal therapy with delayed in men with advanced prostate cancer (locally advanced or disseminated affectation). No differences were found for cancer-specific survival when comparing the immediate and delayed treatment.
The study of 1,352 patients published by Moul et al205 analysed early salvage hormone therapy in patients with PSA relapse after radical prostatectomy (started when PSA values ≤ 5 ng/ml were reached), and compared it with late salvage hormone therapy (when there were signs and symptoms of progression). They noted that for metastasis-free survival at 5 and 10 years there were no statistically significant differences [HR = 0.91 (95% CI: 0.58-1.41)]. Statistically significant improvements were seen only a in a subgroup of patients at high risk: those with Gleason > 7 or a PSA doubling in <1 year (HR = 2.32 [95% CI: 1.14-4.70]).
| 1+ | In men with advanced prostate cancer (locally advanced or disseminated affectation), no differences were found for cancer-specific survival when comparing immediate and deferred treatment217. |
| 2- | In patients with PSA relapse after radical prostatectomy, no statistically significant differences were found when comparing metastasis-free survival at 5 and 10 years for those who received early salvage hormone therapy (started when PSA levels reached levels ≤ 5 ng/ml) vs late (when there were signs and symptoms of progression). The only statistically significant improvement was seen in a subgroup of patients: Gleason> 7 or PSA doubling <1 year (HR = 2.32 [95% CI: 1.14-4.70])205. |
| D | In patients with PSA relapse after radical prostatectomy in which hormonal treatment is decided, if there is a Gleason > 7, PSA ≤ 5 ng/ml and a PSA doubling time of less than 1 year, it is recommended that hormone treatment be applied early. |
| √ | In patients with PSA relapse after radical prostatectomy in which hormonal treatment is decided, if there is a Gleason > 7, PSA ≤ 5 ng/ml and a PSA doubling time of less than 1 year, it is recommended that hormone treatment be applied early. |
The use of intermittent androgen suppression hormone therapy has been justified for various reasons4:
The systematic Cochrane review published by Conti et al218 is a meta-analysis comparing intermittent hormone therapy (HT) (after prostatectomy, radiotherapy or brachytherapy) vs continuous HT in prostate cancer patients. Most studies included patients who had not received radical treatment, except for the Leval et al 219, consisting of men with prostate cancer in PSA relapse after radical prostatectomy. The median progression rate at 3 years was significantly less for intermittent HT (7% vs 38.9%). Cancer-specific mortality was 5.7% for those treated with intermittent HT, and 12.1% for those treated with continuous HT, after a median follow-up of 2.4 years. After analysing the risk of PSA relapse (defined as PSA ≥ 10 ng/ml), no significant differences were found when comparing intermittent and continuous HT. In patients with Gleason> 6, there was a trend towards lower risk of biochemical progression in the intermittent HT group (RR = 0.47 [95% CI: 0.04-4.96]; p = 0.53 ). Most of the patients who received HT experienced mild-moderate side effects due to androgen suppression (hot flashes, loss of libido and erectile dysfunction). These almost always disappeared during the period without hormone treatment in the intermittent HT group. In addition, the emergence of severe gastrointestinal toxicity stopped treatment in 4.4% of the patients who underwent continuous HT treatment and 2.9% for those receiving intermittent HT treatment219.
With regard to sexual function, in a trial published by Calais et al220, 50% of the patients included had been sexually active in the previous month. At 15 months of treatment, 40% of the intermittent HT group and 25% of the continuous HT group maintained sexual activity. Similar results were obtained from both groups in this study using the quality of life scales.
| 1- | For patients in PSA relapse subjected to salvage HT after radical prostatectomy, the median progression rate at 3 years was significantly smaller for intermittent HT compared with continuous HT (7% v 38.9%)219. |
| 1- | For patients with PSA relapse after radical prostatectomy, 5.7% of those treated with intermittent salvage HT died because of the tumour compared with 12.1% for those treated with continuous HT219. |
| 1- | For patients with PSA relapse subjected to salvage HT after radical prostatectomy, there were no significant differences between intermittent and continuous when analysing the risk of the appearance of biochemical progression (defined as PSA ≥ 10 ng/ml). For Gleason > 6, there was a trend towards lower risk of biochemical progression in the intermittent HT group (RR = 0.47 [95% CI: 0.04-4.96]; p = 0.53)219. |
| 1- | Most patients who received HT experienced mild-moderate adverse effects due to androgen suppression (hot flushes, loss of libido and erectile dysfunction), which disappeared during the period without hormone treatment in the intermittent HT group219. |
| 1- | For patients in PSA relapse subjected to salvage HT after radical prostatectomy, severe gastrointestinal toxicity stopped treatment in 4.4% of patients treated with continuous HT and 2.9% of those receiving intermittent HT219. |
| 1+ | In patients with PSA relapse after radical prostatectomy, 50% of men were sexually active in the month prior to the start of the salvage treatment. At 15 months of treatment, sexual activity was maintained in 40% of the intermittent HT group and 25% of the continuous HT group220. |
| 1+ | The quality of life scales had similar results for intermittent and continuous HT in patients with biochemical progression after salvage radical prostatectomy220. |
| A | In patients with PSA relapse after radical treatment in which hormone therapy has been decided, it cannot be determined whether it is better to apply it continuously or intermittently. |
Latest update: May 2009

