The watchful waiting attitude is the conscious decision not to provide any kind of treatment until the progression of the disease or presence of symptoms is apparent. In the latter situation, hormonal or palliative treatment could be started, but any radical treatment option is excluded. This attitude is often adopted with men of an advanced age or with significant comorbidities, with a low probability that the cancer will progress in any meaningful way during their expected lifetime¹⁷.

The randomised clinical trial of Bill-Axelson et al¹²¹compared the efficacy of radical prostatectomy with watchful waiting in patients with localised prostate cancer. The study showed the results with an analysis with intent to treat. The results (accumulated over 10 years) for both groups (radical prostatectomy v watchful waiting) are 19.2% [95% CI = 15.0-24.6] v 44.3% [95% CI: 38.8-50.5] for local progression (RR = 0.33; [CI 95%: 0.25-0.44]); 15.2% [95% CI 11.4-20.3] v 25.4% [95% CI 20.4-31.5] for distance metastasis (RR = 0.60; [CI 95%: 0.42-0.86]), 9.6 % [95% CI 6.5-14.2] v 14.9% [95% CI 11.2-19.8] for cancer-specific mortality (RR = 0.55; [CI 95%: 0.36-0.88]) and 27% [95% CI: 21.9-33.1] v 32% [95% CI: 26.9-38.2] for overall mortality (RR = 0.74: [95% CI: 0.56-0.99]). In other words, surgery is a statistically significant more effective treatment than watchful waiting.

The clinical trial of Steineck et al¹²² compared the quality of life for radical prostatectomy v watchful waiting in patients with localised prostate cancer. The results for both groups (radical prostatectomy v watchful waiting) are 80% v 45% for erectile dysfunction (RR = 1.78 [95%: 1.49-2.12], number needed to treat, NNT = 3 for watchful waiting), 29.1% v 39.6% for difficulties in urination (RR = 0.74 [95%: 0.55-0.98], NNT = 10 for surgery), 15.9% v 1.6% for the losses of urine (RR = 9.89 [95% CI 3.07-31.86], NNT = 7 for watchful waiting), 23.3% v 15% for moderate or severe urinary pain (RR = 1.55 [CI 95%: 1.01-2.39], NNT = 12 for watchful waiting), and 33.9% v 36.4% for perceived quality of life (RR = 0.93 [95% CI: 0.71-1.23]). It is believed that the only clinically significant differences for quality of life between the two treatment s are those relating to the sexual sphere, where there are better results for watchful waiting.

Klotz et al¹²⁰evaluated a series of 299 patients with clinically localised prostate cancer and proposed active surveillance for those meeting the following criteria:

1. Age < 70 years: Gleason < 7 and PSA ≤ 10 ng/ml (definition similar to low risk).
2. Age > 70 years: Gleason ≤ 7 (3 +4) and PSA <15 ng/ml.

These patients received treatment with intent to cure when the PSA doubling time was less than 2-3 years, when a Gleason ≥ 7 appeared in a prostate biopsy or when the patient requested it.

After a follow-up of 5.3 years, 15% of patients experienced early biochemical progression; 3%, clinical progression; 4%, histological progression, and 12% requested radical treatment. After 8 years, overall survival was 85% and cancer-specific survival 99.2% (100% of the deaths from prostate cancer had a PSA doubling time of < 2 years).

The systematic revision of Martin et al¹²³compared active surveillance protocols for patients with localised prostate cancer, including 5 series of cases. They agreed only in the PSA determination and digital rectal examination in active surveillance, with initial checks after each quarter, then every 6 months.

The clinical practice guideline on prostate cancer from the United Kingdom’s National Institute for Health and Clinical Excellence (NICE)^16,17 recommended special active surveillance in patients with clinical stage cT1, Gleason 3 + 3, PSA < 0.15 ng/ml and less than 50% of biopsy cylinders affected. It also proposed offering active surveillance to other low risk patients and considered it as an alternative for patients at intermediate risk.

The initial draft of this guideline recommended following up patients who opt for active surveillance with the following measurements^124,125:

1. Repeated yearly biopsies, after 4 years and 7 years, with at least 10 cylinders in each biopsy.
2. PSA determinations every 3 months during the first 2 years, and every 6 months thereafter.
3. Estimation of the PSA speed with linear regression, using at least 5 PSA determinations extending over at least a year.

It also suggested radical treatment in patients with any of the following data: PSA velocity > 1 ng/ml/year, higher degree or greater extension of the tumour in repeated biopsies, or evidence of locally advanced disease during a rectal examination^124,125.

In the systematic review of Nilsson et al¹²⁶ on the effects of radiotherapy for prostate cancer, the effects of radiotherapy alone are compared with radiotherapy associated with an intervention. It concludes that there are a large series of patients with efficacy results for external beam radiotherapy (ERT) and brachytherapy (BT) which are similar to those for radical prostatectomy (RP) for patients with localised prostate cancer at low risk (cT1-cT2a and Gleason < 7 and PSA ≤ 10 ng/ml).

In the systematic review of the Medical Services Advisory Committee (MSAC) in the Australian Ministry of Health¹²⁷, which includes systematic revisions, retrospective cohort studies and a series of cases, brachytherapy was evaluated with permanent I-125 implants in patients with localised prostate cancer at low risk. The review concluded that the available evidence did not demonstrate any differences in survival or disease progression in these patients compared with ERT v RP v BT.

In another systematic review of the Norway Health Technologies Evaluation Centre (SINTEF)¹²⁸, which analysed brachytherapy in patients with localised prostate cancer, a series of cases of men with low or intermediate risk [cT2b or Gleason = 7 or (PSA> 10 and ≤ 20 ng/ml)] when treated with BT or RP was studied. No differences were found in progression-free biochemical survival (PFBS) after 5 years, although the groups were not entirely comparable in terms of age and clinical stage. They also looked at 3 other studies (one cohort study of 2,222 patients, a case-control study and a series of cases) comparing BT with ERT, and in those where there were no differences found in PFBS at 5 and 7 years, although the groups were not entirely comparable in the case studies and controls, and the follow-up time was very short for the series of cases. When comparing BT + ERT v ERT, a case-control study found a greater PFBS after 5 years for the combined treatment (67% v 44%), although in this study the follow-up was incomplete and the average age of the control group was 5 years older. The authors concluded that BT compared with ERT or RP seems to provide comparable results, although the evidence is scant.

In the systematic revision of Nilsson et al¹²⁶ the use of high dose rate brachytherapy (HDR) in patients with prostate cancer was also studied. This consists of the application of brachytherapy at a high dose rate with Ir-192 in combination with ERT to provide a boost in the prostate. It must be done through transperineal ultrasound guided biopsy (TRUS). The review concluded that the total minimum dose obtained with this technique is far superior to those achieved with 3D-CRT, with an acceptable toxicity, and it induces local healing in most patients, even those at high risk.

Safety

The systematic revision of the MSAC¹²⁷ also compares the toxicity of brachytherapy vs external beam radiation vs radical prostatectomy. It found that, in the short term, brachytherapy is equal to or less toxic than ERT and RP in the area of sexual function (p = 0015); and that, for urinary incontinence, BT is better than RP ( p < 0.0001); for urethral obstruction, BT is worse than ERT (p < 0.0001); and for rectal toxicity, BT and ERT have similar results, both being worse than RP (p = 0.03) . In other words, the toxicity profiles for RP, ERT and BT are different. The authors of this revision concluded that, although it needs more evidence on the safety and efficacy of BT as a treatment for prostate cancer, its use can be recommended for patients with localised prostate cancer at low-risk, with a glandular volume less than 40 cm3 and availability of treatment (it is not possible to implement it in all Spanish public establishments).

The study by Potosky et al¹²⁹ is a retrospective cohort study comparing the adverse effects of RP vs ERT, with 5 years of follow-up. After 2 years, the (adjusted) percentage of patients with impotence is significantly higher in patients who underwent RP (82.1%) than in those treated with ERT (50.3%). Between 2 and 5 years, sexual function in patients who underwent ERT gets worse, although at 5 years there are still significant differences between the two treatments (erectile dysfunction 79.3% for RP v 63.5% for ERT; odds ratio, OR = 2.5 [CI 95%: 1.6-3.8]). There are significant differences in urinary incontinence (14-16% for RP v 4% for ERT; OR = 4.4 [95% CI: 2.2-8.6]), rectal tenesmus (35% for ERT v 20% for RP; OR = 0.56 [95% CI: 0.36-0.87]) and painful haemorrhoids (16% for ERT v 11% for RP; OR = 0.43 [95% CI: 0.25-0.74]).

In the SINTEF systematic review¹²⁸, which analyses brachytherapy in patients with localised prostate cancer, a case-control study comparing BT vs ERT was investigated. Higher rates of urinary obstruction were found in patients treated with BT, but no differences with regard to sexual function or proctitis were found.

A series of cases comparing BT with BT + ERT was also analysed. It found more patients with rectal complications in patients treated with only BT (grade 1: 10.5% v 8.9%; grade 2: 7.1% v 6.5 %; Grade 3: 0.7% v 0,4%).

The study by Robinson et al¹³⁰ is a systematic review comparing rates of erectile dysfunction after RP with preservation of neurovascular bundles (PNB) with other treatments. The results are derived from non-randomised studies of low sample size which may be biased, because they allowed neoadjuvant hormonal therapy (which can block testosterone for up to one year after finishing treatment). It was found that the probability of maintaining erectile function one year after treatment, adjusting for age, was as follows: for BT, 0.80 [95% CI: 0.64-0.96]; for BT + ERT , 0.69 [95% CI: 0.51-0.86]; for ERT, 0.68 [95% CI: 0.41-0.95]; for RP + PNB it was 0.22 [95% CI: 0-0,53]; and for RP without PNB, 0.16 [95% CI: 0.0-0.37].

In the systematic review of Morris et al¹³¹, which includes randomised and non-randomised trials, conformal radiotherapy is compared with conventional for the treatment of localised prostate cancer. In terms of efficacy, the conclusion was that, at similar doses, there were no statistically significant differences for local control of the disease, disease-free survival, biochemical progression-free survival or overall survival. Similar conclusions were found even with added hormonal treatment in both groups.

In the Morris review¹³¹the acute toxicity induced by similar doses of radiation applied by conventional and conformal radiotherapy was also reviewed, and three randomised studies with revealing information were identified:

In the study by Dearnaley et al from 1999¹³², statistically significant differences (p = 0.01) were found in the incidence of acute gastrointestinal toxicity grade ≥ 2 (proctitis with bleeding), with a frequency of 5% for conformal radiotherapy and 15% for conventional, at a dose of 64 Gy. No significant differences were found in bladder function.

In the trial by Koper et al¹³³, which applied a dose of 66 Gy in both groups, a gastrointestinal toxicity of grade 2 was observed in 32% for conventional radiotherapy and 19% for conformal radiotherapy, characterised by anal toxicity and proctitis (p = 0.02).

The randomised study by Storey et al¹³⁴, which compares conventional and conformal radiotherapy with escalating doses, identified no statistically significant differences in acute rectal or bladder toxicity (p = 0.6).

In addition, the Morris review identified 15 non-randomised articles for which no statistically significant differences were found in toxicity when comparing the equivalent dose application of conformal radiotherapy with conventional radiotherapy. This included a minimum follow-up period of 2 years.

In another clinical trial by Dearnaley et al from 2007¹³⁵, improved results were seen for intestinal toxicity (adverse effect frequencies of 8% and 5%, but without statistically significant differences) for the conformal radiotherapy group with escalating doses.

The systematic review of the Galicia Health Technologies Evaluation Agency, evaluation-t¹³⁶ analysed the safety and efficacy of treatment with intensity modulated radiation therapy (IMRT). This is a (more advanced) 3-dimensional conformal radiotherapy technique, evaluated on patients with localised and locally advanced (T1-T3) prostate cancer. Three retrospective localised prostate cancer studies of poor quality were found which compare IMRT and 3D-CRT. No statistically significant differences were found regarding efficacy. As for safety, better (and statistically significant) results were found for IMRT on the quality of life related to the sexual sphere (p = 0003). Patients treated with IMRT also obtained more favourable (and statistically significant) results in connection with late rectal toxicity grade 2-3 (p < 0.001).

The systematic review of Hummel et al¹³⁹ attempts to assess the clinical efficacy of new and emerging technologies for localised prostate cancer. With regard to cryotherapy (cryoablation of the prostate) and HIFU (high intensity focused ultrasound), analysed using non-comparative studies, it concludes that there is no evidence to support their use as a first line of treatment.

Another systematic review of the National Institute for Health and Clinical Excellence (NICE) in the United Kingdom¹⁴⁰ evaluates the safety and efficacy of HIFU for the treatment of prostate cancer. The localised prostate cancer studies were case series with short follow-up periods (less than 2 years). It also concluded that it is an experimental procedure, and not a first choice treatment.

The systematic revision of Shelley et al¹⁴¹ compared the efficacy and adverse effects of cryotherapy with those of other primary treatments (radical prostatectomy, radiation therapy and observation) for the management of patients with T1-T3 prostate cancer. A comparative study only was found. Separate results for localised prostate cancer were not found. It considers cryotherapy to be an experimental procedure, and therefore not a first choice treatment.

The Guazzoni et al study¹⁴⁷ is a randomised clinical trial of 120 patients with clinically localised prostate cancer subject to open (ORP) or laparoscopic radical prostatectomy (LRP)g carried out by the same surgeon with extensive experience in both techniques. No differences in the rates of positive surgical margins could be found when comparing both groups (ORP v LRP), but there are better results with laparoscopic for blood loss (mean ± standard deviation: 853.3 ± 485 v 257.3 ± 177 cm3; p < 0.001); catheter removal rate within 5 days (33.4% v 13.4%); operating time (mean ± standard deviation: 170 ± 34.2 v 235 ± 49.9 min; p < 0.001), and post-operative pain on the first day (p = 0.250). No data were available for long-term safety and efficacy.

A systematic review by the United Kingdom National Institute for Health and Clinical Excellence¹⁴³ evaluated the safety and efficacy of LRP, in comparison with ORP, for localised prostate cancer. It included non-randomised and case series studies. No statistically significant differences were found between LRP (transperitoneal - TLRP, extraperitoneal - ELRP or robot-assisted: RALRP) and ORP for either biochemical progresion free-survival or urinary incontinence with a follow-up of less than 3 years. There are no statistically significant differences with regard to urinary continence. And, although not significant (due to low sample size), there are differences with regard to sexual impotence, with a tendency to get better results for LRP in different studies.

A systematic review of Tooher et al¹⁴³ compares LRP (transperitoneal, extraperitoneal or robot-assisted) and ORP. It includes non-randomised comparative studies. The safety and adverse effects, including urinary incontinence, are very similar for the different types of LRP and ORP: TLRP v ORP, similar (complications average 2% v 0%); ELRP v ORP, similar; RALRP v ORP, higher complication rate for ORP.

The study of Hu et al¹⁴⁶ included 2,702 men treated with LRP v ORP. Those treated with laparoscopic prostatectomy were found to be younger (p < 0.001). This study offered no information on other relevant clinical or anatamopathological data (pre-operative PSA, Gleason score, clinical stage). A lower rate of preoperative complications was found with the minimally invasive treatment (29.8% v 36.4%; p = 0.002), in addition to shorter hospital stays (1.4 v 4.4 days; p < 0.001). However, patients who received LRP received salvage treatment more frequently than for ORP (27.8% v 9.1%; p <0001). Regarding the need for salvage treatment, better results were obtained by surgeons who had performed more laparoscopic prostatectomies the previous year (OR = 0.92; [95% CI: 0.88-0.99]), although the need for subsequent salvage treatment was still higher for ORP. The results of this study in light of the clinical or pathological patient data were not analysed.

The Bhatta-Dhar et al study¹⁵³ is a cohort study with a 6-year follow-up of 336 patients with clinically localised prostate cancer, PSA < 10 ng/ml and Gleason < 7 (low risk) who underwent prostatectomy. The decision to perform a lymphadenectomy (LN) or not was taken by the surgeon. After 6 years no significant difference was found in PSA relapse-free survival between patients with or without LN.

The study by Allaf et al¹⁵⁰ is a retrospective cohort study involving 4,000 patients with clinically localised prostate cancer, comparing extended lymphadenectomy (n = 2,135) v limited (n = 1,865), with each technique applied by a different surgeon. No statistically significant differences in the results for biochemical progression-free survival at 5 years were found. No differences were found when comparing extended vs limited for biochemical recurrence-free survival in patients with positive lymph nodes, although there was a trend for improved survival results in patients who underwent extended dissection (p = 0.07). More positive lymph nodes were detected with the extended treatment (mean 14.7 v 12.4; p = 0.15) as well as more patients with lymph node affectation (p < 0.0001).

The 2003 study from Bader et al¹⁵¹ is a cohort study involving 367 men with clinically localised prostate cancer subjected to prostatectomy, with a comparison of results with and without LN. 25% (92 patients) had positive lymph nodes. 43% of the patients had a pathological stage pT3 (infra staging), this group had a greater chance of having positive lymph nodes than those who were in stages pT1-T2 (39% v 13%). The existence of positive lymph nodes is statistically significantly associated with increased risk of progression, decreased cancer-specific survival (74% at 5 years) and a greater probability of relapse.

The study by Clark et al¹⁵² is a clinical trial that compares extended and limited lymphadenectomy in 123 patients with clinically localised prostate cancer. In this study, the same patient received an extended LN on one side and a limited LN on the other side. No statistically significant differences were found between both groups with respect to unilateral surgical complications. Positive lymph nodes were found in only 8 patients, making it impossible to find any statistically significant differences between each group.

The study by Sofer et al¹⁴⁸ is a retrospective cohort study evaluating the effect of radical prostatectomy (RP) with the preservation of neurovascular bundles (PNB) vs RP without PNB (the surgeon applies PNB when he feels that it is technically feasible, which can skew the results, because patients with PNB may be less at risk). The number of losses is not specified. The percentage of positive surgical margins was 24% in patients with PNB and 31% in those without PNB (no statistically significant differences were found). The cumulative risk of PSA relapse (BF) with PNB at 3 and 5 years of surgery was 9.7% and 14.4%, respectively. No statistically significant differences were found when comparing the BF of patients with PNB v patients without PNB after 3 years of surgery (not even when stratifying according to preliminary risk), nor when comparing unilateral PNB v bilateral PNB v patients without PNB. After adjusting for a number of variables (age, PSA and Gleason), there was no statistical difference in the probability of positive surgical margins between the two groups: OR = 0.89 [95% CI: 0.61-1.31].

The study by Robinson et al¹³⁰ is a systematic review comparing the rates of erectile dysfunction after RP with PNB vs other treatments. The results were obtained from non-randomised studies, with low sample size, and may be biased because they allow neoadjuvant hormonal therapy (which can block testosterone for up to a year after finishing the treatment). It was found that the probability of maintaining erectile function after RP + PNB 1 year after treatment is 0.34 [95% CI: 0.30-0.38], and after 2 years was 0.25 [95% CI: 0.18-0.33]. After adjusting for age, the probability 1 year after treatment is 0.22 [95% CI: 0-0.53]. The probability of erectile dysfunction for RP without PNB is 0.16 [95% CI: 0.0-0.37]. In other words, for patients with localised prostate cancer who have undergone a prostatectomy, the probability of erectile function is greater if the neurovascular bundles are preserved.

The study of Kundu et al¹⁵⁴ includes 1,834 patients who underwent retropubic RP with or without PNB, whether uni- or bilaterally. The neurovascular bundles were retained in only 5% of patients (91 out of 1,834). No statistically significant differences were found (p = 0.3) in the recovery of urinary continence when comparing RP and PNB v RP without PNB (minimum follow-up of 18 months).

The study of Wille et al¹⁵⁵ is a small sample size retrospective cohort study which analyses post-RP urinary continence results according to a number of variables. It consists of a questionnaire completed by 81% of those requested. It concludes that PNB (both uni- and bilateral) does not affect urinary continence results (no statistically significant difference between the performance or not of PNB in RP).

In a randomised clinical trial carried out by Peeters et al¹³⁷, which compared a dose of 68 Gy vs 78 Gy in 664 patients with prostate cancer T1b-T4, it was found that there were statistically significant differences between the two groups for biochemical progression-free survival (BPFS) at 5 years: 54% v 64% (p = 0.01).

The Khuntia et al study¹⁵⁷ is a prospective cohort study, which includes T1-T3 patients treated with external radiotherapy (RT). For patients with T1-T3 and low risk, biochemical progression-free survival at 5 years depending on the dose was 52% (≤ 68 Gy), 82% (68-72 Gy), 93% (≥ 72 Gy); p < 0.001.

The Kupelian et al publication¹⁵⁸ is a prospective dose escalation cohort study that analyses 292 patients with localised prostate cancer at low risk (cT1-cT2a and Gleason < 7 and PSA ≤ 10 ng/ml) treated with external beam radiotherapy (ERT). Statistically significant differences were found in the BPFS at 96 months when comparing ≤ 72 Gy vs > 72 Gy (77% v 95%; p = 0.01). When analysing by dosage subgroups at 4 years, again statistically significant differences were found when comparing <74 Gy (77%) vs > 74 Gy (94%), with p = 0.09. There were no differences when comparing 74 Gy (94%) vs 78 Gy (96%), with p = 0.90.

In a randomised clinical trial by Peeters et al¹³⁷ with T1b-T4 patients, it was found that there were no statistically significant differences for BPFS after 5 years when comparing 68 Gy vs 78 Gy in the low risk group. One cannot rule out that there were no differences, however, because the study had insufficient statistical power to analyse the subgroups and because some patients received a lesser dose than was planned initially.

In another clinical trial by Pollack et al¹⁵⁹ 70 Gy was compared to 78 Gy in patients with T1-T3 prostate cancer, where the minipelvis and prostate with vesicles were radiated. The results are shown according to different risk groups. In patients with PSA <10 ng/ml, there were no statistically significant differences in the BPFS.

The article by Hanks et al¹⁶⁰ is a prospective dose escalation cohort study, which analyses patients with localised prostate cancer treated with external RT (median 9 years of follow-up). For a PSA between 10-20 ng/ml, statistically significant differences were found in the BPFS when comparing 71.5 Gy vs 75.6 Gy vs > 75.6 Gy (19% vs 31% vs 84%); p = 0.0003.

The Peeters et al trial¹³⁷, which analysed patients T1b-T4 and the intermediate risk group, found that there are statistically significant better results for the BPFS at 5 years in the higher dose when comparing 68 Gy vs 78 Gy.

In the Khuntia et al study¹⁵⁷ of T1-T3 patients of intermediate risk, the BPFS at 5 years depending on the dose was 27% (≤ 68 Gy), 51% (68-72 Gy), 83% (≥ 72 Gy, median dose 78 Gy); p < 0.001. It also found that for T1-T3 patients at high risk, the BPFS at 5 years depending on the dose was 21% (≤ 68 Gy), 29% (68-72 Gy), 71% (≥ 72 Gy; median dose 78 Gy); p < 0.001. Moreover, by increasing the median dose from 70 Gy to 78 Gy, the greatest improvement was found in the intermediate and high risk group. In other words, in patients with T1 and T3 prostate cancer and intermediate and high risk, the best BPFS results at 5 years were in the ≥ 72 Gy group (median 78 Gy).

In the Pollack et al study¹⁵⁹, in T1 and T3 patients with PSA > 10 ng/ml, comparing 70 Gy vs 78 Gy, statistically significant better results at 5 years were found for the BPFS: 48% vs 75% (p = 0.011).

In the test Peeters et al study¹³⁷, in the high-risk group for BPFS at 5 years, there is a tendency to find better results for the higher dose when comparing 68 Gy vs 78 Gy.

In another Peeters article¹³⁸, the same patients as in the previous study¹³⁷ are included, but toxicity results are offered instead of efficacy results. It includes patients with T1-T4 prostate cancer. In this study, different volumes and dose limits (VD) are compared and different institutions are involved. When comparing 68 Gy vs 78 Gy (with a volume that includes the anus), no statistically significant differences were found for gastrointestinal toxicity grades 2 and 3 (p = 0.2; p = 0.4). However, statistically significant differences were found for rectal bleeding (3% vs 7%; p = 0.02) and anal incontinence (for faeces, mucus or blood, which require disposable pads more than twice a week; 6% vs 10%; p = 0.03). In other words, a dose of 78 Gy maintains anal bleeding and losses below 10% in patients with T1-T4 prostate cancer.

The publication by Vargas et al study¹⁶⁴ is a multicentre study which includes patients with clinically localised (86.5%) and locally advanced (13.5%) prostate cancer and a high risk of lymph node invasion, ie, above 15% (calculated according to the formulaj proposed by Roach et al163). ERT plus TPV (n = 312) is compared with ERT with PSSV (n = 284). The choice of volume to be used in the study depended on the centre treating the patient: TPV in two centres, PSSV in another centre. When comparing the two groups with a follow-up of 15 years, statistically significant differences were found in clinical failure (univariate p = 0.04, multivariate p = 0.9), but not for PSA relapse (univariate p = 0.8), clinical disease-free survival (p = 0.06), cancer-specific survival (p = 0.8) and overall survival (p = 0.6). In other words, for patients with clinically localised prostate cancer and a high risk of metastasis (greater than 15% risk), when comparing pelvic irradiation with prostate and seminal vesicles, no statistically significant differences for clinical control and cancer-specific survival with a follow-up period of up to 15 years were found.

The study by Jacob et al¹⁶¹ includes 420 men with prostate cancer and pre-treatment PSA of <100 ng/ml, treated with 3-dimensional conformal ERT with or without Androgen deprivation of short duration. The patients had a lymph node invasion risk of ≥ 15% or a cT2 stage with Gleason 6-10. POV fields were applied in 48 cases, MPV in 74, and TPV in 298. In this study, the irradiated volume was not a significant predictor of outcome.

The 2003 study by Roach et al¹⁶³ is a random clinical trial comparing RT with PV + neoadjuvant hormone therapy (HT) vs RT with PV + adjuvant HT vs RT with POV + neoadjuvant HT vs RT with POV + adjuvant HT in patients with prostate cancer (67% were pT2c-pT4). When comparing PV vs POV at 4 years, statistically significant differences for progression-free survival were found (54.2% vs 47.0%; p = 0.02) and biochemical progression-free survival (40.7% vs 33.5%; p = 0.007), but not for overall survival (84.7 vs 84.3%; p = 0.94), PSA relapse rate (34% vs 40%; p = 0.089), lymph node failure nor metastasis at a distance.

In the article by Lawton et al¹⁶⁶, the results from the 2003 Roach study163 were updated with 1,292 cases and a longer follow-up period, of up to 10 years (with a median of 7 years). When comparing PV with POV, no statistically significant differences were found in progression-free survival (p = 0.99) nor for biochemical progression-free survival (p = 0.93).

In another article by Roach et al¹⁶², published in 2006, an analysis was done of the patient subgroups who received neoadjuvant HT from the 2003 Roach study163 (those who had obtained the best results for larger volumes), with a longer follow-up (up to 9 years, with a median of 7).

In this article from 2006, the patients were divided into 3 groups according to the volume received: TPV (n = 309) vs MPV (n = 170) vs POV (n = 131). Of the patients studied, 67% were pT2c-pT4, and all of them received neoadjuvant HT.

Statistically significant differences were found (p = 0.024) for progression-free survival at 9 years when comparing the 3 groups: 40% (TPV), 35% (MPV) and 27% (POV), and also when comparing TPV vs POV (p = 0.010; in favour of TPV), but not for TPV vs MPV (p = 0.06).

No statistically significant differences were found (p = 0.06) for biochemical progression-free survival at 9 years when comparing the 3 groups, nor when comparing TPV vs MPV (p = 0.12). However, statistically significant differences were found for TPV when compared with POV (p = 0025).

Statistically significant differences were also found for TPV in the percentage of PSA relapse at 9 years when comparing the 3 groups with each other (p = 0.025), when comparing TPV with POV (p = 0.029) and with MPV (p = 0.022).

When analysing the results at 4 years according to the type of hormone therapy received in the 2003 article by Roach¹⁶³, when PV + neoadjuvant HT was compared with POV + neoadjuvant HT, statistically significant differences were found in progression-free survival (54.2% vs 47.0%; p = 0022), but not for overall survival (84.7% vs 84.3%; p = 0.94), local progression (9.1% vs 8.0%; p = 0.78), lymph node failure (1.3% vs 2.5%; p = 0.12) nor distant metastasis (8.2% vs 6.6%; p = 0.54).

Lawton¹⁶⁶ did not find statistically significant results for biochemical progression-free survival at 10 years according to the type of HT received, nor when comparing PV + neoadjuvant HT with POV + neoadjuvant HT (p = 0.066), or PV + adjuvant HT against POV + adjuvant HT (p = 0.057). These results were only offered for a definition of biochemical progression different to the global analysisk. Regarding overall survival, there is no statistical difference when comparing PV +neoadjuvant HT with POV + neoadjuvant HT (p = 0.9629). However, POV + adjuvant HT has better results than PV + adjuvant HT (p = 0.01).

In the 2003 article by Roach et al¹⁶³, when comparing the different groups (RT with PV + neoadjuvant HT vs RT with PV + adjuvant HT vs RT with POV + neoadjuvant HT vs RT with POV + adjuvant HT), no statistically significant differences in the following types of toxicity were found: grade ≥ 3, acute (p = 0.06) and late (p = 0.09) gastrointestinal; and acute (p = 0.39) and late (p = 0.85) genitourinary.

In the 2006 article by Roach et al¹⁶², the following toxicity results were found when comparing TPV vs MPV vs POV: Acute gastrointestinal toxicity ≥ grade 2 was 46.5% vs 36.7% vs 20.2%; p < 0.001. The late was 15.2% vs 8.5% vs 7%; p = 0.002. Acute genitourinary toxicity of grade ≥ 2 was 31.4% vs 37.7% vs 22.1%; p = 0.016. The late was 14.9% vs 14.7% vs 5.6%; p = 0.03.

The study by Kupelian et al¹⁶⁹ is a series of cases of 770 patients treated with hypofractionation for 5 years, with a biochemical progression-free survival of 82% [95% CI: 79-85]. In addition, there is another set of 300 cases treated with hypofractionation, from Higgins et al¹⁷⁰, who also received neoadjuvant hormone therapy, which found a biochemical progression-free survival of 57.3%, and cancer-specific survival rate of 83.2% at 5 years.

The study by Kumar et al¹⁷¹ compared the effectiveness and side effects of hormone therapy added to local treatment (radical prostatectomy or radiotherapy) vs local treatment in patients with localised and locally advanced prostate cancer (sometimes without separating the two groups).

In the review by Kumar et al¹⁷¹, for patients with T1 and T2 disease with localised prostate cancer risk at low and intermediate risk [cT2b or Gleason = 7 or (PSA > 10 and ≤ 20 ng/ml)] who received radical prostatectomy, the addition of neoadjuvant HT did not improve overall survival (OR = 1.11 [95% CI: 0.67-1.85]; p = 0.69). There was no available data on disease-associated survival (DAS). A significant limit reduction on relapse rates was found (OR = 0.74 [95% CI: 0.55-1.0]; p = 0.05).

In the only article about the Kumar review¹⁷¹ which analysed patients with T1-T2 Nx localised prostate cancer who received radical prostatectomy (McLeod et al¹⁷²), the addition of adjuvant HT (bicalutamide 150 mg/day) did not improve survival.

The study by D'Amico et al¹⁷³ is a high-quality randomised clinical trial that includes localised prostate cancer patients (most of them at low risk, cT1-cT2a and Gleason < 7 and PSA ≤ 10 ng/ml) in which (3-dimensional conformal) radiation treatment + neoadjuvant androgen suppression HT is compared with 3D-CRT; in both cases at a dose of 30 Gy. The overall survival at 5 years in the group treated with RT + HT was 88% [95% CI: 80-95%] and 78% [95% CI: 68-88%] in those treated with RT. In other words, no statistically significant difference in overall survival between the two groups was found 5 years after treatment.

In the McLeod et al study¹⁷² for patients with localised prostate cancer who received radical radiotherapy the addition of adjuvant HT (bicalutamide 150 mg/day) did not improve survival.

The Hummel et al study¹³⁹ is a systematic review comparing different treatments for localised prostate cancer. In the comparison of local treatment + neoadjuvant HT v local treatment, no differences were identified in terms of biochemical progression-free survival (BPFS), and when comparing local treatment + adjuvant HT vs local treatment, no differences were identified in terms of survival, although there were indications that high-risk patients could benefit from HT added to local treatment.

In the Kumar review¹⁷¹ and the D'Amico study¹⁷³, for patients with localised prostate cancer who received radical treatment, the addition of HT (neoadjuvant or adjuvant) increased adverse events (hot flushes, diarrhoea, fatigue, gynecomastia).

To assess how to monitor men with localised prostate cancer subjected to radical prostatectomy, firstly, the Han et al case series¹⁷⁴ was investigated. The study followed 2,404 such patients over 15 years. It found that no patient experienced local or distant recurrence without the PSA level increasing. In addition, patients with clinical stage T1a or Gleason 2-4 (a subgroup of 50 cases) experienced no PSA relapse1. Patients with clinical stage T1b-T1c did not experience PSA relapse within 10 years of monitoring. In the rest of the clinical stages, no patient had PSA relapse after 15 years.

Another publication by Kupelian et al¹⁷⁵ compared patients with localised prostate cancer treated with prostatectomy vs radiotherapy^m. The percentage of the 1,467 prostatectomised patients with biochemical progression-free survival (BPFS)ⁿ was 79% [95% CI: 77-81%] at 5 years and 67% [95% CI: 64-71%] at 10 years. While the percentage of the 1,049 irradiated patients was 66% [95% CI: 63-69%] at 5 years and 62% [95% CI: 58-65%] at 10 years. The survival curve stabilised around 6.5 years after radiotherapy treatment and 13 years after the operation.

The Hanks et al study¹⁶⁰ is a series of 229 cases of localised prostate cancer treated with 3-dimensional conformal external beam radiation (3D-CRT), which was the standard radiotherapy treatment for these patients. Biochemical progression-free survival^o was 55% at 5 years, 48% at 10 years and 48% at 12 years, with no statistically significant differences when compared with each other. The BPFS curve stabilised around 7.2 years. When stratified into different prognosis groups according to the pre-treatment PSA level, no statistically significant differences for BPFS were found. In patients who had pre-treatment PSA <10 ng/ml, the biochemical progression-free survival at 8 years was between 68% and 74%.

The Albertsen et al case series⁴⁷ followed 767 patients with localised prostate cancer who did not receive treatment with intent to cure, and found that the cancer-specific survival curve for these patients stabilised at 15 years. This result is considered extrapolable to the rest of the patients who did receive treatment.

Since no studies have been found comparing the PSA monitoring frequency guidelines, it is proposed to follow the recommendations of the 2007 prostate cancer clinical practice guide from the European Association of Urology⁴, which proposes reviews at 3, 6 and 12 months after treatment with intent to cure, then every 6 months after the 1st year and annually after the 3rd year.

To be able to establish the recommendations, the 2005 consensus of the International Society of Urological Pathology (ISUP)²⁴, was also taken into account. It was agreed that the diagnosis of a Gleason summation grade^2-4 in the prostatectomy sample should be an exception (only in transition zone tumours) and should always be confirmed.