Does hormone treatment added to radiotherapy improve outcome in locally advanced prostate cancer?

Meta-Analysis of Randomized Trials

Abstract

BACKGROUND:

To quantify the magnitude of benefit of the addition of hormone treatment (HT) to exclusive radiotherapy for locally advanced prostate cancer, a literature-based meta-analysis was conducted.

METHODS:

Event-based relative risks (RR) with 95% confidence intervals (CIs) were derived through a random-effect model. Differences in primary (biochemical failure and clinical progression-free survival) and secondary outcomes (cancer-specific survival, overall survival [OS], recurrence patterns, and toxicity) were explored. Absolute differences and numbers of patients needed to treat (NNT) were calculated. A heterogeneity test, a metaregression analysis with clinical predictors of outcome, and a correlation analysis for surrogate endpoints were also performed.

RESULTS:

Seven trials (4387 patients) were gathered. Hormone suppression significantly decreased both biochemical failure (RR, 0.76; 95% CI, 0.70-0.82; P < .0001) and clinical progression-free survival (RR, 0.81; 95% CI 0.71-0.93; P = .002), with absolute differences of 10% and 7.7%, respectively, which translates into 10 and 13 NNT. cancer-specific survival (RR, 0.76; 95% CI, 0.69-0.83; P < .0001) and OS (RR, 0.86; 95% CI, 0.80-0.93; P < .0001) were also significantly improved by the addition of HT, without significant heterogeneity, with absolute differences of 5.5% and 4.9%, respectively, which translates into 18 and 20 NNT. Local and distant relapse were significantly decreased by HT, by 36% and 28%, respectively, and no significant differences in toxicity were found. Primary and secondary efficacy outcomes were significantly correlated.

CONCLUSIONS:

Hormone suppression plus radiotherapy significantly decreases recurrence and mortality of patients with localized prostate cancer, without affecting toxicity. Cancer 2009. © 2009 American Cancer Society.

Androgen deprivation through hormone suppression with luteinizing hormone–releasing hormone (LH-RH) analogues remains the cornerstone of treatment for patients with hormone-sensitive advanced prostate cancer.1 Clinical studies have shown that 17% to 64% of patients with locally advanced prostate cancer who receive radiotherapy alone experience clinical progression, and 26% to 43% die within 5 years of initiating treatment.2-4

Antiandrogen therapy has an independent cytotoxic effect on prostate cancer cells, and the rationale for combining androgen deprivation with radiation is to act as a “sensitizer” for radiation to enhance tumor cell kill, and to eradicate micrometastatic disease beyond the radiation volume. The optimal duration of androgen ablation therapy remains controversial, despite a growing number of prognostic factors that have been shown to predict disease outcome.

With respect to the timing of androgen deprivation in early stage disease, multiple studies have now been conducted to establish the potential benefit of androgen deprivation therapy combined with radiotherapy (RT).

Seven randomized trials have reported on the use of neoadjuvant androgen deprivation or concurrent hormone suppression in conjunction with RT.5-14 These studies have demonstrated an overall survival advantage with the addition of hormone suppression or antiandrogen to RT, using various treatment schedules in patients with intermediate and high risk. However, given the relatively long natural history of the disease, it takes a considerable amount of time for clinical trials to provide mature data; therefore, the validation of reliable surrogate endpoints is an emerging priority.15

To quantify the potential benefits of hormone therapy (HT) combined with RT in terms of progression-free survival (PFS), overall survival (OS), and cancer-specific survival, a meta-analysis was conducted.

MATERIALS AND METHODS

The analysis was conducted following 4 steps: definition of the outcomes (definition of the question the analysis was designed to answer), definition of the trial selection criteria, definition of the search strategy, and a detailed description of the statistical methods used.16, 17

Outcome Definition

The combination of HT and RT was considered as the experimental arm, and exclusive RT as the standard comparator. Analysis was conducted to find significant differences in primary and secondary outcomes, according to the reported sequence and definitions in the selected trials. Primary outcomes for the magnitude of the benefit analysis were both biochemical failure, time between randomization and prostate-specific antigen increase) and clinical PFS (clinical progression-free survival, time between randomization and clinical appearance of local and/or distant relapse or death by any cause). Secondary endpoints were: 1) cancer-specific survival (time between randomization and death for prostate cancer), 2) OS (time between randomization and death by any cause), 3) local failure rate, 4) distant metastases rate (DM), 5) overall grade 3-4 toxicities, 6) genitourinary grade 3-4 toxicity (GU), 7) gastrointestinal grade 3-4 toxicities (GI), and 8) cardiac deaths.

A sensitivity analysis taking into account treatment duration was accomplished by evaluating the effect of HT on efficacy outcomes in trials adopting either long-term (≥1 year) or short-term (<1 year) treatment duration.

Search Strategy

Deadline for trial publication and/or presentation was November, 2008. Updates of randomized clinical trials (RCTs) were gathered through Medline (PubMed, www.ncbi.nlm.nih.gov/PubMed), American Society of Clinical Oncology (ASCO, www.asco.org), European Society for Medical Oncology (ESMO, www.esmo.org), Federation of European Cancer Societies (www.fecs.be), and American Society for Therapeutic Radiology and Oncology (ASTRO, www.astro.org) website searches. Keywords used for searching were: adjuvant hormone therapy, prostate cancer, radiotherapy, review, meta-analysis, meta-analysis, pooled analysis, randomized, phase 3, comprehensive review, and systematic review. In addition to computer browsing, the reference sections of review and original articles were also scanned for missing trials. Furthermore, lectures at major meetings (ASCO, ESMO, European Cancer Conference [ECCO], and ASTRO) having “hormone treatment and radiotherapy for prostate cancer” as the topic were checked. No language restrictions were applied.

Trial Identification Criteria

Included were all prospective phase 3 RCTs published in peer-reviewed journals or presented at the ASCO, ECCO, ESMO, and ASTRO meetings before November 2008, in which previously untreated patients with locally advanced prostate cancer were prospectively randomized to receive exclusive RT (control arm) or HT (hormone suppression with LH-RH analogues or antiandrogen) plus RT (experimental arm), regardless of drug, schedule, dosages, duration, and RT technique.

Data Extraction

The number of events for primary and secondary endpoints were extracted; the last trial's available update was considered as the original source. All data were reviewed and separately computed by 5 investigators (E. B., F. Cu., D. G., I. S., and P. C.).

Data Synthesis

The log of relative risk ratio (RR) was estimated for each considered endpoint,18 and 95% confidence intervals were derived.19 A random-effect model according to the inverse variance and the Mantel-Haenzel method was preferred to the fixed, given the known clinical heterogeneity of trials; a Q-statistic heterogeneity test was used. Absolute benefits for each outcome were calculated (ie, absolute benefit = exp [RR × log{control survival}] − control survival20; modified by Parmar and Machin21). The number of patients needed to treat for 1 single beneficial patient was determined (NNT: 1/[{absolute benefit}/100]).22 Results were depicted in all figures as conventional meta-analysis forest plots; an RR < 1.0 indicates fewer events in the experimental arm. To find possible correlations between outcome effect and negative prognostic factors (selected among trials' reported factors: lymph node-positive, Gleason score between 7 and 10, and tumor size T3-T4 rates), a metaregression approach was adopted (ie, regression of the selected predictor on the log RR of the corresponding outcome). Calculations were accomplished using SPSS software, version 13.0 (SPSS Inc., Chicago, Ill) and the Comprehensive Meta-Analysis Software, version v. 2.0 (Biostat, Englewood, NJ).17

Correlation

Potential correlations to test surrogacy between primary endpoints and cancer-specific survival/OS were explored according to a linear regression model considering both the outcome rates for each single arm and the calculated RRs and their logs for each outcome in paired comparisons. Correlations were estimated according to the Pearson (r) and R² coefficients (parametric) and the Spearman (ρ) coefficient (nonparametric).

Power Analysis

To evaluate if the performed meta-analysis has enough statistical power to determine the obtained results, a sample size determination was accomplished. Given that differences in median survival across the included trials ranged widely, and to assess how many patients would eventually be required to determine the benefit, a model to calculate the target sample size for a cancer-specific survival benefit of 3%, 4%, 6%, and 7%, using either biochemical failure or clinical progression-free survival beta coefficients according to the regression, was accomplished.

RESULTS

Selected Trials

Eleven trials (7535 patients) were identified5, 7, 8, 10-14, 23-25 (Fig. 1). One paper reported the results of 2 trials8; 1 was excluded because it was designed to compare HT duration without an exclusive RT arm. Three additional RCTs were excluded for the same reason.23-25 Seven and 5 RCTs, respectively, were evaluable for biochemical failure (3956 patients, data lacking for 179 patients) and clinical progression-free survival (4020 patients, data lacking for 2 RCTs8, 13). With regard to secondary outcomes, 1 trial (161 patients) was excluded for lack of data.8 Six RCTs (4266 patients) provided data for cancer-specific survival, OS, and cardiac deaths. Data were available for local failure rate, DM, and toxicity analyses in 4 RCTs (2650 patients with 2 trials excluded,12, 13 and 2050 patients with 2 trials excluded,7, 12 respectively). HT was administered according to an long-term or short-term approach in 35, 10, 12, 14 and 47, 8, 11, 13 trials, respectively. Trials characteristics are listed in Table 1. Median follow-up ranged from 4.5 to 10.1 years.

Figure 1. A flow diagram outlines the search. RCT indicates randomized clinical trial; HT, hormone treatment; RT, radiotherapy; pts: patients; BF, biochemical failure; CSS, cancer-specific survival; OS, overall survival; CPFS, clinical progression-free survival; LR, local relapse; DM, distant metastases.

Table 1. Trial Characteristics

Authors	Pts	HT Duration	Experimental Arm	Median FU, y	Primary Endpoint	Secondary Endpoint	% Node Positive	% Gleason Score 7-10	% T3-T4
Pts indicates patients; HT, hormone treatment; FU, follow-up; LT, long term; EBRT, external beam radiotherapy; Gy, gray; DFS, disease-free survival; OS, overall survival; LR, local relapse; DM, distant metastases; CSS, cancer-specific survival; PFS, progression-free survival; ST, short term; AST, androgen suppression therapy; 3D, 3-dimensional; BF, biochemical failure; STAD, short-term androgen deprivation; AS, androgen suppression; BNED, no biochemical evidence of disease; NR, not reported; ADT, androgen deprivation therapy.
Bolla 20025,200814	415	LT	Goserelin 3 y+EBRT (70 Gy)	9.1	DFS	OS	3.4	34.1	91.0
Pilepich 200510	977	LT	Goserelin until progression+EBRT (65-70 Gy)	7.6	LR, DM, DFS	CSS	26.9	62.2	69.8
See & Tyrrell 200612	1,370	LT	Bicalutamide 150 mg (median 1.8 y)+EBRT (64 Gy)	7.2	PFS	OS	1	24.6	22.0
D'Amico 200813	206	ST	AST 6 mo prior and concurrent with 3D-EBCRT (70 Gy)	7.6	BF	CSS, OS	0	72.3	0
Denham 20057	802	ST	STAD 3 or 6 mo prior and concurrent with EBRT (66 Gy)	5.9	Time to LR, CSS	Time to LR, time to DM	0	38.0	40.1
Laverdiere 20048	161	ST	3 mo AS (Group 2) or 10 mo AS (Group 3)+EBRT (64 Gy)	5.0	BNED	NR	0	26.0	30.0
Roach 200811	456	ST	ADT 4 mo prior and concurrent with EBRT (65-70 Gy)	12.6	LR	DFS, OS	8.1	65.9	70.0

Combined Analysis

Primary outcomes

HT significantly decreased biochemical and clinical failure over exclusive RT by 24% and 19%, respectively. The absolute benefit was 10% for biochemical failure and 7.7% for clinical progression-free survival, corresponding to 10 and 13 NNT, respectively (Table 2). The benefit was obtained regardless of HT duration, as shown by the sensitivity analysis (Table 2, Fig. 2).

Figure 2. Combined results are shown according to sensitivity analysis. BF indicates biochemical failure; HT, hormone treatment; CI, confidence intervals; LT, long-term HT; ST, short-term HT; RT, radiotherapy; CPFS, clinical progression-free survival; CSS, cancer-specific survival; OS, overall survival; LR, local relapse; DM, distant metastases.

Table 2. Combined Efficacy Results According to the Sensitivity Analysis

Outcomes	Pts (RCTs)	RR (95% CI)	P	Heterogeneity P	% AD	NNT
Pts indicates patients; RCT, randomized clinical trial; RR, relative risk; CI, confidence interval; AD, absolute difference; NNT, number needed to treat; BF, biochemical failure; LT, long-term hormone treatment; ST, short-term hormone treatment; CPFS, clinical progression-free survival; CSS, cancer-specific survival; OS, overall survival, LR, local relapse; DM, distant metastases.
BF	3956 (7)	0.76 (0.70-0.82)	<.0001	.08	10.0	10
LT	2656 (3)	0.79 (0.75-0.83)	<.0001	.50	8.6	12
ST	1300 (4)	0.67 (0.55-0.82)	<.0001	.006	14.2	7
CPFS	4020 (5)	0.81 (0.71-0.93)	.002	<.0001	7.7	13
LT	2762 (3)	0.81 (0.61-0.95)	.011	.005	7.4	14
ST	1258 (2)	0.83 (0.67-1.02)	.088	<.0001	—	—
CSS	4266 (6)	0.76 (0.69-0.83)	<.0001	.56	5.5	18
LT	2762 (3)	0.77 (0.60-0.84)	<.001	.89	5.3	19
ST	1464 (3)	0.67 (0.49-0.91)	.022	.25	7.2	14
OS	4266 (6)	0.86 (0.80-0.93)	<.0001	.36	4.9	20
LT	2762 (3)	0.84 (0.75-0.94)	.003	.21	5.6	18
ST	1464 (3)	0.87 (0.79-0.97)	.013	.34	4.1	21
LR	2650 (4)	0.64 (0.54-0.75)	<.0001	.27	9.8	11
LT	1392 (2)	0.65 (0.53-0.78)	<.0001	.36	8.7	12
ST	1258 (2)	0.61 (0.44-0.84)	.002	.09	11.8	8
DM	2650 (4)	0.72 (0.65-0.81)	<.0001	.43	9.5	11
LT	1392 (2)	0.70 (0.61-0.79)	<.0001	.99	11.1	9
ST	1258 (2)	0.80 (0.65-0.99)	.04	.24	5.7	17

Secondary outcomes

HT significantly reduced the risk of death for prostate cancer by 24%, without significant heterogeneity. This corresponds to an absolute benefit of 5.5%, with 18 NNT (Table 2). In the sensitivity analysis, the absolute benefit in cancer-specific survival ranges from 5.3% in the long-term trials to 7.2% in the short-term trials. HT significantly decreased the risk of death by any cause by 14%, regardless of treatment duration, with an absolute benefit of 4.9%, corresponding to 20 NNT (Table 2). With regard to recurrence, both local relapse (LR) and DM were significantly decreased (36% and 28%, respectively) by the addition of HT to RT, with a 9.8% and 9.5% absolute benefit, corresponding to 11 NNT (Table 2). No significant differences in toxicities and cardiac deaths were observed by comparing the 2 arms, without heterogeneity (Table 3).

Table 3. Combined Toxicity and Cardiac Deaths Results

Outcomes	Pts (RCTs)	RR (95% CI)	P	Heterogeneity P
Pts indicates patients; RCT, randomized clinical trial; RR, relative risk; CI, confidence interval; GU, genitourinary; GI, gastrointestinal.
Overall toxicity	2050 (4)	0.92 (0.87-1.11)	.41	.55
GU toxicity	2050 (4)	0.66 (0.36-1.22)	.19	.05
GI toxicity	2050 (4)	0.69 (0.46-1.03)	.07	.71
Cardiac deaths	4266 (6)	0.87 (0.70-1.09)	.24	.69

According to the metaregression analysis, none of the considered predictors significantly affected outcome, with the exception of lymph node positivity and Gleason score, which significantly influenced clinical progression-free survival (Table 4).

Table 4. Meta-regression Analysis Model P Values According to Selected Predictors (ie, Regression of the Selected Predictor on the Log Risk Ratio of the Corresponding Outcome)

Outcome	Node Positive	Gleason 7-10	T3-T4
BF, biochemical failure; CPFS, clinical progression-free survival; CSS, cancer-specific survival; OS, overall survival; LR, local relapse; DM, distant metastases.
BF	.22	.65	.62
CPFS	.026	.00003	.24
CSS	.67	.23	.39
OS	.51	.44	.99
LR	.28	.06	.17
DM	.50	.85	.16

Correlation Analysis

The correlation analysis was performed in 6 RCTs, in which biochemical failure could be considered as a potential surrogate for survival, and in 5 RCTs, which considered clinical progression-free survival. Both primary outcomes significantly correlated with either cancer-specific survival or OS, regardless of the adopted coefficients (Table 5). The correlations between the log of the RRs of primary outcomes and the RRs of cancer-specific survival and OS are shown in Figure 3.

Figure 3. Correlation analysis data are shown. BF indicates biochemical failure; CSS, cancer-specific survival; RR, relative risk; OS, overall survival; CPFS, clinical progression-free survival.

Table 5. Correlation Analyses; Coefficients, and P Values According to Method and Outcome

	CSS (P)		OS (P)
	Pearson/R^2	Spearman	Pearson/R^2	Spearman
CSS, cancer-specific survival; OS, overall survival; BF, biochemical failure; RR, relative risk; CPFS, clinical progression-free survival; NE, not evaluable.
BF
Rates	0.78/0.71 (.003)	0.86 (.03)	0.72/0.53 (.007)	0.71 (.02)
RRs	0.92/ 0.85 (.01)	0.49 (.28)	0.81/ 0.65 (.04)	1.00 (.02)
CPFS
Rates	0.88/0.78 (.0007)	0.94 (.005)	0.77/0.60 (.009)	0.69 (.04)
RRs	0.89/0.79 (.04)	NE	0.99/ 0.97 (.002)	0.90 (.06)

Power Analysis

On the basis of power calculations, 6209, 3467, and 1085 patients would have been required to demonstrate a statistically significant cancer-specific survival benefit of 3%, 4%, and 7%, respectively. With >4000 patients to demonstrate a 5.5% absolute cancer-specific survival benefit, the present meta-analysis largely exceeds the statistical power required (1487 patients needed to determine a 6% cancer-specific survival benefit).

We also derived a model to calculate the numbers of patients needed to demonstrate a 3%, 4%, 6%, and 7% absolute improvement in cancer-specific survival, using biochemical failure or clinical progression-free survival as surrogate endpoints; given the correlation between surrogate endpoints and cancer-specific survival (Table 5 and Fig. 3), the target cancer-specific survival improvements correspond to a 5.3%, 7.1%, 10.6%, and 12.4% improvement and to a 5.6%, 7.5%, 11.2%, and 13.1% improvement for biochemical failure and clinical progression-free survival, respectively. By using such surrogate endpoints instead of cancer-specific survival as a target for trial design, the corresponding required sample sizes would be 2321, 1327, 603, and 447 patients, and 2220, 1254, 569, and 442 patients, respectively.

DISCUSSION

The present meta-analysis demonstrates that the administration of hormone-suppressive therapy in patients affected by prostate cancer who are candidates to receive exclusive RT significantly improves all investigated outcomes. Although with significant heterogeneity in many of the endpoints, the overall absolute benefit is in the range of 7.5% to 10% in favor of HT for both primary outcomes, biochemical failure and clinical progression-free survival (Table 2). The choice of biochemical failure and clinical progression-free survival as primary endpoints is justifiably arguable; however, it was supported by their incorporation into the primary endpoint of most of the trials considered, as well as by a general drive toward the use of surrogate endpoints in oncology. Although a literature-based meta-analysis cannot deeply investigate the surrogacy of an endpoint over another, the correlation rates clearly indicate that both primary endpoints significantly correlate with the more clinically valuable outcomes of cancer-specific survival and OS (Table 5, Fig. 3). Indeed, in addition to biochemical failure and clinical progression-free survival, survival of patients receiving HT is increased, with an absolute reduction in deaths due to prostate cancer or deaths due to any cause of 5.5% and 4.9%, respectively. Although data on LR and DM were available for only a small subset of the patient population considered, the risk of local recurrences and distant metastases were also significantly decreased.

To easily understand the impact of absolute benefits, we also calculated the NNT. Indeed, this method offers an immediate perspective of the magnitude of benefit achievable by implementing a specific medical intervention, and improves the quality of cumulative analyses.26, 27 In the medical oncology scenario, adjuvant treatments rarely provide absolute benefits (5%-10%, depending on the outcome considered) and NNTs10-20 are found in the range of those observed in the present meta-analysis (Table 2 and Refs. 28, 29), calling for rapid and extensive implementation of HT in addition to definitive RT in the daily clinical practice.

Although comparing the duration of hormone suppression was neither an endpoint of the individual RCTs considered nor within the scope of the present meta-analysis, this particular subject was preliminarily explored with a sensitivity analysis approach, in light of its potential relevance for daily clinical practice. No statistically significant interaction was found between duration of hormone suppression (long-term or short-term) and any of the outcomes considered, indicating that HT benefit is consistent across all trials, regardless of treatment duration (Table 2).

According to Sharifi et al,1 the benefits of the combination of hormone suppression and radiotherapy must be “carefully weighed against substantial risks and adverse effects on quality of life.” Although no statistically significant differences in toxicity were observed (Table 3), the 31% to 34% reduction in the RR of GU and GI toxicities observed for patients receiving the combined treatment suggests that the addition of HT to RT may actually prove beneficial in a larger trial population. With regard to treatment duration, data are insufficient (4 trials) to reliably evaluate its impact on treatment tolerability; however, long-term seems to significantly increase the risk of both GU and GI morbidity for patients treated with 3-dimensional conformal RT.30 Obviously, RT dose plays a dramatic role, and the optimal hormone treatment duration to be combined with >70 grays RT is currently not established. Although the addition of short-term (< 1 year) neoadjuvant/concurrent androgen deprivation therapy to a very high radiation dose did not appear to confer a therapeutic advantage but added side effects and cost,31 escalated-dose conformal radiotherapy with neoadjuvant androgen suppression appears clinically worthwhile in terms of PFS and decreased use of salvage androgen suppression.32

The issue of toxicity is particularly important in light of recent warnings of increased mortality when adding HT to RT in prostate cancer.33-35 The Prostate Strategic Urologic Research Endeavor database suggested that neoadjuvant/adjuvant HT was associated with higher rates of cardiovascular death. Greater risk of cardiovascular death was observed in the subset of men who underwent a partial response, but not in the overall study population.36 A pooled analysis of 3 randomized controlled trials of RT with or without androgen deprivation therapy for intermediate-risk and high-risk prostate cancer also showed that HT was associated with shorter time to fatal myocardial infarction in a statistically significant fashion.6, 7, 24 However, such association was observed only in the subset of men older than 65 years.33 A competing risk analysis discriminating cardiac events (and deaths) as a function of age and HT in an individual patient data meta-analysis would actually be required to solve this issue. According to the results reported herein, no significant difference in terms of cardiac deaths was observed when comparing exclusive RT with HT + RT (Table 3); therefore, whether the suggested higher rate of cardiac events is due to HT or patient age remains unknown.

After the pivotal meta-analysis performed by Roach et al, which explored the beneficial role of adding HT in patients receiving RT and provided suggestions for the future trials,4 this is the first meta-analysis weighing the effect of the addition of HT to exclusive RT by gathering those trials specifically designed to answer this question. A recent meta-analysis has analyzed the effect of neoadjuvant HT before surgery (14 RCTs) or RT (4 RCTs).37 According to the authors, the optimal schedule and/or duration of treatment still remain a subject of further studies; however, neoadjuvant HT should use “the quickest and most potent combination for restricted time periods.”37

The conclusions in favor of the combination strategy should be softened in view of the discrepancies in trial number, sample size, treatment duration, and patients' risk profile, as well as in light of known biases of literature-based meta-analyses.38 Such caveats notwithstanding, the magnitude of benefits and evidence demonstrating the absence of major differences between individual patient data and abstracted data meta-analyses, when dealing with a similar patient sample,39 strongly indicate that the results can be considered “carefully reliable.” In the case of the present meta-analysis, this is further supported by the power analysis, demonstrating that the sample analyzed largely exceeds that required to reliably determine the observed differences in outcomes.

An exploratory metaregression analysis, with all its limitations, did not identify clinical predictors with a significant impact on benefit from HT, with the exception of lymph node positivity and Gleason score only with regard to the clinical progression-free survival outcome (Table 4). This issue is of paramount importance in determining which patients would benefit from a combined treatment strategy, and could be appropriately assessed only in the context of an individual patient data meta-analysis.

Although the results of the current study confirm and quantify the benefit of adding hormone suppression to exclusive RT for patients with locally advanced prostate cancer, a “road map” to identify which patients really benefit from such strategy and what is the optimal duration of treatment needs to be drawn. In this regard, gene and molecular profiling play an increasingly important role in the prognostic and predictive classification of prostate cancer.

Conflict of Interest Disclosures

Supported by a grant of the National Ministry of Health and the Italian Association for Cancer Research.