How the Addition of Hormones Improves Outcome: Hormone Therapy and Surgery

1. Introduction 

This paper will discuss the use of hormone therapy in conjunction with radical prostatectomy for the treatment of prostate cancer. These two treatment modalities may be combined in a variety of ways. Hormone therapy can be used in the neoadjuvant setting (NHT) where it has been investigated in both locally advanced (cT3) disease and clinically localized (cT1c–cT2) disease. It can also be used in the adjuvant setting (AHT) or as monotherapy in cases of PSA-only recurrence. One recurring question that arises with the use of hormone therapy in the AHT or PSA-only recurrence setting is whether therapy should be initiated early or delayed until further progression of disease is evident.

2. Neoadjuvant therapy with radical prostatectomy 

There are some consistent advantages to NHT that have been appreciated in studies investigating a 3-month preoperative course of therapy. All have shown a consistent 30–35% prostate gland downsizing effect, with potentially greater volume reductions with longer courses of NHT. This may facilitate surgery in cases apical cancers and/or large glands. NHT has also been shown to decrease the incidence of positive surgical margins by around 50% in patients with clinically localised disease (clinical stage T1c or T2). There is also a potential advantage in terms of the impact of NHT on micrometastatic disease although this has yet to be proven.

There are also some potential disadvantages to NHT that should be considered. Besides the well-known and common adverse effects of androgen deprivation therapy, NHT can also occasionally induce a significant desmoplastic tissue reaction. This obliteration of tissue planes may make surgery more difficult and nerve-sparing in particular may be more challenging. Although markers of cell turnover suggest otherwise, there is also the theoretical risk of tumour progression during NHT from the development of androgen resistant clones. NHT can also significantly effect the evaluation of the radical prostatectomy specimen. Histological changes induced by prolonged androgen deprivation therapy can make it difficult for the pathologist to determine the Gleason grade of the cancer.

3. NHT in patients with locally advanced disease 

There are over 20 published studies investigating 3 months of NHT prior to radical prostatectomy for clinical stage T3 disease. Downsizing of the prostate and a decrease in tumour volume are observed but pathologic downstaging is uncommon: only 20% have organ-confined disease at the time of surgery. Similar rates of positive margins are seen in patients who present with stage T3 disease, with or without NHT. Also, the rates of seminal vesicle and lymph node involvement are the same in the hormone-treated and untreated groups.

The Southwest Oncology Group (SWOG) study 9109 investigated 4 months of NHT in patients with clinical stage T3 and T4 prostate cancer [1]. Sixty-two patients received 4 months of combined androgen blockade and were followed-up for a median of 6.1 years. Organ-confined disease was found in 62% of patients and only 30% had positive margins. The 5-year progression-free survival was 70% and overall survival 90%. The results of this study suggest that a more extended course of NHT may be of benefit in patients with locally advanced disease.

4. NHT in patients with localised disease 

There are at least 7 randomised studies in the literature investigating 3 months of NHT in patients with localised disease (T1c or T2). These studies consistently showed that the rate of positive margins was reduced by about 50% in those treated with NHT. Many of the patients in these series did not reach an undetectable PSA level before radical prostatectomy. Biochemical disease-free outcome at 3 years was similar in those who did or did not receive NHT.

There are several potential reasons why the studies of 3 months NHT did not detect differences in PSA recurrence. First, the concept that androgen ablation may eliminate a sufficient number of cells may be flawed. It may also be due to flawed study design. Because these studies were designed only to detect differences in margin positive rate, not biochemical outcome, the sample size may be insufficient to detect small differences in PSA recurrence. Follow-up may also be insufficient since the number of PSA recurrence events in these studies was relatively small. Finally, the duration of androgen deprivation therapy may not have been long enough to have a significant effect. This is supported by the fact that many of these patients did not reach an undetectable PSA level before radical prostatectomy.

Longer courses of NHT prior to radical prostatectomy have only recently been investigated. Gleave and colleagues have demonstrated that many more patients achieve a nadir PSA level after 8 months of NHT than after 3 months. In a pilot study of 154 patients, with 8 months of NHT, 74% had organ-confined disease, 6% had positive margins and only 2% were node positive [2]. When the risk of PSA recurrence in this population was compared retrospectively to series incorporating 3 or 0 months of NHT, PSA recurrence free survival was significantly better in those receiving the longer (8 month) course of NHT. In another retrospective study comparing radical prostatectomy alone to radical prostatectomy with short (3 months) or longer (5 months) courses of NHT, there was a 40% reduction in the PSA recurrence rate in those receiving the longer course of NHT (Table 1) [3].

Table 1. Risk of biochemical failure in patients treated with either 3 or 5 months of neoadjuvant hormonal therapy prior to radical prostatectomy [3]

	Biochemical failure
	Treatment	Hazard ratio	95%CI
	RP alone	1.00	–
	3-month NHT	1.01	0.70–1.45
	5-month NHT	0.60	0.38–0.94

The Canadian Urologic Oncology Group (CUOG) P95A study was designed to detect differences in biochemical recurrence between 3 and 8 months of NHT. A total of 547 patients were randomised to either 3 or 8 months of leuprorelin and flutamide prior to radical prostatectomy [4]. The preliminary analysis of this investigation reported pathological endpoints and side effects only; it did not look at biochemical outcome. Radical prostatectomy was undertaken in 500 patients and the incidence of positive margins was found to be significantly lower in the 8-month versus the 3-month treatment group. However, patients in the 8-month group experienced a higher number of adverse events and hot flushes.

A recently presented analysis of the PSA recurrence outcomes from this study at 3 years showed that there was no statistically significant difference in outcome between the two treatment groups if all patients were considered together [5]. Sub-analysis of patients stratified by risk group showed that intermediate-risk patients (PSA between 10–20mg/ml; pathological stage T2b disease) seemed to benefit the most from the longer course of NHT.

One interesting finding of this study was that patients who had their surgery performed at high-volume centres did better than those at low-volume centres, regardless of whether they had received 3 or 8 months of therapy. This finding is so far unexplained.

To summarise the findings of this important study of 3 versus 8 months of NHT:

5. Adjuvant hormonal therapy after radical prostatectomy in high-risk groups 

Numerous nomograms are now available to stratify patients into groups at high and low risk of recurrence after radical prostatectomy. Patients in the Center for Prostate Disease Research (CPDR) and Cancer of the Prostate Strategic Urologic Research Endeavor (CapSURE) databases have been stratified by risk of recurrence (based on grade, stage, PSA and race) into very low, low, high and very high risk groups (Fig. 1) [6]. This risk stratification allows the identification of groups with a high likelihood of recurrence, and potentially identifies patients most likely to benefit from adjuvant therapy.

View Large Image Download to PowerPoint

Fig. 1. Disease-free survival in patients stratified by risk of disease recurrence [6]. Reproduced with permission of Journal of Urology.

However, there are limited randomised, prospective trials in the literature exploring adjuvant hormonal therapy (LHRH or antiandrogens) after radical prostatectomy in high-risk patients.

Several retrospective studies from the Mayo Clinic report improved outcome in subsets of high-risk patients who receive early hormonal therapy after radical prostatectomy [7]. In patients with node positive (D1) disease, early AHT improved cancer-specific survival in DNA diploid tumors after 10 years. Adjuvant hormonal therapy improved progression-free survival in patients undergoing radical prostatectomy for locally advanced (clinical stage T3) cancers although overall and cancer-specific death was not affected. In a recent report, patients with seminal vesicle involvement also appeared to benefit from AHT. In those receiving hormonal therapy within 90 days of surgery, cancer specific survival at 10 years was 95% versus 87% in those who did not receive this early therapy.

The bicalutamide Early Prostate Cancer trial evaluated the use of early anti-androgen monotherapy after primary therapy in 8000 men [8]. Although there were some differences in the protocols used in each region, results were pooled from the Scandinavian, North American and European arms of the study. In this placebo controlled, double-blind trial patients were randomised to receive either bicalutamide 150mg or placebo in addition to standard care. The results showed that in the bicalutamide treated group the risk of disease progression was reduced by 42%. In the pooled analysis, this advantage was seen for all forms of standard care when combined with bicalutamide: radiotherapy, radical prostatectomy or watchful waiting. When the results were stratified by cancer stage (localised or locally advanced), and other markers indicative of disease aggressiveness, (PSA or grade), patients with more aggressive disease seemed to have the greatest benefit of adjuvant therapy, regardless of primary therapy.

More recently, 5-year follow-up data have become available from the EPC study. These data have demonstrated that in the subgroup of patients with localised prostate cancer who would otherwise have been managed only by watchful waiting, there was a trend for an increase in the number of deaths in patients receiving bicalutamide 150mg when compared with patients who received placebo [196 (25.2%) deaths versus 174 (20.5%) deaths, hazard ratio (HR)=1.23, 95% confidence interval 1.00–1.50] [unpublished data]. As a result, the Health Authorities in Canada have withdrawn the approval for the use of bicalutamide 150mg for the treatment of patients with localised disease. A similar withdrawal of approval has also been enforced in several European countries. The use of bicalutamide 150mg in locally advanced prostate cancer is not affected by these new data.

6. Timing of hormonal therapy for rising PSA after radical prostatectomy 

There is some uncertainty about the most appropriate PSA cut-point to define recurrence after radical prostatectomy. In one study, a PSA of 0.4ng/ml was proposed as the most appropriate level at which to begin to consider treatment [9]. There are several potential explanations why low detectable but stable PSA values after radical prostatectomy do not indicate disease recurrence. Often, normal prostate tissue is found at the surgical margins, especially at the apex, and benign prostate tissue is found in 15% of prostatic fossa/vesicourethral biopsies after radical prosatectomy. In addition, there are detectable urinary levels of PSA in 77% of patients after radical prosatectomy and ectopic prostatic (PSA producing) tissue has been found in urethra, bladder, spleen and perivesical spaces.

The natural history of prostate cancer progression after biochemical recurrence can be varied. A study by Pound et al implied a relatively slow natural history of disease progression in this setting [10]. For PSA recurrence after radical prostatectomy it was generally about 8 years to develop clinical metastases with no treatment and then another 5 years until death from prostate cancer. PSA kinetics (timing of recurrence and PSA doubling times) may help to identify subsets of men with biochemical recurrence who are more likely to develop systemic progression of disease.

In the Pound study, patients with higher Gleason scores had a shorter time to development of distant metastases, especially if time to development of biochemical recurrence was shorter (<2 years). The same pattern was seen for those with a low Gleason score.

Unfortunately, there are no randomised, prospective studies yet available which address the timing of hormonal therapy in patients with biochemical recurrence. However, if we believe that the majority of men with PSA-only recurrence actually have occult stage D1 disease, then early traditional hormonal therapy may provide a disease-specific survival advantage. This is supported by the results of the Messing et al study of immediate hormone therapy versus observation in node-positive prostate cancer after radical prostatectomy [11]. Patients receiving immediate hormonal therapy had better disease-free and overall survival than those receiving deferred treatment.

7. Conclusion 

While NHT can significantly decrease the incidence of positive margins at the time of radical prostatectomy, 3 months of treatment is not long enough to have any significant effect on biochemical recurrence rates. The results of studies investigating longer courses (8 months) of NHT are anxiously awaited. High-risk patients should be considered for early AHT after surgery as they may be most likely to benefit. There may also be an advantage to adding chemotherapy to the treatment regimen in these high-risk patients and further investigation of multimodal treatment in both the adjuvant and neoadjuvant setting is ongoing. PSA recurrence after radical prostatectomy has a varied natural history but it is possible to identify patients at high risk of disease progression who may benefit most from early hormonal therapy.