Definitive Therapy for Localized Prostate Cancer: An Overview

William J. Catalona, MD, Misop Han, MD, MS




The purpose of this chapter is to provide an overview of the definitive treatment of clinically localized prostate cancer. In reporting differences in treatments, we have attempted to be objective but have included our editorial perspective. For balance, we have included in our suggested reading list several recent articles focusing on the relative advantages and disadvantages of the different management strategies, some expressing opinions contrary to ours. We have grouped treatments into established treatments, such as radical prostatectomy and radiotherapy, for which abundant published information is available, and other treatments, including active surveillance, primary hormone therapy, cryoablation, radiofrequency ablation, and high-intensity focused ultrasound (HIFU), for which data are more limited for localized disease.



Background


Prostate cancer is the most common nonskin cancer and the second-leading cause of death from cancer in men in the United States (Jemal et al, 2009). In 2009, it was estimated that more than 192,280 new cases would occur; for more than 27,000, it was predicted to be the cause of death (Jemal et al, 2009). Because prostate cancer is prevalent in many countries and exhibits a wide spectrum of aggressiveness, different methods of treatment have been developed, and the preferred methods for detection and treatment are controversial. The prevalence of prostate cancer increases strikingly with age. Autopsy studies have documented microscopic foci of prostate cancer in about one fourth to one third of men in the fourth and fifth decades of life and in more than three fourths in the ninth decade (Sakr et al, 1993; Yin et al, 2008). Yet, a disproportionately lower but still substantial number of men (about one in six) are diagnosed with prostate cancer during their lifetime (Jemal et al, 2009). Because of effective treatment of some prostate cancers and the biologic indolence relative to life expectancy of others, only about 16% of men diagnosed with prostate cancer ultimately die of it. Prostate cancer is the cause of death in about 3% of the U.S. male population (Jemal et al, 2009). An additional (unquantified) proportion suffers from prostate cancer but dies of other causes.


The marked disparity between prevalence and incidence rates of prostate cancer, on the one hand, and morbidity and mortality rates, on the other, has led some to conclude that many prostate cancers are harmless and perhaps would better be left undetected. Nevertheless, if the present trends of increasing life expectancy continue, given the current age-specific incidence, morbidity, and mortality rates of prostate cancer, this disease will become a far greater public health problem in the future.


Since the 1980s, the methods of diagnosis of clinically localized prostate cancer have changed. Widespread screening with serum prostate-specific antigen (PSA) and digital rectal examination (DRE) has allowed earlier detection (Catalona et al, 1991, 1993). Furthermore, with a remarkable stage migration, approximately 90% of cases are being detected in a clinically localized stage, whereas metastases at the time of diagnosis are now rare in the United States (Han et al, 2001a). The natural history of prostate cancer varies from indolent disease that might not cause symptoms during a patient’s lifetime to highly aggressive cancer that metastasizes quickly and causes terrible suffering and untimely death. The challenge for the physician who treats patients with prostate cancer is to advise effective treatment in those for whom treatment is necessary. Selection of the appropriate treatment requires assessment of the tumor’s potential aggressiveness and the general health, life expectancy, and quality of life preferences of the patient.


Patients whose tumor has a low malignant potential are predetermined to fare better with most treatments. Therefore the treatment outcomes in any patient series may be influenced by the malignant potential of the tumors as well as by the treatment used. Accordingly, it is difficult to compare the results of different reports because the populations of patients are usually heterogeneous and not strictly comparable. Furthermore, outcome measurements are not necessarily comparable between different forms of therapy (e.g., different definitions of biochemical progression for surgery and radiotherapy), confounding comparisons between them.



Characterization of the Primary Tumor


DRE and prostate ultrasound findings usually provide useful information about the extent of the primary tumor. The serum PSA data, including the total PSA level, the rate of change of PSA (PSA velocity and doubling time), the PSA density (serum PSA divided by prostate volume), and the percentage of PSA in the free or complexed isoforms, are significantly associated with prostate cancer aggressiveness (Benson et al, 1992; Carter et al, 1997; Catalona et al, 1998; D’Amico et al, 2004; Thompson et al, 2004; Kundu et al, 2007). The biopsy findings—Gleason score; the number of cores containing cancer; the distribution and volume of cancer in the biopsy cores; the presence of perineural space invasion, lymphovascular invasion, or ductal or neuroendocrine differentiation—also correlate with cancer aggressiveness and the likelihood of the cancer being organ confined. Prediction tables and algorithms have been developed to assist in this assessment (Partin et al, 1997, 2001). However, such statistical aids are more useful in groups of patients than in individual patients, and wide confidence intervals surrounding estimates of outcomes sometimes limit the usefulness of risk assessment for an individual patient. In this regard, it has been claimed that the simultaneous assessment of multiple variables in nomograms provides more accurate predictions than tables do for individual patients (Kattan, 2003).




Conservative Management



Active Surveillance or Watchful Waiting


Active surveillance and watchful waiting are almost unique to prostate cancer. There are different concepts of deferred treatment of prostate cancer. Watchful waiting refers to monitoring the patient until he develops metastases that require palliative treatment. Active surveillance or expectant management allows delayed primary treatment if there is biochemical or histologic evidence of cancer progression (Dall’Era et al, 2008). Active surveillance is a less established strategy in patients with a long life expectancy, because criteria for selecting candidates and trigger points for instituting treatment have yet to be defined and validated. Currently, treatment is frequently initiated because of the patient’s anxiety from living with untreated cancer combined with a rising PSA level or biopsy findings that suggest an increase in the volume or Gleason grade of the cancer. Findings on DRE or transrectal ultrasonography are seldom the sole indications for intervention.


Traditionally, deferred treatment has been reserved for men with a life expectancy of less than 10 years and a low-grade (Gleason score 2 to 5) prostate cancer. However, active surveillance is now being evaluated as a management strategy in younger patients with low-volume, low- or intermediate-grade (up to Gleason score 3 + 4 = 7) tumors to avoid or to delay treatment that might not be immediately necessary. In one study, approximately 16% of patients with newly diagnosed prostate cancer would fulfill the criteria for active surveillance, about 10% chose surveillance, and an additional 4%, who did not meet all criteria, chose active surveillance (Barocas et al, 2008).


Statistical models have been generated in an attempt to predict which tumors can be observed without aggressive treatment. For example, Epstein and associates (1994, 1998) proposed a model involving preoperative clinical and pathologic features that would predict “insignificant tumors” (tumor volume less than 0.2 cc, Gleason score below 7, and organ-confined cancer). The preoperative features used in the model include no Gleason pattern 4 or 5 in the biopsy specimen, PSA density of 0.1 ng/mL/g or less, fewer than three biopsy cores involved (with a minimum of six total cores being obtained), and no core with more than 50% involvement or PSA density of 0.1 to 0.15 ng/mL/g and cancer smaller than 3 mm on only one prostate biopsy sample. Characteristically for statistical models, this model was reported to have a predictive value of 95% for identifying a “significant” cancer but a predictive value of only 66% for identifying an “insignificant” cancer (Epstein et al, 1994). Approximately 16% of the men in this series met criteria for an insignificant cancer (Epstein et al, 1994). Subsequently, Epstein and colleagues (1998) updated the model to include a free/total PSA ratio (≥0.15) and favorable needle biopsy findings (fewer than three cores involved, no core with more than 50% tumor, and Gleason score of 6 or lower).


Kattan and associates (2003) proposed another statistical model to predict small, moderately differentiated, organ-confined cancer on the basis of PSA, clinical stage, biopsy Gleason score, ultrasound-determined prostate volume, and variables derived from systematic biopsies. They defined indolent cancer as being organ confined, having less than 0.5 cc tumor volume, and without poorly differentiated elements. Approximately 20% of their patients treated with radical prostatectomy met the criteria for indolent tumors according to their prediction model.


Some authors have claimed that even patients who do not fulfill such criteria may be legitimate candidates for active surveillance (Eastham et al, 2003). A potential untoward consequence of recommending active surveillance for all men without obviously aggressive, clinically localized disease is that only men with clearly aggressive and often incurable disease would be treated immediately, whereas a substantial proportion of those with curable disease destined to progress would be monitored, often with multiple extended biopsy procedures that could complicate subsequent attempts at nerve-sparing surgery or delay treatment until the window of opportunity for cure has closed.


All prostate cancer patients are at risk for progression. In reports of deferred therapy for prostate cancer, patients are usually observed with semiannual PSA determinations, DRE, and annual biopsies (Zietman et al, 2001; Choo et al, 2002; el-Geneidy et al, 2004; Klotz, 2004; Patel et al, 2004; Carter et al, 2007; Dall’Era et al, 2008). Intervention is recommended if Gleason pattern 4 or 5 is present, more than two biopsy cores are involved, or more than 50% of a biopsy core is involved. Progression is more likely in patients who have cancer present on every biopsy procedure. The absence of cancer on repeated biopsy significantly decreases the likelihood of progression (Carter et al, 2007). In this regard, biopsy criteria have been reported to be more accurate than PSA criteria in predicting progression. No study has found DRE or imaging studies to independently predict progression.


The percentage of patients with curable cancer at the time of progression has been reported to vary from 33% to 92%. In most studies of active surveillance, approximately 25% to 50% of patients, depending on their individual risk factors, develop objective evidence of tumor progression within 5 years (Neulander et al, 2000; Patel et al, 2004; Warlick et al, 2006). Carter and colleagues (2007) reported that 59% remained on surveillance, 25% underwent curative treatment, and 16% either withdrew, were lost to follow-up, or died of other causes. Although some studies suggest that patients with Gleason score 2 to 4 tumors do not suffer or die of prostate cancer with conservative management, those with higher Gleason score tumors have a substantial risk for morbidity and mortality (Albertsen et al, 1995, 2005; Johansson et al, 2004). Klotz and colleagues (2006) reported that of patients who underwent radical prostatectomy for evidence of cancer progression during active surveillance, 58% had tumor extension beyond the prostate and 8% had lymph node metastases.


A prospective, randomized clinical trial from Scandinavia reported that patients with clinically localized prostate cancer managed with watchful waiting have significantly higher rates of local cancer progression, metastases, and death from prostate cancer and a shorter cancer-specific (lower death rate from prostate cancer) and overall survival than those treated initially with radical prostatectomy (Bill-Axelson et al, 2008). Similarly, an observational study of Medicare patients treated with observation, radiation, or surgery showed a survival advantage for active treatment of men aged 65 to 80 years; however, the absolute difference in cancer-specific death at 12 years was small (Wong et al, 2006).


One rationale for active surveillance is the belief that there is substantial overdiagnosis of prostate cancer as a result of widespread PSA screening coupled with aggressive biopsy regimens. Overdiagnosis often refers to a cancer detected by screening that would not be detected during the patient’s lifetime without screening or would never cause disability or death. It is axiomatic that any effort to detect cancer early will involve detection of some cancers that would not have been otherwise detected. Therefore some overdiagnosis is necessary to reduce suffering and death from prostate cancer.


There are two methods of estimating the extent of overdiagnosis. The epidemiologic method applies only to populations, not to individuals. With this method, statisticians look at population trends in prostate cancer cases and use statistical models to estimate whether more cases are being diagnosed than should be, considering past incidence rates of prostate cancer and prostate cancer deaths. Statisticians estimate lead time in prostate cancer diagnosis with PSA screening (3 to 12 years) and use the lead time to estimate overdiagnosis; however, statistical models have not accurately predicted the observed incidence of prostate cancer nor fully explained the observed decrease in advanced disease.


The second method of estimating overdiagnosis is for a pathologist to examine a surgically removed cancerous prostate gland and determine whether it contained only a tiny amount of cancer with no high Gleason pattern glands and that was completely encapsulated within the prostate gland. If so, they call it an overdiagnosed cancer.


Some reports have estimated that 50% or more of prostate cancer cases are overdiagnosed (Etzioni et al, 2002; Draisma et al, 2003). However, recent studies suggest that epidemiologic estimates of overdiagnosis are exaggerated. Epidemiologic estimates based upon both statistical models and data from the United States yield a 23% to 28% incidence of possible overdiagnosis (Draisma et al, 2009). Estimates based upon clinicopathologic data range from 6% to 20% (Graif et al, 2007; Pelzer et al, 2007).


Estimates of overdiagnosis derived from older men should not be generalized to younger men. Prostate cancers diagnosed in younger men are less likely to be harmless in the long term, and it is uncertain whether all cases labeled as overdiagnosed are clinically insignificant. Contrary evidence suggests that screening with low PSA thresholds for biopsy in young patients detects tumors that fulfill the criteria for insignificant cancer in only 12% of patients (Krumholtz et al, 2002). Even among those that do, some tumors are multifocal or do not have a diploid complement of chromosomes. Presently, no tumor marker can identify indolent tumors with certainty.


An accumulating body of increasingly compelling evidence shows that PSA screening can reduce prostate cancer deaths. First, there would be no point in trying to detect prostate cancer early if there were no effective treatment for early stage disease. The study on the long-term results of a Swedish randomized trial of radical prostatectomy versus watchful waiting for early prostate cancer showed that surgery can reduce the rates of metastases and prostate cancer deaths, and even increase overall survival (Bill-Axelson et al, 2011).


The “holy grail” of screening is the randomized trial. In the initial report from the European Randomized Study of Screening for Prostate Cancer (ERSPC), men randomized to screening had a 20% lower prostate cancer death rate. Not surprisingly, this mortality benefit was observed largely in men under age 70 (Schroder et al, 2009).


The initial report of the ERSPC contained the sobering estimate that 48 men had to be treated to prevent one prostate cancer death. This high number-needed-to-treat (NNT) raised many eyebrows, because it is shocking to think that 48 men would have to undergo surgery or radiation to prevent one man from dying of prostate cancer. However, because the risk of dying from prostate cancer is influenced by a patient’s life expectancy, the NNT to save one life is also strongly influenced by a patient’s age and health, and the length of follow-up of the trial. The NNT is lower in younger, healthier men and in trials that have longer follow-up. Subsequent studies have reported a far lower NNT to save one life.


The mortality curves of the screened and control groups from the ERSPC began to separate at 9 years of follow-up when the trial was first reported. By 12 years follow-up, the NNT had decreased to 12 (Schröder FH, presented at the 2011 American Urological Association Meeting in Washington, DC).


In the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO) (Andriole et al, 2009), the results were initially reported on the entire study group—young and old, healthy, and not so healthy. They concluded, “…we now know that prostate cancer screening provided no reduction in death rates at 7 years… .” Many authors summarized these two large screening trials as yielding conflicting results; however, this conclusion is not correct. From a subsequent data analysis of the healthier participants, a far different picture emerged. A follow-up analysis of PLCO data showed a striking mortality benefit in the roughly one third of PLCO participants with minimal or no co-morbidities (Crawford et al, 2010). This analysis showed the prostate cancer death rates of relatively healthy PLCO participants randomized to screening were 44% lower than in the control arm, and the NNT to save one life in these healthier men was reported to be only 5.


The Norrkoping Trial from Sweden (Sandblom et al, 2011) also was widely publicized as a negative screening trial. This trial has many limitations. First, it is a small study with only 85 screen-detected prostate cancer cases. Men were screened every 3 years, but the first two screens involved only digital rectal examination. There was no PSA screening until 6 and 9 years into the trial. Fine-needle aspiration cytology (rather than core needle biopsy) was used for diagnosing prostate cancer. The median age of the patients at diagnosis was nearly 70 years old, and the median follow-up of prostate cancer patients was only 6 years. Moreover, nearly half of the screen-detected patients were managed with watchful waiting, and only one third were treated with radical prostatectomy. Despite these limitations, the data actually provides considerable support for PSA screening. The screened patients had far fewer advanced tumors at diagnosis. There was a strong trend for better prostate cancer–specific survival in the screening arm that approached statistical significance, and there was a significantly lower risk ratio for prostate cancer death in the screening arm.


The Gotebörg Screening Trial used perhaps the best methodology (Hugosson et al, 2010). It is a population-based study of younger patients—20,000 men aged 50 to 64. The contamination rate (i.e., opportunistic PSA screening of controls) was only 3% when the trial started versus an estimated 15% contamination rate in ERSPC and 40% to 50% in PLCO. Trial participants were screened every 2 years with progressively lower PSA cutoffs over time, beginning with a 3.4 ng/mL cutoff and ending with a cutoff of 2.5 ng/mL. Ninety-three percent of the men complied with the biopsy recommendation when they had an abnormal screening test, in contrast to PLCO, in which only 40% underwent prompt biopsies. Seventy-seven percent had 14 years of follow-up. The results showed that 41% fewer screened patients presented with advanced disease, and the screening arm had a 44% lower prostate cancer death rate. Moreover, the NNT was 12, which compares favorably with the NNT for breast cancer (NNT = 10 at 10 years of follow-up). The Gotebörg trial has not yet demonstrated a significant improvement in overall survival; however, overall survival benefit takes the longest time to demonstrate of any survival parameter.


Pound and colleagues (1999) reported that the median time from PSA failure to the development of metastases after radical prostatectomy is 8 years, and the median time from metastases to death is 5 years (Pound et al, 1999). Thus the median time from diagnosis to death is more than 13 years. The median follow-up time for patients in U.S. screening trial was only 5.2 years for controls and 6.3 years for cases—2 years before the expected metastases and 6 years before prostate cancer death. In the European trial, there also was no improvement in survival until after 7 years.


The United States trial has been portrayed as the underpinning for a proposed shift in the mind-set of physicians and patients that screening and treatment of prostate cancer do more harm than good. However, this trial was flawed at the beginning, is a mere snapshot taken halfway through the relevant follow-up period, and will never be informative of the true impact of screening on prostate cancer death of healthy men managed with intelligent screening, prompt biopsy, and effective treatment.


There is other compelling evidence to suggest that early diagnosis through PSA testing and prompt, effective, and high-quality treatment saves lives. This evidence comes largely from the cancer registries of the United States and World Health Organization databases. The percentage of men who have advanced prostate cancer at the time of diagnosis has decreased by 75% in the United States during the PSA era, and there has been a 40% decrease in the prostate cancer death rate during the PSA screening era (National Cancer Institute, 2010). Statistical studies suggest that 45% to 70% of this decrease is due to PSA screening (Etzioni et al, 2008).


Other studies have shown that in regions of the United States where more PSA testing is performed, there is a lower rate of advanced prostate cancer and prostate cancer death rates (Colli and Amling, 2008). In a population-based study, PSA screening reduced the prostate cancer-specific mortality rate by 62% (Agalliu et al, 2007; Kvale et al, 2007). Globally, the death rate also has fallen in countries where PSA testing is practiced, while it continues to rise in countries where it is not (Bouchardy et al, 2008).


The frequently cited decrease in prostate cancer mortality reported in the United Kingdom from 1992 to 2004 in the absence of widespread PSA screening is due largely to the method of attributing the cause of death in the United Kingdom databases during that time. Before the PSA era, if a man with metastatic prostate cancer died of pneumonia, the cause of death was attributed to prostate cancer, but during the early years of the PSA era, the cause of death was attributed to pneumonia. Thus there was a spurious decrease in the prostate cancer mortality rate in the PSA era in the absence of widespread PSA screening (Hussain et al, 2008).


The physician confronting the patient with newly diagnosed prostate cancer must decide on management based on the PSA level, the estimated tumor volume, and the Gleason score (up to one third of patients are upgraded on the basis of the radical prostatectomy specimen (Pinthus et al, 2006) to select patients for immediate treatment or active surveillance. Repeated biopsies are always subject to sampling errors (Harnden et al, 2008) and may induce fibrosis in and around the prostate gland that could compromise subsequent nerve-sparing surgery, rendering it impossible to perform in more than half of patients (Barzell and Melamed, 2007), as well as trigger inflammation leading to PSA fluctuations that are difficult to interpret.


Treatment is more likely to be successful if given earlier while the tumor is smaller and the prospects for potency-sparing surgery are greater. Deferred treatment is more appropriate for older patients with a limited life expectancy or comorbidities. Additional clinical and laboratory research are needed to define the parameters for safe use of active surveillance in younger men, including the appropriate selection criteria, follow-up procedures, and trigger points for intervention (Carter et al, 2003; Allaf and Carter, 2004; Wilt, 2008). It will also be necessary to determine the proportion of patients that would still have curable disease when they are treated at the time of objective disease progression. In many instances, active surveillance delays the treatment by only a few years; however, Freedland and colleagues (2006) reported that delays of more than 6 months conferred a 2.73-fold increased risk of progression in patients with low-risk prostate cancer. Active surveillance frequently amounts to delayed treatment, and patients selected for active surveillance have cancers that are most curable with the fewest side effects. Some with curable disease would have surveillance until the window of opportunity for cure is closed, and it would be a mistake to treat only patients with incurable disease. For the present, patients who opt for active surveillance should be evaluated with DRE and PSA testing quarterly or semiannually, and they should consider having repeated prostate biopsy procedures yearly or biennially. In patients with a consistent PSA velocity of greater than 0.35 ng/mL/year, there is a fivefold increased risk of prostate cancer death in the next two to three decades (Carter et al, 2006). Although it is assumed that quality of life should be largely preserved with active surveillance, studies have demonstrated significant decrements in quality of life with time, including waning erectile function, diminished urinary continence, and adverse psychologic effects from living with untreated cancer (Penson et al, 2005). For example, in the Scandinavian study, men randomized to watchful waiting had a significantly worse quality of life than men randomized to radical prostatectomy (Johansson et al, 2008). This was especially true in men who received androgen deprivation (Johansson et al, 2008). If the PSA level is rising, the DRE suggests tumor growth, or surveillance biopsy specimens show evidence of increased involvement by cancer, treatment should be instituted. Patients may change their minds about remaining on an active surveillance protocol; therefore the physician should review management options on follow-up visits.


In patients with a long life expectancy, there is a certain risk associated with active surveillance. Clearly, it can avoid or delay treatment for some patients, but there will inevitably be those who will miss their opportunity for cure and, tragically, ultimately progress to metastases and death from prostate cancer. Favorable outcomes have been reported in potential candidates for active surveillance treated with prompt nerve-sparing radical prostatectomy (Loeb et al, 2008).


The ERSPC trial has provided an unequivocally affirmative answer to half of the often quoted aphorism of Willett Whitmore Jr. concerning prostate cancer—“Is cure possible when it is necessary?” To answer the second part, “Is cure necessary when it is possible?,” will require more clinical research evaluating the trade-offs.



Radical Prostatectomy


Radical prostatectomy was the first treatment used for prostate cancer and has been performed for more than 100 years (Kuchler, 1866; Young, 1905). It is a technically formidable operation, and as a result, simpler treatments have been sought for the treatment of early-stage disease. However, no treatment has supplanted radical prostatectomy, and it still remains the gold standard because of the realization that hormone therapy and chemotherapy are never curative, and not all cancer cells can be eradicated consistently by radiation or other physical forms of energy, even if the tumor is contained within the prostate gland.



Innovations have led to the wider use of radical prostatectomy:





In recent years, the laparoscopic approach has been developed for performing the operation.


The main advantage of radical prostatectomy is that when skillfully performed, it offers the possibility of cure with minimal collateral damage to surrounding tissues (Han et al, 2001b; Hull et al, 2002). Further, it provides more accurate tumor staging by pathologic examination of the surgical specimen. Also, treatment failure is more readily identified, and the postoperative course is much smoother than in the past. Few patients require nonautologous blood transfusions. The hospital stay is usually 1 to 3 days, and operative mortality is rare in the modern era. Moreover, radical prostatectomy significantly reduces local progression and distant metastases and improves cancer-specific and overall survival rates compared with watchful waiting (Bill-Axelson et al, 2008). Some patients with tumor recurrence after radical prostatectomy can be salvaged with potentially curative postoperative radiotherapy (Stephenson et al, 2004b; Trock et al, 2008).


The potential disadvantages of radical prostatectomy are the necessary hospitalization and recovery period; a possibility of incomplete tumor resection, if the operation is not performed properly or if the tumor is not contained within the prostate gland; and a risk for erectile dysfunction and urinary incontinence. However, erectile dysfunction and rectal complications are less likely with nerve-sparing surgery than with radiotherapy, and good treatment options are available for both urinary incontinence and erectile dysfunction. Results reported from high-volume centers are more favorable than those from national surveys (Rabbani et al, 2000; Stanford et al, 2000, Kundu et al, 2004; Sanda et al, 2008).



Surgical Approaches to Radical Prostatectomy




Perineal


Total perineal prostateactomy is an acceptable surgical treatment when performed by a surgeon familiar with this approach (Scolieri and Resnick, 2001). It is usually associated with less blood loss and a shorter operative time than the retropubic approach. The disadvantages are that it does not provide access for a pelvic lymph node dissection, there is a higher rate of rectal injury, and there is occasional postoperative fecal incontinence that does not occur commonly with other approaches (Bishoff et al, 1998). It is more difficult to spare the cavernous nerves through the perineal approach.




Laparoscopic


The laparoscopic approach is the most daunting method of performing radical prostatectomy. It has been suggested that laparoscopic prostatectomy may be associated with less bleeding, better visualization, less postoperative pain, and shorter convalescence than the standard open approach. However, the claimed advantages are not always realized with many surgeons and are usually less important to the patient than the cure of his cancer, the preservation of his potency, and the avoidance of other more serious complications. Laparoscopic prostatectomy can be performed through a transperitoneal or extraperitoneal approach, but the extraperitoneal approach poses logistical limitations, especially with the use of robotic assistance. The transperitoneal approach facilitates the lymphadenectomy but carries a higher risk of intestinal and vascular injury, urinary ascites, and postoperative intestinal obstruction.


Furthermore, laparoscopic prostatectomy is associated with a higher risk for severe complications. Hemostasis in the neurovascular bundles is difficult to achieve without thermal injury to the neurovascular bundles because of the relative difficulty in rapidly placing hemostatic sutures or applying hemostatic clips laparoscopically. Heat from a harmonic scalpel or electrocautery can irreversibly damage the cavernous nerves. Although intraoperative blood loss is less with laparoscopic surgery, postoperative bleeding may occur after the release of positive pressure in the operative field. Rectal, ureteral, and vascular injuries and anastomotic leaks have also been more common with laparoscopic prostatectomy (Rassweiler et al, 2003).


When laparoscopic prostatectomy is performed by a skilled laparoscopic surgeon, reported incontinence and anastomotic stricture rates are comparable to those achieved with open surgery. It has been claimed that nerve sparing is equivalent or even better with laparoscopic surgery, but direct comparisons and validated results are lacking (Menon et al, 2007). The early reported rates of positive surgical margins have been higher with laparoscopic prostatectomy, and the adequacy of cancer control is as yet uncertain because of lack of long-term results.



Robotic


Remotely controlled laparoscopic surgery has recently been popularized because of its greater technical ease for the surgeon, especially for tying sutures and performing the vesicourethral anastomosis. It has been aggressively marketed as a less invasive, technologically more advanced method of performing the operation with less pain and a quicker recovery. The availability of three-dimensional visualization is an advantage over standard laparoscopic techniques. Early reported results are favorable but have not been validated (Menon et al, 2002, 2007). It is uncertain whether the results of laparoscopic prostatectomy are as good as those achieved with open prostatectomy (Smith, 2004; Webster et al, 2005; Hu et al, 2006; Rogers et al, 2006; Touijer and Guillonneau, 2006).


There has been a dramatic shift from open to robotic prostatectomy in the United States that has been driven largely by marketing. However, until recently, there has been little objective data to substantiate these claims (Smith et al, 2007; Weizer et al, 2007; Wood et al, 2007). Recent reports based on comparative data have shown disadvantages in laparoscopic and robotic prostatectomy in tumor recurrence, urinary incontinence, and patient satisfaction.


In performing a radical prostatectomy, the surgeon must dissect in the proper tissue plane to remove the prostate from between the neurovascular bundles without permanently damaging the nerves or cutting into the prostate, or worse, leaving part of the prostate behind.


Most prostate cancer patients have similar priorities. First, they want to survive; second, they want to remain continent; and third, they want to preserve their potency. These are their main priorities, but they want all three. This constellation of favorable outcomes is known as the “trifecta” (Eastham et al, 2008), and, understandably, patients want to achieve it as quickly and painlessly as possible. Often, patients with newly diagnosed prostate cancer would like to avoid any treatment, if possible, or, after realizing that treatment is necessary, seek the easiest way out. This frequently leads to consideration of options such as active surveillance, cryoablation, high-frequency ultrasound, or minimally invasive prostatectomy.


It is important to examine the claims concerning laparoscopic and robotic prostatectomy. Usually six incisions are made in laparoscopic surgery, and, in most cases, the procedure is performed within the peritoneal cavity, while for open surgery, one incision is made—4 to 5 inches in length and that does not enter the peritoneal cavity. Comparative studies have shown that open prostatectomy has a similar recovery time and return to normal activity (Weizer et al, 2007; Wood et al, 2007).


It is also more difficult to perform a lymph node dissection with laparoscopic approaches. Other factors come into play as well, such as lack of a sense of touch or tactile feedback. Furthermore, some surgeries that start out as laparoscopic must be converted to open, because of failure to progress or complications that cannot be handled laparoscopically. Perhaps the most important consideration is that the laparoscopic approaches have no long-term track record of cancer control. If small amounts of cancer are left behind, it may not become apparent for years.


A comparison of a sample of patients from the Medicare database who underwent minimally invasive or open prostatectomy between 2003 and 2005 revealed a lower rate of medical complications with the laparoscopic approaches; however, the men undergoing minimally invasive prostatectomy had more than a threefold higher rate of requiring salvage therapy for tumor recurrence within 6 months of surgery (Hu et al, 2008). In this study, the more experienced surgeons in minimally invasive radical prostatectomy had better results than less experienced surgeons; however, even the highest-volume laparoscopic surgeons had twice the rate of patients requiring salvage treatment for cancer recurrence when compared with all surgeons performing open radical prostatectomy. Patients undergoing minimally invasive surgery also were 40% more likely to develop anastomotic strictures.


Another study comparing the results of laparoscopic and open radical prostatectomy revealed that the laparoscopic approach was associated with less blood loss, but, of more concern, had a higher rate of postoperative emergency room visits, readmissions to the hospital, and further surgery for complications (Touijer et al, 2008). Also, patients who underwent a laparoscopic radical prostatectomy were less likely to become continent than those treated with open prostatectomy.


In a few studies including a small number of patients, potency rates have been reported to be similar with both techniques when experienced surgeons performed the operations (Menon et al, 2007). However, in general, potency rates have not matched the best rates achieved by the most experienced surgeons performing open surgery.


Schroeck and colleagues (2008) compared patient satisfaction and regret following radical prostatectomy. Patients who underwent a robotic prostatectomy were more than four times more likely to regret their decision. The authors suggested that these patients were more likely to be regretful and dissatisfied because of the higher expectations from an “innovative” procedure. These results raise concerns that patients are being misled about the true risks and benefits of minimally invasive procedures to treat prostate cancer.


Thus the outcomes of cancer control, potency, and continence are well documented for open prostatectomy, but the jury is still out with laparoscopic approaches.



Salvage Radical Prostatectomy


Radical prostatectomy can be performed in patients for whom other local treatments have failed (Pontes, 1994; Chen and Wood, 2003). However, the rate of complications is far higher, and the complications are more serious and much more difficult to manage (Stephenson et al, 2004a; Sanderson et al, 2006). Moreover, the prospects for long-term disease-free survival are more limited for salvage prostatectomy than for primary radical prostatectomy.


Most of the reported experience with salvage radical prostatectomy is from the pre-PSA era. Contemporary series of patients selected because of biochemical recurrence have lower morbidity and better cancer control rates (Stephenson et al, 2004a; Ward et al, 2005). Nevertheless, postoperative incontinence rates are as high as 44% and bladder neck contracture rates as high as 22% (Ward et al, 2005). The incontinence rate is even higher after brachytherapy, presumably because of the higher dose of radiation administered. Long-term progression-free survival rates after salvage prostatectomy in the absence of hormone therapy have not been well documented.



Selection of Patients for Radical Prostatectomy


An ideal candidate for radical prostatectomy is healthy and free of comorbidities that might make the operation unacceptably risky. He should have a life expectancy of at least 10 years, and his tumor should be deemed to be biologically significant and completely resectable. The generally accepted upper age limit for radical prostatectomy is about 75 years.


Because imaging studies are not accurate for staging prostate cancer, preoperative clinical and pathologic parameters are often used to predict the pathologic stage and thus identify patients most likely to benefit from the operation (Partin et al, 1997, 2001). These parameters are frequently used in tables and nomograms designed to predict pathologic tumor stage or post-treatment recurrence-free survival probabilities (Han et al, 2003; Kattan et al, 1998, 2000; Ross et al, 2001). New methods of predicting the outcome after radical prostatectomy that incorporate cellular and biologic features to improve accuracy have been reported (Donovan et al, 2009).


Patients with a low probability of resectable disease or a short life expectancy should not be advised to have surgery. Neoadjuvant hormone therapy does not enhance the resectability of prostate cancer and often increases the difficulty of performing nerve-sparing surgery (Soloway et al, 2002). Similarly, neoadjuvant chemotherapy rarely produces pathologically complete responses (Chi et al, 2008).


The surgeon should realistically counsel the patient on the nerve-sparing aspects of the operation. Nerve-sparing prostatectomy does not materially compromise cancer control in appropriately selected patients; however, it is inappropriate in men with advanced disease. The feasibility of performing nerve-sparing surgery is questionable when there is extensive cancer in the biopsy specimens, palpable extraprostatic tumor extension, serum PSA level above 10 ng/mL, biopsy Gleason score higher than 7, poor-quality erections preoperatively, current and future lack of a sexual relationship, or other medical conditions that may adversely affect erections (e.g., diabetes mellitus, hypertension, psychiatric diseases, neurologic diseases, or medications that produce erectile dysfunction).


Postoperative treatment of erectile dysfunction also should be discussed, including information on phosphodiesterase type 5 (PDE-5) inhibitors, intraurethral and intracorporeal administration of vasodilators, vacuum erection devices, venous flow constrictors, and implantable penile prostheses. The discussion also should include the timing of the return of erections. The patient should be warned about the risk for development of Peyronie disease from injury to the penis during sexual activity without a rigid erection (Ciancio and Kim, 2000). He should also be informed that early postoperative use of intracavernosal injection therapy with vasodilating drugs allows most patients to have normal erections (with arterial blood) shortly after the catheter is removed. This protects against the occurrence of atrophic changes in the penis and allows the resumption of sexual activity early in the postoperative period. If erectile function is a high priority for the patient, he should be reassured that erections almost always can be restored, regardless of whether nerve-sparing surgery could be performed.


The preoperative evaluation should consider the likelihood of success in achieving all goals of surgery and in determining whether nerves can safely be spared. The surgeon should also discuss the possibility of performing cutaneous nerve graft interposition if one or both cavernous nerves must be resected (Kim et al, 1999; 2001; Quinlan et al, 1991; Secin et al, 2007). There are limited data concerning the effectiveness of such grafts. In most patients with a tumor so advanced that one or both nerves must be resected, postoperative radiotherapy or hormone therapy is likely to be required, which could nullify the potential benefits of nerve grafting. Some studies question whether nerve grafting is effective (Davis et al, 2008). The surgeon should also discuss the possible need for and potential side effects of adjuvant postoperative radiotherapy and/or hormone therapy, if the final pathology specimen reveals adverse prognostic features.



Surgical Technique


Radical prostatectomy involves complete removal of the prostate gland and seminal vesicles and usually includes a modified pelvic lymph node dissection as well. The key steps in performing anatomic nerve-sparing radical retropubic prostatectomy are as follows:











The surgeon should strive for complete apposition of the bladder neck and the urethra, with a watertight, tension-free closure.


Pelvic lymphadenectomy is optional in patients at low risk for lymph node metastases. Patients who elect to undergo lymphadenectomy should decide in advance whether they wish to proceed with the prostatectomy if there are nodal metastases. If they do not wish to proceed, the excised lymph nodes are sent for frozen-section examination during the operation. Otherwise, intraoperative frozen-section analysis of pelvic lymph nodes is unnecessary. Some have argued that a more extensive pelvic lymphadenectomy yields better outcomes, but compelling evidence for this is lacking (Weight et al, 2008), and more extensive lymphadenectomy carries a greater risk for postoperative genital and lower extremity lymphedema and lymphocele (Bader et al, 2002; Allaf et al, 2004; Musch et al, 2008). Thromboembolic events and reinterventions are more common in patients with symptomatic lymphocele (Musch et al, 2008).


The key to preserving urinary continence is to perform a meticulous dissection, avoiding injury to the external urinary sphincter. Preservation of the bladder neck is unnecessary to achieve good urinary continence. In patients with high-volume or high-grade tumors involving the base of the prostate, preservation of the bladder neck may risk positive surgical margins.


Meticulous dissection is required to preserve the neurovascular bundles. In performing nerve-sparing surgery, the neurovascular bundles are identified at the apex of the prostate (the dissection can also be performed in an antegrade fashion beginning at the base), and the bundles are dissected free of the posterolateral surface of the prostate gland. Hemostatic sutures or clips may be used to control bleeding from the neurovascular bundles. Use of electrocautery or a harmonic scalpel risks irreversible thermal injury to the neurovascular bundles.


The prostatic pedicles are suture ligated or hemoclipped and divided close to the gland, avoiding incision into the prostatic capsule. In performing the seminal vesicle dissection, care must be taken to avoid injury to the neurovascular bundles situated immediately lateral and posterior to them.




Cancer Control


The principal objective of radical prostatectomy is to completely excise the cancer. Important cancer control end points are pathologically organ-confined disease with clear surgical margins, biochemical recurrence (detectable serum PSA), local progression, metastases, cancer-specific survival, and overall survival. As discussed above, depending on the Gleason score and the PSA doubling time, biochemical (PSA) evidence of recurrence usually precedes clinical metastases by a mean of about 8 years and cancer-specific mortality by about 13 years (Pound et al, 1999).


Nonprogression rates vary with clinical and pathologic risk factors. Independent clinical prognostic factors are tumor stage, Gleason score, preoperative PSA level, and calendar year of diagnosis and treatment. Adverse prognostic features include non–organ-confined disease, perineural or lymphovascular space invasion, extracapsular tumor extension, positive surgical margins, seminal vesicle invasion, and lymph node metastases (Grossfeld et al, 2000; Shariat et al, 2004). In the PSA era, there has been a dramatic stage migration and improvement in prognostic features and treatment outcomes (Han et al, 2001a; Moul et al, 2002).


A rising serum PSA level is usually the earliest evidence of tumor recurrence after radical prostatectomy (Pound et al, 1999). Biochemical recurrence is frequently used as an intermediate end point for treatment outcomes; however, not all patients with biochemical recurrence ultimately develop metastases or die of prostate cancer. In rare instances with high-grade or neuroendocrine tumors that do not produce much PSA, there can be palpable evidence of recurrence despite an undetectable PSA level, indicating a role for DRE in monitoring of patients.


In the first author’s personal series, including more than 5000 open anatomic radical retropubic prostatectomies dating to 1983, the actuarial 10-year cancer progression-free survival probability was approximately 85% for patients with organ-confined disease, 65% for men with extracapsular tumor extension without cancerous surgical margins, 55% for men with extracapsular tumor extension and cancerous surgical margins, 25% for patients with seminal vesicle invasion, and 10% for patients with lymph node metastases. Patients treated in the PSA era have approximately five percentage points more favorable results within each pathologic category. Within these pathologic tumor stages, cancer progression rates also were strongly associated with other parameters. Case selection and the duration and frequency of follow-up monitoring are also important determinants of postoperative outcomes.


Radical prostatectomy also provides long-term cancer control in about half of highly selected men with clinical stage T3 disease (Freedland et al, 2007; Loeb et al, 2007).


Jun 4, 2016 | Posted by in ABDOMINAL MEDICINE | Comments Off on Definitive Therapy for Localized Prostate Cancer: An Overview

Full access? Get Clinical Tree

Get Clinical Tree app for offline access