Presenting Features of Prostate Cancer

The most common presenting features result from urinary tract obstruction due to the enlarged prostate gland. Urgency, nocturia, frequency, and hesitancy are the most common signs of an enlarged prostate gland with bladder-neck obstruction. Aside from the abrupt appearance of these symptoms, there is little to differentiate symptoms that are secondary to cancer from symptoms secondary to benign causes of prostatic enlargement. Other presenting signs and symptoms, although uncommon, should be recognized. New onset of impotence or less firm penile erections (which is often not elicited in history taking and mistakenly ascribed as normal in an aging man) should raise the possibility that prostate cancer may be present and is involving periprostatic tissue involved in erectile function. On occasion, distant metastatic disease heralds the diagnosis of prostate cancer. In these cases, back pain secondary to vertebral metastases is the most common initial feature. In a small percentage of patients, acute urinary retention caused by bladder-neck obstruction, bilateral hydronephrosis caused by periaortic lymph-node enlargement, or spinal cord compression caused by epidural extension from prostate cancer may be initial presenting features. Fortunately, these cases are unusual. However, prostate cancer should remain in the differential diagnosis of these symptoms in male patients.

Rarely, patients have supraclavicular lymphadenopathy or abnormalities of hepatic function. Biopsy of a lymph node or liver may show adenocarcinoma of uncertain anatomic origin. The presence of positive staining for prostate-specific antigen (PSA) on paraffin-embedded or fresh tissue may provide a clinical and pathologic clue that the adenocarcinoma originated in the prostate.

Screening for Prostate Cancer

Prostate-specific antigen, a protease produced by prostatic epithelium, has gained wide acceptance as a serum tumor marker in the management of prostate cancer and is also potentially an important screening tool [6-8]. Prostate-specific antigen is measured by an immunoassay and may be elevated in various benign and neoplastic prostate disorders [9-22]. A normal range of PSA is from 0 to 4 ng/mL. Elevations occur in patients with benign prostatic hyperplasia and prostatitis as well as prostate cancer. Although PSA is prostate specific, it is not prostate cancer specific because 20% to 50% of men with benign prostatic hyperplasia may have slightly elevated values, usually in the range of 4 to 10 ng/mL. Conversely, normal PSA values may be found in 25% to 45% of men with localized prostate cancer [9]. Although routine digital rectal examination does not usually cause clinically important increases in PSA, determinations of PSA should, if possible, be obtained before prostatic manipulation [23].

The value of screening for prostate cancer has become controversial because of the recent widespread availability of PSA and transrectal ultrasonography. As with any screening program, a successful program for prostate cancer should demonstrate that cancer deaths are decreased as a result of detecting more cancers earlier. Similarly, the potential side effects of treatment resulting from screening must be weighed against the morbidity resulting from the disease itself. Two fundamental hypotheses underscoring most screening programs for prostate cancer are 1) when prostate cancer is treated early and organ confined, substantial cure rates may occur; and 2) prostate cancer follows an orderly progression in a local, regional, and disseminated fashion, which, if untreated, may result in morbidity and death. Prospective trials are under way to determine if, in fact, screening programs can reduce the number of prostate cancer deaths. Until such studies are completed, the following practical approach is recommended by many experts in the field.

A digital rectal examination suggesting cancer can be followed by transrectal ultrasonography, regardless of the level of PSA. Transrectal ultrasonography, if abnormal, can help direct the prostate biopsy that is obtained; if transrectal ultrasonography abnormalities are not identified, a biopsy of the physically abnormal area can be performed.

The likelihood of improving the detection rate of prostate cancer with transrectal ultrasonography in a patient with normal results of PSA and digital rectal examination is low. Most experts would not recommend routine transrectal ultrasonography in this clinical situation. If the PSA is elevated (usually more than 10 ng/mL), the likelihood of detecting cancer is much greater regardless of the digital rectal examination findings. A PSA value greater than 10 ng/mL usually calls for transrectal ultrasonography and biopsy of suspicious areas if present. If no abnormalities are noted on sonographic or physical examinations, some experts recommend obtaining multiple "blind" biopsy samples.

The greatest dilemma results in patients who are found to have a PSA level between 4 and 10 ng/mL. Transrectal ultrasonography may be useful if the digital rectal examination is negative. Whether a biopsy should be performed if transrectal ultrasonography is negative is controversial. One reasonable approach is to continue to follow the patient with repeated PSA determinations and to consider a biopsy if the PSA continues to increase.

The American Cancer Society recommends that asymptomatic men older than 40 years be screened for prostate cancer with digital rectal examination. More recent recommendations are for men older than 50 years to obtain both a PSA test and a digital rectal examination and to undergo further evaluation if either is abnormal [22]. Prostate-specific antigen can improve the sensitivity and specificity of digital rectal examination and thus continues to warrant vigorous study as a screening modality for asymptomatic men. Although PSA has not yet been approved by the Food and Drug Administration for detection of early prostate cancer, many additional consultants do recommend a PSA determination with an annual digital rectal examination for men older than 50 years. However, not all advisory groups endorse screening of patients for prostate cancer.

International consensus for routine screening is lacking. Consensus conferences in Sweden, France, and Canada and statements by the International Union against Cancer and the United States Public Health Service Task Force have not endorsed routine screening [24]. Until definitive answers can support that screening and aggressive treatment programs for detecting cancer can lower mortality rates, the criteria for screening will continue to be controversial.

Transrectal ultrasonography of the prostate gland has been used more frequently during the last decade [12, 25-31]. This test can detect small hypoechoic lesions that are missed by digital rectal examination; conversely, transrectal ultrasonography may also fail to detect those lesions discovered on physical examination. It can help localize lesions for biopsy and provide an excellent assessment of tumor size and response to therapy (if the prostate gland is an indicator lesion). Magnetic resonance imaging is also being evaluated for defining local prostate anatomy. Endorectal surface coil magnetic resonance imaging may offer some advantages to ultrasound in staging local prostate cancer.

Value of Prostate-specific Antigen Determination in Established Prostate Cancer

In patients with prostate cancer, serum PSA levels are proportional to the clinical stage of the disease and the volume of prostate cancer found in the gland. An increase in PSA of more than 80 ng/mL generally correlates with more advanced tumor stage, including metastatic disease. The actual serum PSA level before radical prostatectomy may help predict the likelihood of discovering lymph-node or seminal-vesicle involvement in resected radical prostatectomy specimens. Increasing PSA values after definitive radiotherapy for localized prostate cancer can predict residual localized cancer or the development of metastases. Following the initiation of hormonal therapy in metastatic disease, PSA values, if elevated, can decline to normal values. The persistence of such a decrease may help predict a favorable, prolonged response to therapy [7, 32-42].

Biopsy and Histopathologic Grading

Biopsy Procedures

Several biopsy procedures for establishing the diagnosis of prostate cancer exist. These include digitally guided biopsies using a core needle; ultrasound-guided biopsy; fine-needle aspiration; and chip sections obtained at the time of transurethral resection. Although biopsies have usually been done with the Tru-Cut (Baxter, Newark, New Jersey) or Vim Silverman needles, smaller-bore needles placed in spring-loaded biopsy devices are more commonly used because the procedure can minimize patient discomfort and lower infection risk [43-46].

Although both digitally guided transperineal and transrectal routes can be used, the latter is more popular. Both the transperineal and transrectal routes are used in association with ultrasound guidance. Multiple planes of prostate anatomy can be visualized, resulting in a more accurate sample procurement. Samples obtained for both digital and ultrasound biopsy techniques allow accurate histologic grades of the sample neoplasm. Complications of core-needle biopsies, which include urinary tract infection and sepsis, can be substantially diminished with the aid of antibiotic coverage before and after biopsy, cleansing enemas before biopsy, and the use of smaller-bore biopsy needles. Rectal bleeding and transient blood in the ejaculate commonly occur.

Fine-needle aspiration, which is usually done on an outpatient basis as a diagnostic modality, may have the potential for lessening the morbidity and hospitalization costs associated with the more standard biopsy procedures [47, 48]. However, the use of this modality requires experienced cytopathologists to interpret the aspirated contents. Another potential drawback of fine-needle aspiration is the inability to obtain an accurate Gleason score on the aspirated cellular contents.

Pathologic Findings

The pathologist plays an important role in both establishing the pathologic diagnosis and in determining the differentiation of the lesion. Although many different pathologic grading systems for adenocarcinoma of the prostate have been developed, the Gleason scoring system has gained the widest acceptance. This system divides prostate cancer into five histologic patterns, ranging from 1 to 5. Gleason patterns 1 and 2 represent a well-differentiated prostate adenocarcinoma; Gleason pattern 3 represents a moderately well-differentiated cancer; and Gleason patterns 4 and 5 represent a poorly differentiated or anaplastic lesion [49].

The Veterans Administration Cooperative Urological Research Studies (VACURG) cooperative group [50-57] attempted to establish the optimal therapeutic program for all stages of prostate cancer and has served as the foundation for prostate cancer management for 20 years. In these studies, a "primary" score and a "secondary" score were given to each prostatic specimen. Thus, if a biopsy lesion consisted of 70% pattern 3 and 30% pattern 4, the primary plus secondary pattern would be a 3 + 4 or a Gleason 7. Thus, the most well-differentiated cancer would consist entirely of a Gleason pattern 1 (primary + secondary = 1 + 1 or Gleason 2), and the most poorly differentiated cancer would be a 5 + 5 or Gleason 10. Hence, many pathologists report the score as 3 + 4 or 2 + 3; others use 7/10, 6/10, and so forth; others use "well-differentiated," "moderately well-differentiated," or "poorly differentiated" prostatic adenocarcinoma. The primary histopathologic finding of the prostate cancer predicts overall survival.

Clinical Evaluation of Prostate Cancer

The most useful staging evaluations after prostate cancer is diagnosed are routine serum chemistries, urinalysis, liver function studies, blood urea nitrogen, creatinine, and complete blood count. It is also useful to obtain a baseline chest radiograph and electrocardiogram because many patients with localized disease have comorbid disease or underlying cardiopulmonary disease. This information may be helpful in determining whether surgery or radiation therapy should be selected in cases of localized prostate cancer.

A radiographic evaluation of the upper urinary tracts should be done. In the past, this was most commonly obtained by an intravenous pyelogram. Although many experts still find this very useful, the use of abdominal-pelvic computed tomographic scanning has in many cases supplanted the intravenous pyelogram. Proponents of intravenous pyelography cite the ability to evaluate a large proportion of vertebral bones, as well as the kidney, urinary tract, and bladder. However, the information obtained from an abdominal-pelvic computed tomographic scan provides additional data in the overall staging evaluation of patients with prostate cancer [58-63]. The computed tomographic scan is useful in evaluating abnormalities of the urinary tract and the presence of para-aortic and pelvic adenopathy. In addition, the liver and other visceral organs are easily visualized.

Magnetic resonance imaging of the prostatic and periprostatic bed may have some advantages in selected patients in assessing surgical suitability of these patients. Transrectal ultrasonography, if used initially to help establish the diagnosis of prostate cancer, may sometimes be useful in the follow-up evaluation to determine response of primary indicator lesions following therapy.

The mainstay of metastatic assessment in patients with prostate cancer relates to evaluation of the radionuclide bone scan and correlative plain radiograph films for any suspicious areas. Because prostate cancer often metastasizes to axial and appendicular bones, bone scanning is an important diagnostic tool in initial staging and follow-up evaluation [64, 65].

In the past, lymphangiography was suggested as a useful tool in assessing retroperitoneal and pelvic adenopathy and may help in localizing suspicious pelvic lymph nodes for fine-needle aspiration. However, most patients today do not undergo routine lymphangiography because of the availability of computed tomographic scanning.

Recently, laparoscopic pelvic lymphadenectomy has been used to assess pelvic lymph-node involvement [66, 67]. Although early experiences with the procedure were associated with moderate complications (vascular injury, viscus injury, deep venous thrombosis, lymphedema, infection, small-bowel abnormalities, urinary retention, and scrotal swelling), these problems have lessened with greater operator experience. The optimal candidate to undergo a laparoscopic pelvic lymph-node dissection is still being debated, but current recommendations include patients with clinically localized disease who have a poorly differentiated lesion and an elevated PSA of more than 30 ng/mL. Patients who are undergoing perineal prostatectomy can undergo laparoscopic lymphadenectomy before the procedure, and patients who will have radiation therapy are excellent candidates, especially if their clinical disease is diagnosed as stage B2 or C. If lymph nodes are found to contain cancer (thus classifying the patient's condition as stage D1), the physician can thus avoid unnecessary radiation therapy and radical surgery given that cure using these treatments is unlikely.

Prostate Cancer Staging

Although there are multiple prostate cancer staging systems, the following provides the physician with a framework to engage in meaningful communication with specialist physicians who are also involved in the patient's care—the urologist, radiation oncologist, and medical oncologist.

Prostate cancer is conceptually divided into stages A, B, C, and D. A comparison of the current conventional clinical staging and tumor, nodes, metastases (TNM) system is shown in Table 1. The TNM staging system is gaining in popularity because it more accurately allows more specific comparison of staging evaluations across various investigations. More accurate measurement of primary prostatic lesions by transrectal ultrasonography and assessment of lymph node status by laparoscopic lymphadenectomy should enhance use of the TNM system even further.

Patients with clinical stage A disease are usually subdivided into A1 and A2 disease. Patients with clinical stage A1 disease are usually diagnosed after a transurethral resection is performed for evaluation of urinary tract symptoms (hesitancy, urgency, nocturia, frequency). Thus, stage A1 disease is clinically unsuspected by history and physical examination; it is focal in its anatomic distribution and it is usually well differentiated. Pathologic focality has been defined as less than or equal to three "chip" sections or less than 5% of the entire number of chips resected (although this is not uniformly accepted). In A1 disease, the combined Gleason primary and secondary score is usually 4 or less. Stage A1 disease is probably present in many aging men, and it is the type of lesion most likely to be diagnosed at autopsy in asymptomatic patients who have died of other causes. Stage A1 is also likely to be diagnosed during routine screening evaluations.

Patients with clinical stage A2 disease usually have a clinical presentation similar to that of patients with stage A1. However, the disease involves the chips more diffusely and involves more than 5% of the resected specimens or the Gleason primary and secondary score is greater than 4. In contrast to patients with stage A1 disease, a large percentage of patients with stage A2 disease, when surgically staged, are found to have positive lymph nodes.

Stage B

An asymptomatic patient who on routine rectal examination has a prostatic nodule or induration limited to the prostate gland has clinical stage B disease. A patient with a relatively small nodule is part of a subset of stage B patients, stage B1 nodule. Other subdivisions of stage B prostate cancer exist, usually relating to increasing volume and glandular extension of cancer and include categories B1 and B2.

Patients with both stage A and B cancer usually do not have any acute urinary symptoms, although some individuals may have signs and symptoms of urinary obstruction thought to be secondary to benign prostatic hyperplasia.

Stage C

A patient with clinical stage C disease usually has a more acute onset of objective urinary symptoms. These patients may have a precipitous change in their urinary frequency, nocturia, hesitancy, and urgency, spanning a few weeks to months. In addition, patients may experience new onset of impotence, which results from cancerous involvement of periprostatic tissue that controls erectile function. On physical examination, these patients usually have bilobar disease. There may be areas of diffuse induration or nodularity throughout the prostate. Local contiguous extension to the pelvic side wall and seminal-vesicle involvement may be found on digital rectal examination. It should be emphasized that patients with clinical stages A to C disease have shown no evidence of metastatic disease on physical examination or clinical workup.

Stage D

Stage D is divided into subcategories, D0 (disease clinically limited to the prostate gland with elevated acid phosphatase), D1 (positive pelvic lymph-node involvement following diagnosis), and D2 (extrapelvic soft-tissue metastases or, more commonly, bony metastases). Several investigators have used the term stage D3 disease to indicate metastatic disease that has become refractory to hormonal therapy.

Discrepancies between Clinical versus Surgical Staging

When clinically staged with history, physical examination, blood tests, and radiographic studies, patients with prostate cancer are generally understaged when compared with the data obtained from surgical or laparoscopic staging. Patients may be surgically staged with a lymph-node dissection, fine-needle aspirate of the involved pelvic lymph node, and more recently, with laparoscopic pelvic lymphadenectomy. These procedures are purely diagnostic and have no therapeutic value. Rarely, a suspicious focus on bone scan may be biopsied. Approximately 25% to 30% of individuals with stage A2 or B disease have positive pelvic lymph nodes if surgically staged. Nearly 40% to 50% of patients with stage C have positive pelvic lymph nodes. Thus, even though results of the physical examination and blood and radiographic evaluations are all negative, nearly 25% to 50% of men with stage A2 to C have evidence of metastatic spread to pelvic lymph nodes, thus making them stage D1 [68-72]. It should be emphasized that variations exist in assessing the extent of the primary tumor when digital rectal examination is used by specialists of competing treatment modalities. Similarly, the necessity of obtaining surgical staging of lymph nodes continues to be hotly debated. Some investigator bias often is a factor in these decisions.

The prognostic importance of pelvic lymph-node involvement in patients with prostate cancer is profound. After 4 to 5 years of follow-up, nearly 80% of patients with positive lymph nodes eventually develop metastatic, extrapelvic disease. In contrast, only 20% of patients with negative lymph nodes develop distant metastatic disease. Thus, lymph-node involvement is highly predictive of micrometastatic and subsequent systemic spread of disease.

In addition, DNA tumor ploidy, determined by flow cytometry, has been shown to be an important predictor of progression in those patients with localized prostate cancer who have had a radical prostatectomy. However, this technique has not yet found its way into everyday therapeutic decision making in the overall management of patients with prostate cancer [73, 74].

Recommended staging evaluations and frequencies of tests are outlined in Table 2 [75]. It should be emphasized that the recommendation for initial and subsequent follow-up evaluations is intended for patients in clinical protocols. Patients managed outside of clinical investigation settings might be managed differently.

Prostate Cancer Management

In general, if a patient is planning to undergo radical prostatectomy, evaluation of the pelvic lymph nodes is recommended. Most commonly, if frozen section results of the lymph node are negative, urologists proceed with the radical prostatectomy. Others perform two separate operations. After the final pathologic report is received following lymphadenectomy (approximately 4 to 5 days later), the radical prostatectomy is done. If macroscopic disease is found in the pelvic lymph nodes, most urologists would not proceed with the radical prostatectomy because it is not considered to be a "curative" procedure. Some data, however, suggest that if a single microscopic focus of disease is found, removal of the lymph node can be therapeutic; however, most urologists would not recommend continuing with the prostatectomy if the lymph nodes are positive at the time of contemplated prostatectomy. In such instances, "local" control to treat urinary symptoms of prostate gland dysfunction can be addressed with either local radiation therapy, hormonal therapy, or transurethral resection.

If a patient has chosen radiation therapy, surgical lymphadenectomy is usually not performed because it may add to the morbidity of radiation therapy. Thus, many patients in the past who were being treated with radiation therapy had suboptimal information regarding the stage of their disease. This creates difficulties in making meaningful comparisons in treatment outcomes between surgically and nonsurgically staged patients. However, laparoscopic pelvic lymphadenectomy is becoming a more common procedure, thus allowing more accurate staging information. In addition, complications previously arising from surgical lymphadenectomy staging procedures can be minimized with laparoscopic lymphadenectomy.

Management of Localized Disease (Stages A2, B, and C)

The optimal treatment for localized disease, definitive radiation therapy or radical prostatectomy, has been controversial for decades. Internists have an important role to play in advising patients and in assuring that patients are hearing "both sides of the story," as most often these patients have only been seen by a urologist. A frank and thorough discussion should take place between the patient's primary care physician, consulting radiation oncologist, and urologist before a decision regarding therapy is made. The controversy has become even more heated based on a recent publication from Swedish investigators in which a consecutive sample of 223 patients who had early-stage prostate cancer were monitored for progression-free, disease-specific, and overall survival without any definitive therapy for early localized disease [76, 77]. In this study, 19 of the 223 patients had died of prostate cancer after a follow-up period of about 10 years. One hundred five of a total of 124 deaths were from other causes. The 10-year, disease-specific survival rate was 86.8% and was equally high in a subgroup of patients who met current indications for radical prostatectomy or radical radiation therapy. The authors concluded that the low disease-specific mortality rate suggests that untreated controls should be used for comparison when evaluating the effects of radiation therapy or radical prostatectomy.

In 1987, a National Institutes of Health Consensus Development Conference was organized to discuss the issue of localized prostate cancer management and to try to develop recommendations [78]; however, no consensus was reached regarding the most appropriate treatment. The conference identified an urgent need to initiate randomized, prospective studies in surgically staged patients to receive either radiation therapy or radical prostatectomy; such studies are underway.

Long-term survival data comparing radiation-treated patients and surgically treated patients suggest equivalence in prostate cancer-specific survivorship Figure 1 [79]. It should be emphasized that selection criteria for radiation therapy treatment have been less stringent because many patients who receive radiation therapy would not have been candidates for radical prostatectomy because of comorbid disease. The prognosis according to stage and treatment of disease is summarized in Table 3. These data are from studies in which adequate patient numbers (usually more than 50) have been evaluated and followed for survival for more than 5 years [78].

View larger version (15K): [in this window] [in a new window]   Figure 1. Long-term survival. Comparison of long-term survival of patients with stage T2a (B1) disease treated by radiation therapy in the Stanford University Hospital series (134 patients, ) and those treated by radical prostatectomy in the series at The Johns Hopkins Hospital (57 patients ) and Virginia Mason Medical Center (195 patients \#9679;) according to a personal communication (Shipley WU). (Reprinted with permission from Bagshaw MA, et al. NCI Monogr. 1988; 7:47-60.).

Radiation Therapy

Although several techniques of radiation therapy have been used, the most common is external-beam radiation to the prostate gland and periprostatic tissues. In the past, interstitial implants of radioactive gold or iodine were commonly used [80, 81]. Using the linear accelerator, 67 to 70 Gy is given to the prostatic bed over a period of 6 to 7 weeks [82-89]. Improvements in the technique of radiation therapy may allow specially shaped radiation fields to strictly limit radiation dosage to the prostate gland and to minimize exposure of normal tissue to ionizing radiation therapy, thus reducing gastrointestinal and genitourinary side effects. This results in the greatest local control and local eradication of prostate cancer.

Radiation therapy is often selected for patients with coexisting medical conditions that would preclude major surgery. Until recently, radiation therapy was the preferred modality for patients with stage B and C disease because of operative complications and side effects from radical prostatectomy. Patients with clinical stage C lesions are usually treated with radiation therapy because surgical removal of the prostate gland is difficult because gross and microscopic nodal cancer cells are often left behind.

Recently, data about positive prostatic biopsies after definitive radiation therapy have become available. In some series, nearly 50% to 60% of patients had residual prostate cancer cells 18 months to 2 years after definitive radiation therapy [90, 91]. A positive prostate biopsy after radiation treatment has been associated with a worsened prognosis in terms of the development of distant metastatic spread and overall survival. The management of a patient after radiation therapy with a positive prostate biopsy poses a particular challenge; therapeutic choices are "salvage radical prostatectomy," which is a difficult surgical procedure (associated with the high risk of incontinence, rectal injury, and impotence), or the use of hormonal therapy.

The main, although infrequent, complications of radiation therapy are listed in Table 4. Results of treatment for radiation therapy and radical prostatectomy are also shown in Figure 1 and Table 3. Strict comparisons are difficult to make because many patients in all the series have not been surgically staged and were thought to be inappropriate candidates for radical surgery. The 10- and 15-year overall survival, disease-free survival, and cause-specific survival rates are comparable between the two treatment modalities.

Radical prostatectomy has gained popularity in the past decade. This is due primarily to an advance in surgical technique that has allowed the maintenance of the neurovascular bundle responsible for sexual potency. In the past, these nerves were severed during the procedure. A sexually active man in his fifties or sixties with a B lesion now has a reasonable likelihood of maintaining his sexual potency after radical extirpative surgery. As discussed previously, the surgical approach allows the sampling of pelvic lymph nodes before performing the operation, thus leading to more precise surgical staging [92-100].

The complications of radical prostatectomy have in the past included impotence, a small incidence of incontinence, other operative-related morbidity, and death (Table 4). Improved surgical techniques have decreased these complications. Autologous blood donation by the patient obtained before the surgical procedure has also decreased the need for intraoperative homologous blood transfusion and its attendant risks.

Despite decades of treatment with either surgery or radiation therapy, there is only one randomized comparison of radiation therapy versus surgery for localized disease in surgically staged patients [92]. This study measured the time to first treatment failure as a primary end point, with an advantage for radical prostatectomy. Although many urologists use these data to suggest that surgery is preferable for stage A2 and B disease, many other investigators think that this study had many shortcomings, making any firm conclusions tenuous [100].

Management of Biopsy-Proven Prostate Cancer and Elevated Acid Phosphatase (Stage D0)

No standard recommendation exists for treating Stage D0 disease. A patient with local symptoms may need some local control, usually in the form of radiation therapy or transurethral resection of the prostate. However, some individuals may prefer a radical prostatectomy with pelvic lymph-node sampling. Nearly 65% to 70% of patients have positive lymph nodes in the presence of an elevated acid phosphatase level [101]. With time, these patients generally develop metastatic disease and will require systemic therapy.

Management of Stage D1 Disease

The management of patients with lymph-node-positive disease is controversial. The standard therapy is to "watch and wait." A substantial number of these patients will relapse systemically, and hormonal therapy is generally instituted at that time. However, many believe that more aggressive and earlier intervention should be recommended in patients with stage D1 disease [73]. Recent information suggests that the administration of hormonal therapy may be associated with a crude survival benefit if instituted earlier in the disease. In addition, patients with minimal metastatic stage D2 disease with an excellent performance status, if treated early in their disease, may have improved chances for survival. These data may have some relevance for even more minimal stage D1 disease and emphasize the need for initiation of prospective studies [102-107].

If a patient is sexually active, however, the institution of hormonal therapy needs to be considered carefully because this therapy results in loss of libido and impotence. Patients will also lose "vigor" and feel fatigued and may experience hot flashes and sweats after androgen ablation. In the absence of clear-cut evidence that early androgen ablation for stage D1 disease prolongs life, a decision can only be made after weighing all available issues including those pertaining to quality of life.

View larger version (40K): [in this window] [in a new window]   Figure 2. Scheme of achieving castrate androgen levels. Estrogens such as diethylstilbestrol inhibit the release of luteinizing hormone-releasing hormone (LHRH) from the hypothalamus, thus also diminishing the release of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) from the anterior pituitary gland. Analogs of LHRH initially stimulate but ultimately inhibit the release of FSH and LH from the anterior pituitary gland. Ketoconazole and aminogluthethimide inhibit steroid synthetic pathways. In the prostate cells, testosterone is converted into dihydrotestosterone (DHT) by the enzyme 5--reductase. Anti-androgens block the binding of DHT to its cytoplasmic receptor. (Garnick MB. Urologic Cancer. In: Rubenstein E, Federman DD; eds. Scientific American Medicine. New York, Scientific American. Section 12, IX:1-17, with permission.).

The use of hormonal therapy for prostate cancer dates back nearly 50 years. Huggins and Hodges [108] observed that prostate cancer was under the trophic influence of male hormones. The administration of androgens causes prostate cancer progression; elimination of androgens causes prostate cancer regression. The principles of metastatic prostate cancer therapy have remained basically unchanged since, although clinical investigators have attempted to refine the type and timing of hormonal therapy and modulation of side effects.

Methods of Androgen Ablation

Bilateral Orchiectomy

Bilateral orchiectomy is the easiest and most direct way to achieve castrate testosterone levels [108]. It can be done under local anesthesia. Many patients prefer the "medical" means of achieving castration partly due to the psychologic effect of the removal of the testes. Although the cosmetic aspects of bilateral orchiectomy can be minimized by performing a subcapsular orchiectomy, patients can be offered a testicular prothesis if necessary. The main side effect is hot flashes, which occur in approximately 50% of patients.

Diethylstilbestrol

Diethylstilbestrol inhibits the release of luteinizing hormone releasing hormone (LHRH) from the hypothalamus. This hormone controls the release of follicle-stimulating hormone and luteinizing hormone from the anterior pituitary. These gonadotropins act at the Leydig and Sertoli cells in the testes to control spermatogenesis and androgen production. Three milligrams of diethylstilbestrol per day is the current recommended dose. Cardiovascular complications and gynecomastia are the most important side effects of therapy [52]. In one randomized study of leuprolide versus diethylstilbestrol, there was a substantial increase in the incidences of thrombosis, phlebitis, pulmonary embolus, and peripheral edema in patients receiving diethylstilbestrol [110]. Because of the excessive cardiovascular morbidity [50] that can occur in patients with estrogen moieties, many authorities today believe that an LHRH analog plus an antiandrogen agent is preferable to avoid the cardiovascular complications despite the substantial cost differential between these preparations and estrogen.

Luteinizing Hormone Releasing Hormone

The use of LHRH analogs and antiandrogen agents have been recently introduced as a medical means to cause androgen ablation. Analogs of LHRH, such as leuprolide, initially stimulate, but ultimately inhibit, the release of follicle stimulating hormone and luteinizing hormone from the anterior pituitary. In one randomized, prospective study, leuprolide was favorably compared with diethylstilbestrol in achieving a similar overall survival and remission rate in patients with prostate cancer with a lower side-effect profile [110]. Specifically, gynecomastia, nausea and vomiting, peripheral edema, and cardiovascular side effects were substantially greater in the diethylstilbestrol-treated group compared with the LHRH group. Hot flashes were more common in the latter group [111].

Antiandrogens Agents

The adrenal gland produces about 5% of circulating androgens. Testosterone is converted into the active androgen moiety, dihydrotestosterone, in prostate cells by the enzyme 5- reductase. Anti-androgens, such as flutamide, cyproterone acetate, and certain progestational agents block the binding of dihydrotestosterone to its receptor and consequently prevent androgen action [112, 113].

Other Hormonal Therapies

In addition to diethylstilbestrol, LHRH analogs, and antiandrogen agents, a spectrum of other hormonal therapies have been used in either first- or second-line therapy or for patients with metastatic prostate cancer. One such therapy is megesterol acetate, an agent that can inhibit pituitary gonadotropin secretion [114]. An occasional patient may achieve some improvement in pain when the disease has become hormonally refractory. Estramustine, a hybrid molecule (the combination of an estrogen moiety with an alkylating agent), has limited activity in hormonally refractory prostate cancer [115]. The anti-fungal agent, ketoconazole, can cause inhibition of steroidogenesis and can provide occasional improvement in patients' symptoms [116, 117]. The doses selected are generally high, and great caution should be used if ketoconazole or its congeners are used in patients with prostate cancer. Aminoglutethimide has been used in some trials with limited success [118].

Maximal Androgen Blockade

Recent data suggest that the combination of an LHRH analog with an antiandrogen agent may actually improve the overall survival of patients with excellent performance status in minimal metastatic disease. Maximal androgen blockade attempts to eliminate all circulating androgens (testicular and adrenal) by decreasing the exposure of prostate cancer cells to these potent trophic hormones [113].

National Cancer Institute Intergroup Study on Combined Therapy

In an important study sponsored by the National Cancer Institute, patients with stage D2 disease were randomly allocated to receive either leuprolide plus flutamide or leuprolide plus placebo [119]. In this study, 603 evaluable patients were entered at many institutions, with approximately 300 patients in each arm. Multiple variables were monitored. Criteria of the National Prostate Cancer Project were used to evaluate response. The end points of the study were progression-free survival and overall survival. The results showed that diarrhea was more common in the combined leuprolide-flutamide group than in the leuprolide-placebo group. Subsequent analysis showed that use of maximal androgen deprivation with the LHRH analog plus the antiandrogen agent was more effective than the LHRH analog alone for metastatic prostate cancer. Both progression-free survival and overall survival with combined therapy were better than the results with LHRH analog monotherapy.

Given the available data, patients with minimal metastatic disease may indeed benefit from combined hormonal therapy. Although it is difficult to say that every patient should receive combined hormonal therapy, the data already have shown clearly that many patients can have prolonged disease-free survival and that overall survival is increased when combined hormonal therapy is used.

Because the use of an LHRH analog plus an anti-androgen have been associated with improvement in overall survivorship in selected patients, many experts now believe that the combination of an LHRH analog, such as leuprolide, and an antiandrogen agent, such as flutamide, represent the best first-line therapy for the treatment of newly diagnosed metastatic disease. Although not all studies have confirmed the advantages of combination versus single modality therapy, the use of maximal androgen blockade (orchiectomy plus flutamide or LHRH analog plus flutamide) are rapidly gaining widespread use. A recent update of this study shows a substantial survival advantage for the combined-modality treatment. The median survival of the entire group was 35.1 months in the leuprolide-flutamide arm compared with 29.3 months in the leuprolide-placebo arm. In the subset of patients with minimal disease and excellent performance status, the median duration of survival is 61 months compared to 41 months for leuprolide plus placebo [120].

Combined Therapy with or without Orchiectomy

Several additional studies have shown a tendency toward improvement in rates of response and time to progression using combination therapy compared with monohormonal therapy. The results of several ongoing randomized clinical trials on maximal androgen deprivation have shown some early signs of potential clinical improvement [121, 122]. An additional study evaluated orchiectomy with or without the addition of the anti-androgen nilutamide [123]. All patients were stage D2. Those treated with a combination of antiandrogen agent plus orchiectomy showed a significantly greater number of positive objective responses than did the patients who received orchiectomy alone. Progression-free survival was improved in the antiandrogen group, but the median time to progression was 12 months for both groups. In addition, there was some suggestion of an improvement in median length of survival in the nilutamide group. These data were interpreted as suggesting some benefit for combined treatment or orchiectomy plus nilutamide versus orchiectomy alone. The currently available data suggest some "cautious optimism" for combined androgen blockade or maximal androgen deprivation compared with monohormonal therapy. With longer follow-up and additional patient accrual, such therapy may become more widely accepted.

How should the practicing physician today manage patients with newly diagnosed metastatic prostate cancer and when should hormone therapy be initiated? With a patient who is ambulatory, has an excellent performance status, and is compliant, a case can be made for treatment with combined hormonal therapy, which would include an antiandrogen agent plus an LHRH analog or orchiectomy plus an antiandrogen agent. Based on earlier Veterans Administration Cooperative Urologic Research Group studies, a recommended treatment strategy was to initiate hormonal therapy in symptomatic patients. The goal was to palliate symptoms, as patients with symptomatic prostate cancer will often have pain from bone involvement, hydronephrosis secondary to retroperitoneal lymphadenopathy, or, rarely, visceral metastases. However, recent data have questioned this dogma. Many men now receive hormonal therapy once the diagnosis of prostate cancer has been established, when they may still be asymptomatic. This earlier timing strategy attempts to enhance the effectiveness of hormonal treatment on a smaller volume of metastatic disease.

A patient with newly diagnosed metastatic prostate cancer (stage D2) with weight loss, bony pain, or visceral involvement should have hormonal therapy instituted immediately. Similarly, if a patient has impending bilateral hydronephrosis and azotemia or impending spinal cord compression, the use of urgent hormonal therapy, usually in the form of a bilateral orchiectomy, is generally indicated. An LHRH analog should not be the initial therapy under these circumstances because of the possibility of causing initial worsening of cord compression and urinary obstruction. The goals of hormonal therapy are to improve a patient's symptoms and performance status and to cause tumor regression. Achievement of this goal is often difficult to measure because patients may feel much better but show no or little improvement on bone scan. Improved performance status and symptoms accompanied by a decrease in PSA are reliable indicators that hormonal therapy is having a beneficial effect.

Stage D3: Hormonally Refractory Disease

After receiving hormonal therapy, patients often have a relapse, usually in the form of an increasing PSA level, a worsening of bone scan abnormalities, and clinical symptoms. It is imperative at this point to determine whether the patient's disease is truly refractory to hormonal therapy. A serum testosterone level may be obtained to ensure that the patient has been compliant with taking his medication. If the serum testosterone value is not at the castrate level (usually 35 to 50 ng/100 mL), there may be a role for additional hormonal manipulation or reinstitution of the originally prescribed medicine. If, however, the patient has castrate levels of serum testosterone and is truly refractory based on increasing PSA level, worsening bone scan and symptoms, or performance status, it is usually appropriate to institute additional therapy. Many physicians continue hormonal therapy to maintain castrate levels while considering the addition of other therapies. However, an individual patient, once refractory to hormonal treatment, rarely responds to a second hormonal therapy.

On occasion, patients who have initiated therapy with an LHRH analog may have a transient worsening of bone scan appearance despite improvement in overall performance status and a decrease in acid phosphatase level. Bone scan "flare" must be considered before the physician can consider a patient hormonally refractory. With continued treatment, a patient's well-being and radiographic abnormalities may improve after an original "flare" [124].

Special Management Problems with Hormonally Refractory Metastatic Disease

When patients with hormonally refractory disease experience new back pain and neurologic symptoms, the diagnosis of spinal cord compression should be strongly considered. The use of spinal magnetic resonance imaging and computed tomographic scanning have greatly facilitated this diagnosis. The immediate institution of dexamethasone and judicious palliative radiation therapy (and in unusual circumstances, decompressive laminectomy) can help prevent or delay the development of paraplegia. Acute urinary obstruction requires the consideration of urinary tract relief. Despite the high prevalence of bone metastases, which are usually osteoblastic, it is uncommon for hypercalcemia to develop in patients with metastatic prostate cancer. So rare is hypercalcemia, in fact, that clinicians should consider and rule out a second underlying diagnosis, such as sarcoidosis and hyperparathyroidism, as a cause of the hypercalcemia. Patients with tumorous involvement of weight-bearing areas such as hips and acetabula or humerus lesions should be considered for palliative radiation therapy to prevent pathologic fracture, especially if the weight-bearing areas have suffered substantial cortical bone loss. Discussion with orthopedic surgeons and radiation oncologists will help determine which patients are best suited for prophylactic radiation therapy to weight-bearing areas.

Pain management in patients with metastatic prostate cancer requires a compassionate physician and family. These patients often suffer excruciating bone pain that can be controlled with localized palliative radiation therapy and narcotic analgesics. It is common for patients to require increasing doses of morphine or other narcotics to control their bone pain while receiving radiation therapy. The use of systemic chemotherapy has never been associated with any meaningful improvement in overall survival or overall objective benefit [125, 126]. It is often used, however, and should be considered in selected patients who have become refractory to hormonal treatment and as a means to help control pain. Long-term and high-dose corticosteroids should be used sparingly because the development of peripheral myopathy may impair already-compromised locomotor ability.

Several new agents are being investigated for the treatment of metastatic disease that has become refractory to standard agents. Suramin, an antiparasitic agent, can inhibit the action of growth factors with their receptors, including epidermal growth factor, platelet-derived growth factor, transforming growth factor ß, and basic fibroblast growth factor [127]. In-vitro work has shown that suramin may inhibit various prostate cancer cell lines [128]. In one recent study, doses of suramin were administered to maintain a plasma level of 150 to 300 ng/mL. Of 26 patients evaluable for response, a substantial number had objective remissions in bone metastases and soft-tissue disease, with most patients having a decrease in PSA level. Toxicity of the agent can be prohibitive and patients should be considered for suramin therapy only in the context of very strictly controlled clinical trials [129].

Other investigational agents include somatostatin analogs [130], bisphosphonates [131, 132], radionuclides [133-135], and gallium nitrate [136]. Radiation therapy (localized and hemibody) may be used for control of pain in patients with widespread metastatic disease [137]. Physicians should consider referring patients for participation in one of these investigational programs if conventional management is unable to control symptoms and if the patient is a candidate for investigational treatments.

The management of patients with hormonally refractory disease is difficult. The physician must manage pain, help in providing adequate nutrition, and preserve patients' mobility. Patients may be severely impaired in locomotion because of the extensive amount of bone disease and experience an extraordinary amount of pain from their bone disease. They often lose a substantial amount of weight. At this time, it is appropriate to consider short-term courses of steroids and use of hospice services. Clearly, in a patient whose disease has become refractory to hormone therapy and the role of palliative radiation therapy treatment for pain control has been exhausted, the patient, family, and physician must frankly discuss issues regarding quality of life versus measures to prolong life.

Psychosocial Support

Internists can and should play a key role in providing patients with psychosocial support. At the time of initial diagnosis, the physician should convey optimism for cure to the patient after localized prostate cancer is surgically removed. Similar optimism should be communicated to patients undergoing radiation therapy, especially if the patient has had negative pelvic lymph nodes on laparoscopic evaluation. However, even after undergoing curative therapy, patients may experience feelings of depression, especially sexually active patients who have become impotent. The physician must engage in a frank discussion with the patient and the patient's sexual partner about the role of penile prostheses and intrapenile injection of vasoactive substances to obtain erection. The goals and side effects of both localized and systemic therapy should be discussed early in the course of the disease.

Hospice services are very useful in providing both psychosocial support and medical care in the terminal phases of the disease. Maintaining adequate nutrition, pain control, and management of constipation (usually induced by heavy analgesic use) are three clinical problems that, if managed well, can substantially improve quality of life. Patients should be provided with educational materials (or how and where to obtain them), be part of the decision-making process regarding treatment options, and be directed to other specialists, if necessary, for counseling.

Cost Associated with Therapy

The treatment of patients with prostate cancer is costly. For individual patients undergoing a curative course of radiation therapy, the cost is approximately $8000 to $11 000 [138, 139]. The cost of radical prostatectomy and associated hospital costs ranges from $10 000 to $18 000. A laparoscopic lymphadenectomy may cost $3000. Treatments for metastatic disease are expensive. The annual cost for an antiandrogen agent such as flutamide and the LHRH analog, leuprolide, is $6000 to $7000 ($2500 for flutamide, $3700 for leuprolide). In contrast, a 1-year supply of diethylstilbestrol costs less than $100. However, avoiding cardiovascular complications associated with diethylstilbestrol and eliminating the need for prophylactic radiation therapy to the breasts to avoid gynecomastia may offset the cost associated with LHRH analog therapy and antiandrogen agents. Bilateral orchiectomy costs approximately $2000 to $3000, including the surgeon's fees and associated hospital fees. Clearly, substantial costs can be incurred during the terminal phases of a disease when the patient may become debilitated and immobilized and requires tremendous amounts of supportive care. Managing complications can also become expensive. Surgical procedures for impotence and incontinence can result in additional physician and hospitalization costs.

Future Challenges in Prostate Cancer: Basic and Clinical Research

A greater effort must be made to promote earlier diagnosis of prostate cancer when it is localized to the prostate gland and is still curable. In addition, however, it is important to differentiate those patients with latent stage A1 disease who can be left "untreated" from those who will develop progressive and metastatic disease.

The management of patients with localized disease should be resolved through clinical trials in which surgically or laparoscopically staged patients are randomized to receive either radical prostatectomy or radiation therapy.

The optimal timing and type of hormonal therapy should be addressed in clinical investigations. Should an asymptomatic patient with stage D1 disease receive early hormonal therapy? Similarly, does earlier institution of hormonal therapy alter the natural history of prostate cancer, improve survival, or prevent the development of metastatic disease? The development of effective therapies for the hormonally refractory patient should receive high priority; the current cost of caring for these patients is high both in actual dollars and in the emotional strain on patients and their families.

There is very little to suggest that patients with metastatic disease will enjoy an improvement in overall survival. A greater understanding of the molecular biology and genetics surrounding prostate cancer development and the development of both hormonal responsiveness and refractory mechanisms should be a major focus of basic research efforts. It is unlikely that any new, effective treatment modalities will emerge until these mechanisms of hormonal resistance are better understood.