Management of Progressive Metastatic Prostate Cancer
Management of Progressive Metastatic Prostate Cancer
Prostate cancer is the most common noncutaneous malignancy in US men. One in five men will develop prostate cancer in their lifetime, with those over age 60 most at risk. In 1997, an estimated 334,500 new cases will be diagnosed and 41,800 men will die from the disease. At presentation, 11% to 20% of prostate cancer patients have distant metastases (African-Americans being at higher risk), and up to 40% develop metastases during their clinical course.
In over 50 years, the premise on which advanced prostate cancer treatment is based has remained essentially unchanged. Since the seminal observations reported by Huggins and Hodges, the initial management of metastatic prostate cancer has involved testicular androgen suppression. This is accomplished by medical or surgical castration, resulting in tumor regression or stabilization in 60% to 80% of men. The median time to progression is 18 to 24 months and the median survival is 24 to 30 months.
A more controversial issue involves the role of complete androgen deprivation, commonly referred to as maximal androgen blockade (MAB). This can be accomplished by the addition of a nonsteroidal androgen-receptor antagonist, such as flutamide (Eulexin), bicalutamide (Casodex), or nilutamide (Nilandron), which blocks the binding of residual circulating androgens to the androgen receptor. Several large trials have shown an improvement in time to progression and survival compared to monotherapy.[5-7] However, two recent meta-analyses attempting to address this issue were inconsistent. The Prostate Cancer Trialists Collaborative Group reviewed the data of 5,710 patients from 22 randomized trials and found no evidence of a survival advantage for MAB over castration alone. The other report, in abstract form, examined 12 trials comparing MAB to medical or surgical castration alone. The reviewers noted a large subgroup of patients, felt to be terminally ill, among the largest trials not favoring MAB. However, the studies carefully describing masked randomization did reveal a benefit for those who received MAB.
Invariably, almost all patients with metastatic disease will eventually progress. The disease, presently, is referred to as hormone-refractory or androgen-independent and is associated with a dismal prognosis and a median survival of 6 to 12 months after relapse. These terms are misleading because some patients do respond to alternate hormonal therapies. Failed or progressive metastatic disease is a more accurate description of the clinical course. Clearly, for those receiving antiandrogen therapy at the time of progression, the first step is antiandrogen withdrawal (AAWD) while maintaining testicular androgen suppression, followed by a period of observation.
It is far less clear what treatment to initiate in the setting of progressive metastatic disease. Management is best classified as palliative or investigational, as no therapy has shown a survival benefit.
For those patients who demonstrate progression after initial therapy, a subset will not have castrate levels of testosterone. This group may benefit from alternative therapies that completely suppress the serum testosterone to castrate levels. Even low levels of testosterone have the potential to persistently stimulate prostate cancer. Significantly longer survival has been reported in men treated with second-line diethylstilbestrol (DES) who had higher, noncastrate levels of testosterone than in who had appropriate castrate levels of serum testosterone (20.8 vs 4 months). Bilateral orchiectomy has resulted in subjective benefit in up to 70% of patients who progressed after initial estrogen, but the objective response was 15% to 20%.[11,12] Hence, it is recommended that serum testosterone be assessed at the time of initial progression.
For those treated with medical castration (luteinizing hormonereleasing hormone [LHRH] analogs or DES), it is recommended that testicular androgen suppression be continued, as some prostate cancer cells may remain partially androgen-sensitive. Fowler and Whitmore reported that, in 52 evaluable men with symptomatic, metastatic, prostate cancer, 94% experienced subjective and objective worsening after the administration of exogenous testosterone. Furthermore, Silver et al noted a rapid increase in prostate-specific antigen (PSA) and serum testosterone in their patient population, along with clinical worsening after LHRH analog monotherapy was discontinued at the time of disease progression.
Two retrospective reviews examined the response of maintained testicular androgen suppression vs cessation of medical castration at the time of progressive metastatic prostate cancer. Unfortunately, the two study groups reported conflicting data. The Eastern Cooperative Oncology Group (ECOG) reported a modest survival advantage for those maintained on testicular androgen suppression, while the Southwest Oncology Group (SWOG) failed to demonstrate a survival advantage.[15,16] A prospective randomized trial is necessary to resolve this issue definitively, but such a trial is unlikely to be done. Based on current evidence, however, continued testicular androgen suppression seems to be prudent.
At some point, most patients who initially respond to antiandrogen therapy will progress, despite castrate levels of testosterone. A mutation in the androgen-receptor binding domain has been reported in some patients with progressive metastatic prostate cancer. In vitro studies, using the LNCaP prostate cancer cell line, indicate that the mutated receptor can be stimulated rather than inhibited with certain antiandrogens, such as flutamide, androgenic steroids, estradiol, and dihydrotestosterone (DHT).[18,19] Withdrawal of flutamide has been shown to lead to a 50% or greater decline in PSA, improvement of clinical symptoms, or stabilization in approximately 30% of patients (range, 15% to 63%).[20-23] The duration of AAWD response is short, with most investigators reporting a median duration of 3.5 to 5 months.[20-22] At least two studies have shown a trend toward an increased likelihood of response to AAWD in patients who received antiandrogen therapy for an extended period prior to disease progression.[20,21]
Bicalutamide has also been reported to have an AAWD response.[22,23] A large double-blind trial of 813 patients with D2 prostate cancer randomized men to an LHRH analog plus flutamide, 250 mg three times daily, vs an LHRH analog plus bicalutamide, 50 mg/d. Of 150 patients who experienced disease progression, 22 agreed to have their LHRH analog continued while AAWD was undertaken. Of those patients who had flutamide withdrawn, 50% had a 50% or greater decline in PSA within days of drug cessation. The bicalutamide group that underwent AAWD showed a 50% or greater PSA decline in 29% of patients, but their PSA levels continued to rise for up to 7 weeks before declining. This probably reflects, in part, the longer half-life of bicalutamide. For those patients progressing on bicalutamide, AAWD should be followed by a period of observation of at least 2 months after drug cessation.
The list of hormonal therapies showing a withdrawal response is evolving. There are reports of PSA declining after withdrawal of megestrol acetate[25,26] and a number of other antihormone agents. Scher and colleagues recently reviewed published reports of steroidal and nonsteroidal agents withdrawn in relapsing prostate cancer patients. The list of drugs compiled to date showing a withdrawal response upon therapy cessation include: flutamide, bicalutamide, flutamide plus aminoglutethimide, DES, chlormadinone and megestrol acetate. Again, most patients who responded to AAWD had been exposed to the drug for a prolonged period.
Several studies have shown an exaggerated response to flutamide withdrawal when carried out concomitantly with the initiation of another therapy, such as aminoglutethimide. This phenomenon may explain, in part, the 63% response to AAWD reported by Dupont and colleagues, as 29 of their 40 patients were taking hydrocortisone and aminoglutethimide at the time that they were subjected to AAWD.
Aminoglutethimide, an inhibitor of aromatase and adrenal androgen production, was initiated in 29 patients with progressive metastatic prostate cancer. All had progressed while receiving an LHRH analog, suramin, and hydrocortisone initiated as second-line therapy after progression on MAB. The LHRH analog and hydrocortisone were continued, and aminoglutethimide, 250 mg four times daily, was initiated at the time flutamide was stopped. Fourteen (48%) of these patients had a 80% or greater PSA decline for at least 4 weeks.
Subsequently, this same group found that flutamide withdrawal with concomitant introduction of aminoglutethimide led to a 65% response rate in 17 patients who progressed after initial treatment with an LHRH analog, flutamide, suramin, and hydrocortisone for previously untreated metastatic prostate cancer. The median duration of response was 344 days.
Thus, the first therapeutic intervention in patients who progress during therapy with an androgen antagonist to which they previously responded should be discontinuation of the antiandrogen with a period of observation and documentation of progression, particularly prior to enrollment in any study. The frequency and clinical significance of this is unclear, although it is important to recognize this occurrence to avoid false perceptions of a response to subsequent therapy.