27Intermittent Androgen Ablation in Biochemical Relapse of Early Stage Prostate Cancer
DEFINITION OF BIOCHEMICAL RELAPSE
The definition of biochemical relapse (BCR) depends on the type of definitive treatment the patient received for initial management of localized prostate cancer. Since all prostatic tissue is removed by radical prostatectomy (RP), the prostate-specific antigen (PSA) levels should be undetectable by 4 weeks after RP, as the half-life of PSA is about 3 days. If the PSA fails to nadir after RP, this indicates residual prostate tissue either locally or as distant metastases, and is called biochemical failure. The most widely accepted definition of BCR after RP is that of the American Urological Association (AUA) and is defined as a serum PSA greater than 0.2 ng/mL, which is confirmed by a second determination with a PSA of greater than 0.2 ng/mL. In contrast to RP, with radiation treatment, there is still some residual prostatic tissue and the PSA slowly nadirs over a period of 18 months. Also, in intermediate and high-risk disease, after completion of radiation treatment short- or long-term androgen deprivation therapy (ADT) is used, and this can confound the PSA values. The Phoenix criteria are currently used to determine BCR after radiation treatment. A PSA rise of 2 ng/mL or more above the nadir PSA is considered BCR after external beam radiation therapy (EBRT), regardless of whether or not a patient receives short-term or long-term ADT. This is generally confirmed by a repeat measurement of PSA in 3 to 6 months.
28INCIDENCE OF BCR
About 40% of men diagnosed with prostate cancer undergo RP as their primary cancer therapy. About one-third of men undergoing RP will develop detectable PSA levels (BCR) within 10 years after the operation. The BCR rate after EBRT is 40% to 60%. If biochemical failure or recurrence goes untreated, about one-third of patients will develop detectable metastatic disease within 8 years after RP (1). After EBRT, about one-fourth of patients will develop detectable metastatic disease within 5 years and the median time to prostate cancer–specific mortality is 10 years (2).
TREATMENT OF BCR
The approach to treatment of BCR varies, as development of BCR does not necessarily predict development of metastases, and many patients may have indolent disease that does not require treatment. The decision to start systemic therapy, that is ADT, is influenced by a number of prognostic factors, namely age, underlying comorbidities, psychosocial factors/patient anxiety, tumor characteristics at time of initial definitive therapy (Gleason’s score, stage, PSA level, margin status), absolute PSA level at the time of BCR, and PSA doubling time (PSADT). The most important factors are a short PSADT and high Gleason’s score (3,4).
• If the PSADT is greater than 12 months, and the patient is elderly, observation may be considered.
• If there is no evidence of distant metastatic disease clinically or radiographically, then some patients (stage T1–T2, N0 or NX, PSA now < 10 ng/mL, life expectancy > 10 years) are candidates for salvage local treatment, which may include radiation after failed RP or cryosurgery after failed EBRT.
• ADT is considered when the risk of systemic disease is high based on the pathologic and clinical factors, and hence not treatable by salvage local therapy; or if the patient is not a candidate for local treatment due to underlying comorbidities or patient preference.
29ANDROGEN DEPRIVATION THERAPY
Once the decision to give ADT for treatment for BCR has been made, the timing of when to start ADT presents a therapeutic dilemma. There is no consensus to the definition of early ADT versus late ADT (i.e., at what PSA level should ADT be initiated). Early ADT may prevent progression and hence may prolong survival, though there is no clear evidence of survival advantage by this approach. Delayed ADT until clinically overt metastasis is seen may be advocated, as ADT is associated with significant side effects and there is no clear survival advantage (5,6). Hence, treatment should be individualized until more evidence is available to support one approach over the other. Earlier systemic therapy is recommended for younger age, high grade disease (Gleason’s score of 8–10), and shorter PSADT of less than 10 to 12 months.
Options for ADT include complete androgen blockade versus monotherapy. There are no randomized trials comparing complete androgen blockade (Gonadotropin-releasing hormone [GnRH] analogue + antiandrogen) to either GnRH or orchiectomy alone. Most of the studies looking at treatment of BCR with ADT used complete androgen blockade (5).
INTERMITTENT VERSUS CONTINUOUS ADT
Long-term ADT is associated with significant side effects including but not restricted to loss of libido and potency, hot flashes, and loss of bone and muscle mass. The rationale for trying intermittent (as opposed to continuous) ADT is 2-fold: to decrease these side effects and hence improve quality of life (QOL); and theoretically, continuous ADT may facilitate progression from androgen dependence (castrate-sensitive disease) to androgen independence (castrate-resistant disease), which may be delayed by use of intermittent ADT.
Intermittent ADT involves use of cyclic ADT: ADT is initiated and continued till serum PSA is lowered to a certain predefined level. This is followed by temporary withdrawal of ADT. Another cycle of ADT is restarted when serum PSA increases to a predefined level (usually 2.5–20 ng/mL) or serum testosterone is no longer at castrate levels (>20–50 ng/mL). Patients who are on intermittent ADT may be off ADT 35% to 50% of the time.
30The most extensive data for intermittent androgen deprivation (IAD) come from the Canadian led PR 7 trial (7). In this international randomized phase 3 trial, 1,386 men with a rising PSA greater than 3 ng/mL after RT were randomized to receive either IAD or continuous ADT. Patients assigned to IAD received therapy for 8 months and treatment was restarted when the serum PSA reached greater than 10 ng/mL off treatment. The primary end point was overall survival. The study was designed to show noninferiority. At a median follow-up of 6.9 years, the intermittent approach was noninferior to the continuous approach with respect to overall survival (median 8.8 vs. 9.1 years, hazard ratio [HR] 1.02, 95% confidence interval [CI] 0.86–1.21). Men treated with IAD had more prostate cancer–related deaths (120 out of 690 patients) than in the continuous ADT arm (94 of 696 patients) but this was balanced by more non-prostate-cancer-related deaths in the continuous ADT arm. QOL analyses showed statistically significant modest benefits in the IAD group when it came to symptoms of loss of libido, urinary symptoms, and hot flashes. However, only 214 out of 1,386 (15%) patients had died when the results of the trial were analyzed and longer follow-up may be required for disease-specific deaths to outbalance deaths by other causes. An unplanned Cox regression analysis showed that men with a Gleason’s score greater than 7 had a 14-month increased survival with continuous ADT (8.0 years) as compared to IAD (6.8 years). Hence, IAD should be offered to patients only after explaining the full risks of potential harm versus benefits from improved QOL.
Currently, the American Society of Clinical Oncology (ASCO) recommends continuous ADT over IAD, due to lack of sufficient long-term evidence (5). However, National Comprehensive Cancer Network guidelines recommend consideration of intermittent androgen deprivation (8).