COMPLICATIONS OF ANDROGEN-DEPRIVATION THERAPY IN PROSTATE CANCER: THE OTHER SIDE OF THE COIN
Claude Schulman, Department of Urology, University Clinics of Brussels, Erasme Hospital, Belgium.
(intermittent) androgen-deprivation therapy
It has been 67 years since Huggins and Hodges published their Nobel Prize-winning article showing the hormone dependence of prostate cancer. Andrew Schally described the structure and principle of LHRH and was rewarded in 1977 with a Nobel Prize. This led to the development of LHRH agonists, which since then, have become in most countries the standard androgen-deprivation therapy (ADT) for use in patients with advanced and metastatic disease and is recommended by many international guidelines on prostate cancer. Nevertheless, ADT remains a palliative treatment with limited effect on survival with almost all patients dying from prostate cancer having hormone-refractory disease with castrate levels of testosterone.
ADT: LACK OF EVIDENCE
The British Medical Research Council conducted a trial in >900 M0/M1 patients with prostate cancer. Early results suggested a survival benefit in favour of immediate vs delayed hormonal treatment for M0 patients only . However, the overall survival was not significantly longer on follow-up. A prospective randomized Swiss trial (SAKK) also did not show any improvement in overall survival . The recently published European Organization for Research and Treatment of Cancer (EORTC) 30891 trial compared immediate ADT (with orchidectomy or LHRH) vs deferred ADT until symptomatic progression in 985 patients without overt metastatic disease not eligible for radical treatment . The trial showed a modest increase in overall survival in favour of immediate ADT but could not demonstrate any improvement in cancer-specific survival. However, only 20% of the patients died of prostate cancer in both groups after a median follow-up of 7 years. In addition, 25% of the patients in the deferred arm died of unrelated causes without ever receiving any hormonal treatment. Another recent retrospective study including 19 271 men could not show an improved survival among most of the elderly men with localized prostate cancer receiving primary ADT when compared with conservative management . Considering the use of antiandrogen monotherapy, the Early Prostate Cancer trial clearly showed that patients with localized disease receiving bicalutamide (150 mg per day) had no improvement in progression-free survival and that there was even a trend towards decreased overall survival (Hazard ratio 1.16; P = 0.07) compared with patients receiving placebo .
COMPLICATIONS OF ADT
In the last decade, pharmacological ADT has been increasingly used in the earlier stages of prostate cancer, despite a paucity of evidence from large-scale randomized trials suggesting a survival advantage from earlier intervention. ADT may eventually improve cancer-specific survival in a defined subset of patients; nonetheless, it is associated with complications in almost all patients. The well-known symptomatic side-effects of castration include loss of libido/erectile dysfunction, fatigue, hot flashes, gynaecomastia and also cognitive dysfunction, depression, loss of muscle mass and increase of fat mass, which compromise physical function, independence, and quality of life. Furthermore, ADT reduces bone mineral density and increases the risk of fractures in men with prostate cancer, another possible cause of death. The 1-year mortality ratio of elderly men with hip fracture is nearly four times that in women . Bone mineral density loss is rapid during the first year of ADT and longer duration of treatment confers greater fracture risk. Indeed, men receiving ADT ≥1 year increase their fracture risk from 10% to 16%.
METABOLIC SYNDROME AND CARDIOVASCULAR (CV) MORTALITY
Besides these well-recognized side-effects of ADT, more recent accumulating evidence indicates that patients treated with ADT have an increased risk of dying from CV disease and developing metabolic syndrome, thus affecting overall survival of patients. The five diagnostic criteria for metabolic syndrome are: a fasting plasma glucose level of >110 mg/dL, serum triglyceride level of ≥150 mg/dL, serum high-density lipoprotein level of <40 mg/dL, blood pressure of ≥130/85 mmHg, and waist circumference of >102 cm. Middle-aged men with metabolic syndrome are three times more likely to die of coronary heart disease and other CV problems even after risk factor adjustment .
The prevalence of metabolic syndrome in men treated with ADT was shown by Braga-Basaria et al.  to affect more than half of men. Keating et al.  analysed a large observational study of a population-based cohort of 73 196 men (SEER), aged ≥66 years with loco-regional prostate cancer treated with GnRH analogues or orchidectomy. Men were observed for up to 10 years with a median of 4.6 years. The authors showed an association of GnRH analogue treatment with an increased risk of diabetes (44%; P < 0.001), coronary heart disease (16%; P < 0.001), sudden cardiac death (16%; P < 0.004), and myocardial infarction (11%; P = 0.03). Moreover, the increased risk of coronary heart disease and diabetes was apparent with as little as 1–4 months’ treatment. However, men treated with orchidectomy did not present any increased risk of CV disease (all P > 0.20) but only in developing diabetes (34%; P < 0.001). These correlations between ADT and CV disease were confirmed by Saigal et al.  who used the SEER Medicare database, with 22 816 subjects. A multivariate model was used to calculate the risk of subsequent CV morbidity in men with prostate cancer who where treated with ADT. In those patients receiving ADT there was a 20% increased risk of CV morbidity, and this was seen within the initial 12 months of ADT. Pooled results from three randomized trials of radiation therapy with or without ADT including 1372 men were analysed by D’Amico et al. . They found that men who received 6 months of ADT had shorter times to fatal myocardial infarction compared with men who did not receive ADT (P = 0.017). Furthermore, in men aged ≥65 years, there was no significant difference in the time to fatal myocardial infarction when comparing 3 months of ADT to 6–8 months of ADT, thus indicating an increased risk of CV complications even after short-term ADT. Lastly, Efstathiou et al.  assessed the relationship between duration of GnRH therapy and CV mortality in a large randomized trial including 1554 men with locally advanced prostate cancer receiving external beam radiotherapy with or without ADT. After a median follow-up of 8.1 years, multivariate analyses did not show a difference in CV mortality for men receiving short-term therapy (4 months) compared with longer-term ADT (28 months). These results seem to indicate that metabolic effects and CV complications occur early after initiating ADT.
SIMILARITIES WITH HYPOGONADISM
All the above studies reporting an association of metabolic syndrome and CV morbidity with ADT should not be surprising, as hypogonadism has been known to be associated with similar complications. Indeed, low testosterone is inversely associated with concentrations of insulin, glucose and triglycerides, and positively associated with levels of high-density lipoprotein cholesterol. Stellato et al. examined a possible association between low testosterone and the development of type 2 diabetes mellitus in men aged 40–70 years in the Massachusetts Male Aging Study over a mean follow-up period of ≈9 years. The authors reported that baseline free testosterone and sex hormone-binding globulin levels were significantly lower in men who subsequently developed type 2 diabetes. Furthermore, a study by Rodriguez et al.  analysing a cohort of 618 men in the Baltimore Longitudinal Study of Aging also linked lower baseline androgen levels to a higher incidence of metabolic syndrome. Finally, low testosterone was also shown to be a predictive marker for a high risk of CV disease in a prospective study including 11 606 men. Men with the highest levels of testosterone had a 41% reduction in CV mortality. In addition, an increase of testosterone by each 6 nmol/L was associated with an 19% reduction in mortality risk .
There is no evidence that ADT as monotherapy is beneficial at all stages of prostate cancer. Nowadays, ADT is predominantly prescribed in asymptomatic patients and these recommendations are not supported by evidence-based data. Most recent studies clearly indicate that CV complications are of significant concern in patients with prostate cancer receiving ADT. Before initiating treatment, patients with prostate cancer should carefully be selected as the eventual slight survival benefit in patients with prostate cancer is negatively counterbalanced by increased CV mortality leading to no significant difference in survival. Studer et al.  suggested that patients, not suitable for radical treatment, with a baseline PSA level of >50 ng/mL are more likely to die from prostate cancer and may be considered as candidates for immediate ADT to delay complications from progressive disease. In addition, patients with a PSA level of 8–50 ng/mL at diagnosis and PSA doubling time of <12 months could potentially benefit from immediate ADT. However, patients with a PSA level of ≤8 ng/mL are at very low risk of dying from prostate cancer within 7 years after diagnosis and should therefore not receive early ADT and benefit from active surveillance, avoiding the side-effects of ADT.
Another important issue with ADT is the development of androgen-independent prostate cancer. Most patients with metastatic disease will develop within 2–3 years of ADT, disease progression and lineage of hormone-independent cancer cells. ADT could provide selective pressure between heterogeneous mixture of prostate cancer cells and thus leading to development of androgen-independent cancers . Preclinical evidence suggests that the use of hormonal therapy on an intermittent basis might delay androgen independence and hormonal resistance. Another benefit of intermittent ADT (IADT) includes attenuation of side-effects through the ‘off-period’ treatment. Unfortunately, there is still insufficient data from phase III randomized controlled studies and confusion exists on which patient might benefit from IADT. Boccon-Gibod et al.  reviewed recent studies of IADT and proposed potential target groups benefiting of IADT, showing comparable results with continuous ADT.
Increased use of ADT in the earlier stages of prostate cancer is lacking sufficient evidence and may reflect in great part influence from the pharmaceutical industry. Should surgical castration have remained the standard of ADT, most probably, most of the increased use would not have been considered in the same way. Financial aspects might have influenced the use of LHRH agonists. Indeed a recent study in the USA reported a decline in the prescription of LHRH and an increase in surgical castration after a decrease in reimbursement . Patients should receive ADT monitoring and therapy aimed at preventing or treating ADT- induced complications, focusing on metabolic syndrome (high blood pressure, insulin resistance, central obesity, high triglycerides and low high-density lipoprotein-cholesterol). Moreover, patients undergoing ADT should be controlled for CV risk factors before therapy as CV toxicity occurs early after initiating treatment. Precautionary measures should be given such as refraining from smoking, having regular physical exercise, doing ‘cerebral training’, having a balanced diet, and blood pressure, glucose and serum lipids monitoring.
The authors are grateful to Dr Addla Sanjai for linguistic support.
CONFLICT OF INTEREST