Safety and tolerability of intermittent androgen deprivation therapy: A literature review


Damian Gruca M.D., Abbott, Knollstrasse 50, 67061 Ludwigshafen, Germany. Email:


Androgen deprivation therapy is commonly used in men with advanced prostate cancer; however, it is associated with many short- and long-term side-effects. Intermittent androgen deprivation therapy was first suggested as an alternative regimen in the early 1990s and is now part of treatment guidelines as a result of its ability to reduce adverse events associated with continuous androgen deprivation therapy without decreasing its efficacy. Although many publications evaluated intermittent androgen deprivation therapy's efficacy, the safety and tolerability information of this regimen is relatively limited. The goal of this literature review was to analyze clinical trials that have reported safety and tolerability data in prostate cancer patients treated with intermittent androgen deprivation therapy, as well as assessing quality of life outcomes. A literature search was carried out using biomedical and pharmaceutical databases for published information comparing intermittent androgen deprivation therapy with continuous androgen deprivation therapy. A total of 13 randomized and non-randomized studies were selected and reviewed based on their relevance to the safety, tolerability and quality of life of intermittent androgen deprivation therapy. Benefits for intermittent androgen deprivation therapy were observed for the short-term side-effects (hot flushes and sexual functions) mainly during the off-treatment phase, whereas the data for the long-term side-effects were not as conclusive. Quality of life evaluations are more in support of intermittent androgen deprivation therapy. Although there are some safety, tolerability and quality of life benefits associated with intermittent androgen deprivation therapy, the overall evidence is still limited.

Abbreviations & Acronyms

androgen deprivation therapy

AE =

adverse events


American Urological Association


bone mineral density


body mass index


continuous androgen deprivation therapy


Cancer of the Prostate Strategic Urologic Research Endeavor


European Organization for Research and Treatment of Cancer


Food and Drug Administration

GnRH =

gonadotropin-releasing hormone

Hb =



high-density lipoprotein

HR =

hazard ratio


intermittent androgen deprivation therapy


low-density lipoprotein

NA =

not available

OS =

overall survival

PCa =

prostate cancer


progression-free survival


prostate-specific antigen

QoL =

quality of life


serious adverse events


South European Uronocological Group


time to progression


PCa cell growth is regulated by endogenous androgens (e.g. testosterone), and ablation of these hormones through surgical castration or pharmacotherapy is crucial for the treatment of this disease.1 ADT is commonly used in approximately 50% of men2 with advanced PCa to lower levels of androgen directly or to block androgen–androgen receptor binding.3

Potent reduction of androgens is achievable through the pharmacological inhibition of the hypothalamic–pituitary–gonadal axis using a GnRH agonist, such as leuprorelin, goserelin, busereline, triptoreline or histrelin, or a GnRH antagonist, such as degarelix or abarelix.4–6 Traditionally, GnRH agonist therapy was given continuously; however, multiple studies have reported the beneficial effects of intermittent ADT, a cyclic therapy consisting of an on-treatment period (normally fixed) lasting 6–9 months followed by a treatment interruption or off-treatment period of variable length, dependent on PSA response.7 IAD is an attractive option focusing on reducing side-effects of ADT8–14 that are usually observed in CAD (Table 1). In many cases, early side-effects (hot flushes and sexual dysfunction) are immediately reported; however, long-term consequences (anemia, bone loss, cardiovascular and metabolic changes) are not as easily detected by either the patient or physician, and yet these side-effects are crucial to the management of PCa during ADT.

Table 1.  Common AE of CAD
Early side-effects
Sexual dysfunction 15–18
 Libido loss↑ up to 95% 
 Erectile dysfunction↑ up to 80% 
Hot flushes50–80%18–20
Fatigue14%8, 21
Long-term side-effects
Anemia90% have 10% Hb↓18, 22
Bone loss88% if ADT > 1 year23
Metabolic changes  
 Fat mass↑ by 11.2% ± 1.5% (P < 0.001)13, 24
 Triglycerides↑ by 19.4% ± 9.7% (P = 0.039) 
 HDL↑ by 9.7% ± 3.6% (P = 0.002) 
 LDL↑ significantly (P = 0.03) 
 Insulin sensitivity↓ significantly 
 Insulin fasting level↑ by 25.9% ± 9.3% (P = 0.04) 
Cardiovascular disease 10
 Coronary heart disease↑ (HR = 1.16, P < 0.001) 
 Myocardial infarction↑ (HR = 1.11, P = 0.03) 
 Sudden cardiac death↑ (HR = 1.16, P = 0.004) 
Cardiovascular disease 25
 Myocardial infarctionNo risk ↑ (HR = 0.91; 95%CI, 0.84 to 1.00) 
 Sudden cardiac deathNo risk ↑ (HR = 0.96; 95%CI, 0.83 to 1.10) 

A sequence of studies suggested that IAD was a reasonable management option,26,27 as patients observed over 21–47 months had stable disease and full PSA responses with each consecutive suppression of testosterone.26 A study supporting IAD for PCa treatment was that of Goldenberg et al., who reported that PSA levels can act as a marker for determining eligibility and guiding the IAD regimen.27 It was, however, recommended that the patient's disease stage and the PSA nadir ultimately guides hormonal therapy.7,28

Disease stage and PSA levels are interconnected and play a crucial role in IAD; however, the PSA level that determines when treatment should resume is based empirically, and this guidance was reflected in observations by Goldenberg et al. and Tunn et al.27,29 Recently, Sanchez-Salas et al. presented preliminary data, from a European study in patients with biochemical recurrence of localized PCa, at the annual AUA 2011 meeting. The investigators concluded that the favorable survival rate combined with the few side-effects observed contributed to the efficacy of IAD.30 Although the first goal of IAD to reach comparable efficacy versus CAD seems to be achieved, little has been published about the second goal of IAD, namely to reduce the side-effects of CAD. We have taken this opportunity to compare the safety and tolerability data from PCa patients treated with IAD, and to evaluate their data during on-treatment and off-treatment periods, with those on the CAD regimen from a selected number of recent studies.


A literature search was carried out using the following search criteria terms: IAD, intermittent androgen suppression, intermittent hormonal treatment, side-effects, metabolic side-effects and cardiovascular risk; and biomedical/pharmaceutical databases (MEDLINE [R] 1950–2010/Aug 24, Derwent Drug File 1983–2010/Jul W4, Derwent Drug File 1964–1982, Pascal 1973–2010/Aug W4, Biosis Previews [R] 1926–2010/Aug W4, Int.Pharm.Abs 1970–2010/Jul B2, EMBASE 1974–2010/Aug 26, SciSearch [R] Cited Ref Sci 1990–2010/Aug W4) for published information comparing IAD with CAD. References were not limited by language nor publication date, and studies were chosen based on their focus on QoL measures, side-effects or long-term consequences of ADT. Dosing regimens, drug types used for ADT, patient populations, severity of diseases and efficacy parameters were also not part of our selection criteria. The majority of data evaluated for the present review were obtained from large, randomized, actively controlled studies; however, some smaller trials were included, as they provided additional information with regards to the safety and tolerability of IAD. The reviewed studies had various designs, from single-arm observational trials of IAD in PCa patients qualifying for hormonal treatment to extensive double-arm, actively-controlled randomized studies comparing in parallel IAD and CAD regimens. For studies that had only an IAD arm, the on-treatment period's data was categorized in the same group as the CAD data, whereas the off-treatment period was considered similar to IAD.

From our literature search, physical side-effects were categorized into those that occurred soon after the start of ADT, termed as “early side-effects”, and those that occurred after long-term ADT of at least several weeks or months duration.


We identified and reviewed publications from 25 IAD studies; of these, 13 were selected for the present literature review based on their relevance to assess safety and tolerability of IAD (Tables 2,3). Of these 13 studies, eight were randomized, active-controlled studies comparing IAD with CAD (Table 2; n = 68 to 1386 patients),28,29,31–39 and five were observational, non-randomized studies (Table 3; n = 19 to 250 patients).9,40–43 Of these eight active-controlled studies, two31,32 were published in full, whereas preliminary results have been reported in abstract format for the remaining six studies.29,33,35–38

Table 2.  Published randomized studies that addressed tolerability and safety aspects of IAD
ReferencenPrimary end-pointDuration of follow upStudy designPatient populationMajor findings
  1. †Number of evaluable patients. ‡Validated QoL questionnaire used in the analysis (QLQ-C30 or disease specific questionnaire QLQ PR24).

31626Time to subjective or objective progression/Median follow up at 51 months Maximum 12 yearsPhase 3 randomized active controlled (SEUG9401) studyLocally advanced/metastatic hormone naïveNo difference in survival in both treatment arms
No clinically meaningful QoL differences‡
3268Time to androgen-independent PCaMean follow up at 30.8 monthsRandomized open-label – active controlled studyLocally advanced/metastatic/recurrent hormone naïve3-year overall progression rate was lower in the IAD than in the CAD – androgen independent progression rate lower in IAD group vs CAD
– less serious side effect reported in IAD
28, 29, 33184Clinical or PSA progression under treatmentMean follow up at 28.4 monthsPhase 3 randomized active controlled (EC507) studyPSA relapse after radical prostatectomyNo difference in TTP or mean PFS – better QoL under IAD‡
341386OSNAPhase 3 randomized active controlled (NCIC CTG PR.7) studyPSA relapse after radiationImproved time to hormone refractory stage in the IAD arm and no difference in overall survival between both arms – more PCa related & less unrelated deaths in IAD arm
35335Time to clinical or biochemical progressionNARandomized active controlled studyAdvanced or locally advanced Patients with PCaNo difference in TTP or PFS
– patient's self assessment on overall health & sexual activity better in IAD group – no differences in AE/SAE or other safety parameters between groups
36, 37193Time to PSA escape or clinical progressionMean time to follow up at 66 monthsRandomized active controlled (TULP) studyLocally advanced/metastatic PCaNo difference in OS, but a trend for longer TTP in CAD
– no differences in safety or QoL‡
38366Clinical or PSA progressionNARandomized controlled studyMetastatic PCaQoL scores were better for physical and emotional functions in IAD, but worse for cognitive‡
39173Overall survivalBoth groups were followed every 3 months until progressive diseaseRandomized controlled studyMetastatic PCa with bone metastasesEquivalence of both treatments in terms of OS and median PFS
Table 3.  Published non-randomized studies that addressed tolerability and safety aspects of IAD
ReferencenPrimary endpoint§Study designPatient populationMajor findings
  1. †Number of evaluable patients. ‡Validated QoL questionnaire used in the analysis (QLQ-C30 or disease specific questionnaire QLQ PR24). §Follow Up Duration was not available. ¶Patients were evaluated at baseline by bone scan, computed tomography of the abdomen and pelvis, chest X-ray, and BMD determination by dual energy X-ray absorptiometry scan.

4095Efficacy of IAD and SAE recovery & prevalencePhase 2 studyMinimally metastatic or recurrent PCa– Sexual potency regained in 47% during off-treatment intervals;
– hemoglobin recovery in 50% of cycles
– progressive osteopenia; not improved BMD by IAD
– no change in BMI
4143Feasibility of IADProspective non-randomized studyMetastatic PCaLess SAE in off-treatment periods, but no QoL differences measured by QLQ-C30 questionnaire‡
42109QoL, morbidity & mortalityProspective non-randomized studyLocally advanced, recurrence after radiationSignificant improvement of QoL during off-treatment periods§
919BMD changesNABiochemical relapse PCaA decrease in bone mineral density from baseline after 9 months of ADT. The off-treatment period slowed down the bone loss, but did not balance it to the baseline BMD values¶
43250QoLLongitudinal studyLocally advanced or recurrentIn off-treatment periods the number of SE decreased and QoL increased‡

Early side-effects

Early side-effects of ADT usually occur immediately or soon after the reaching of the testosterone suppression. Hot flushes and sexual dysfunction are among the most common early side-effects described during ADT, and are important factors impacting patients' quality of life.

Hot flushes

Hot flushes were evaluated among the early side-effects of ADT in five of the 13 studies reviewed. Among these, the Calais Da Silva et al. South European Uronocological Group study31 reported that in a 2008 analysis in patients with a maximum follow up of 7 years and a median follow up of 2 years, hot flushes were reported in 7% of patients in the IAD arm compared with 23% in the CAD arm (P < 0.0001).44 Hence, the occurrence of hot flushes was approximately threefold greater in PCa patients receiving CAD than in those receiving IAD. An analysis carried out 1 year later, in the same population with a maximum follow up of 12 years and a median follow up of 4.25 years, observed hot flushes in 19.7% of patients in the IAD group and 30.0% in the CAD group (P < 0.01).31 Similar findings were noted by de Leval et al. Bruchovsky et al. Bouchot et al. and Spry et al.32,41–43 In these studies, hot flushes were reported as mild to moderate during IAD treatment,32 and were resolved or improved in the off-treatment periods.29,42,43 In the Bouchot et al. study, hot flushes improved in the first IAD off-treatment cycle from a score of 52 to a score of 78 (100 = baseline), and from a score of 44 to a score of 62 (P < 0.01) in the second cycle.41 Furthermore, Bruchovsky et al. reported that hot flushes decreased from 8.3% in the on-treatment period to 4.6% in the off-treatment (P = 0.1).42 The de Leval study did not provide additional insight or suggest which ADT increased the incidence of hot flushes.32 Based on all available data, the frequency of hot flushes in IAD patients was reduced from 34.3% to 69.9%. Similarly, the severity of hot flushes was positively affected and showed a relative improvement of 29% to 33.3%. These benefits for IAD patients were mainly observed in the off-treatment phase (Table 4).

Table 4.  Safety and tolerability overview from IAD studies
  1. †The de Calais references cited are two separate publications that report the same end-point and patient population, but at different points in time.

Hot flushes     
 626†7.0%23.0%P < 0.0001; maximum follow-up 7 years44
 62619.7%30.0%P < 0.01; maximum follow-up 12 years31
– no difference in OS
 68Not reportedNot reportedMost patients had slight to moderate levels of hot flushes that resolved in the majority of IAD patients who stopped receiving ADT32
 1386Not reportedNot reportedIAD patients reported less hot flushes34
 1094.6%8.3%P = 0.1; improvement of hot flushes in QoL scale (80.3 in on- vs 88.3 in off-treatment period)42
 43Not reportedNot reportedImprovement of hot flushes in QoL scale reported (52 of 100 in on-treatment period vs 78 in the off-treatment period)41
 250Not reportedNot reportedQoL score change for hot flushes ≥44.5 during on- & −30.1 during off-periods43
Summary of relative differenceFrequency: 34.3% to 69.9%  
Severity: 29.0% to 33.3%
Sexual activity     
 62628.0% active patients10.0%P ≤ 0.0131
 1090.9% patients impotent5.5% patients impotentP = 0.0342
 250Not reportednot reportedP < 0.01; −15.8 score change on and +8.4 during the off-treatment period43
 95Potency regained in 47%Not reportedMen with normalized testosterone values were more likely to recover their sexual activity (64 vs 50%; P = 0.51)45
Summary of relative differenceFrequency: 23.3% to 64.2%  
Bone loss     
 95Not reportedNot reportedBMD assessed at 43 months (mean) showed osteopenia/osteoporosis in 37% of patients in at least one site and 8–31% in each of the sites evaluated40, 45
 19Not reportedNot reportedBMD loss was attenuated but did not stop with IAD9
 1386Not reportedNot reportedNo difference in osteoporotic fracture rates34
 95Not reportedNot reportedAnemia was present in 4% at baseline, but increased to 20–50% later on. Only 50% of patients showed recovery after stopping treatment45
 1092.8%1.8%P = 0.5; on-treatment period vs off-treatment period42
 95Not reportedNot reportedWeight gain during on-treatment periods was balanced by weight loss during off-treatment periods40, 45
 250Not reportedNot reportedWeight gain during on-treatment periods (+9.8 QoL score points; P < 0.01) compared to baseline and weight loss (−4.6 points; P < 0.05) in the off-treatment period43
 1098.3%2.8%P = 0.6; not significant42
Cardiovascular diseases     
  2.8%4.6%P = 0.4; myocardial infarction; 
 1095.5%0.9%P = 0.6; cerebrovascular accidents42
  3.7%1.8%P = 0.3; deep vein thrombosis; 
 626Not reportedNot reportedP = 0.21; CAD patients were more likely to die of cardiovascular diseases or other causes31
 1386Not reportedNot reportedNo differences in myocardial infarctions and other vascular events34
Quality of Life     
 43Not reportedNot reportedNo differences41
 173Not reportedNot reportedNo differences39
 626Not reportedNot reportedNo differences31
 193Not reportedNot reportedNo differences36
 335Not reportedNot reportedNo differences35
 184Not reportedNot reportedImproved QoL in the IAD group28, 29, 33
 109Not reportedNot reportedImproved QoL scores in 11 modules through a shift of results towards baseline data42
 250Not reportedNot reportedQoL improved in the off-treatment periods43
 366Not reportedNot reported2 out of 5 scores were better in IAD group than CAD, 1 was worse (cognitive function) and other 2 were not different in both groups38

Sexual dysfunction

Improved sexual dysfunction was observed in six of the 13 studies.32,35,40,42,43,45 In 2002, de Leval reported that the loss of libido and erectile dysfunction, which occurred during CAD, was improved during IAD.32 Similar observations were reported by Malone et al., who noted that libido and erection were generally lost during the on-treatment periods of IAD; however, 47% of patients were able to regain sexual function in the off-treatment periods, and this correlated with testosterone level normalization.40 These findings were consistent with those of Miller et al., who reported that patients' self assessments of their overall health and sexual activity were more favorable with the IAD treatment group compared with those receiving CAD,35 and later by Calais da Silva et al.31 The study by Calais da Silva reported that men treated with IAD had better sexual function (activity and libido) when compared with those receiving CAD. Furthermore, 15 months after randomization, 28% of patients who had received IAD reported sexual activity compared with 10% of the CAD patients. In both arms, the sexual activity decreased, but in the CAD group, this decrease was significantly greater (P ≤ 0.01).31 Finally, improvement in sexual function, captured by QoL questionnaires, also showed enhanced sexual activity in patients receiving IAD during the off-treatment period in studies by Spry et al. and Bruchovsky et al.42,43 Sexual activity was decreased in both groups during the study; however, patients receiving IAD had significantly better scores than their CAD counterparts (P < 0.01).31,44 Overall, these studies showed that IAD therapy, especially during the off-treatment periods, maintained or restored sexual activity, libido and satisfaction from 23.3% to 64.2% (Table 3).

Long-term consequences and risks

In contrast to the early side-effects of ADT, which are observed almost immediately after the initiation of treatment, there is a group of side-effects connected with ADT that occur later, and are referred to in the present review as long-term consequences. Bone loss, obesity and cardiovascular diseases are the most common long-term consequences, and their rate of occurrence during IAD is analyzed below.

Bone loss

Of the 13 studies reviewed, three examined BMD and bone loss in PCa patients.9,34,40 Higano et al. carried out an observational non-randomized study in hormone-naive patients with prostate cancer, and reported significant BMD decrease in patients receiving IAD. Mean (± SEM) baseline lumbar spine BMD decreased from 1.10 ± 0.04 g/cm2 to 1.05 ± 0.04 g/cm2 after 9 months of treatment with a total of 4.5% ± 1.0% (P = 0.0007) decrease from baseline. A 2.5% decrease in BMD from baseline was also measured at the hip; however, this observation was less prominent and not significant. In the same study by Higano et al., the rate of bone loss did decline during the off-treatment period of 7.9 months, but BMD values did not return to those originally measured at baseline. Post-CAD, BMD values at the lumbar spine were 1.5%; P = 0.06, and at the hip −0.01%; P = 0.09.9 Malone et al. carried out a study in 95 IAD patients and reported that osteoporosis was seen in 37/41 patients who were evaluated at least at one site. Furthermore, serial bone scans showed progressive osteopenia.45 Recently, Klotz et al. reported the strongest clinical observation from a randomized trial comparing patients receiving IAD (n = 690) to CAD (n = 696) patients. That study found no evidence of differences in osteoporotic fractures between regimens.34


Although anemia was not a key side-effect monitored in the 13 studies reviewed, a study by Malone et al. did assess anemia in on-treatment periods of patients with PCa receiving IAD.40,45 Anemia caused by ADT is a normachromic and normocytic event; however, the mechanism of action is currently unclear. In this non-comparative study, mild anemia was assessed during each treatment cycle (C) and defined as hemoglobin <130 g/L. These IAD patients showed anemia rates of 33% in C1, 44% in C2 and 67% in C3, respectively. Hemoglobin recovery was seen in 50% of these cycles, and observed more often when testosterone levels did not fully normalize after ADT was stopped (47% vs 14%, P = 0.01).40,45 Bruchovsky et al. also evaluated anemia as one of the AE, and showed that it appeared in 4.6% of patients studied. Anemia was also observed during on-treatment periods in 2.8% patients compared with 1.8% of patients during the off-treatment period (not statistically significant; P = 0.5).42 Other IAD studies did not report anemia or any differences in hemoglobin levels between the treatment groups.


Changes in BMI were end-point measurements in three of the 13 studies outlined in Table 2.40,42,43,45 Malone et al. observed that the typical ADT-driven weight gain that normally occurs during the on-treatment period appeared to be balanced by the off-treatment weight loss. Hence, PCa patients in this study did not experience a significant change in overall BMI by the end of the IAD treatment period.40,45 In this study, 14% of patients had an increase in BMI of more than 10% during the on-treatment, but just 6% of patients had a decrease in BMI > 10% when off treatment, making the observed trend of BMI change inconsistent.40,45 Also, Bruchovsky et al. reported weight gain during the on-treatment and weight loss during the off-treatment periods. Theses differences were not statistically significant (P = 0.06).42 However, patients evaluated by Spry et al. experienced significant weight gain during the on-treatment period (+9.8 score points; P < 0.01) compared with baseline (measured by the QoL questionnaire). Weight gain was, however, decreased by 4.6 score points during the off-treatment period.43 To date, results from a randomized study by Klotz et al. containing metabolic changes and lipid alterations as secondary end-points are pending.34

Cardiovascular disease

Cardiovascular diseases were examined in three of the 13 studies evaluated.31,34,42 Of these three, only the Bruchovsky et al. study reported a significant difference in the percentage of cardiovascular events when comparing on and off-treatment periods. Patients had an increased incidence of myocardial infarction during off-treatment period (4.6%) than during the on-treatment IAD period (2.8%). In contrast, the incidence of cerebral vascular accidents was higher during therapy (5.5%) than during the off-treatment period (0.9%), and deep vein thrombosis occurred at similar frequency during both periods.42 Both the Calais da Silva et al. and Klotz et al. studies noted increased risk of dying from cardiovascular disease in the CAD group,31 or having myocardial ischemia/infarction in both treatment arms (10% in IAD vs 11% in CAD arm);34 however, these observations were not statistically significant. Further to this, Klotz et al. reported that patients in the IAD arm were 9% more likely to die from PCa than those in the CAD arm, who had an 8% higher chance of dying from other causes.34


QoL was assessed as an end-point parameter in 64% (9/13) of the studies evaluated. Of the nine studies that evaluated QoL, four showed a QoL benefit in patients treated with IAD,28,38,42,43 whereas the remaining five did not show any significant difference between IAD and CAD.31,35,36,39,41

In 2006, Spry et al. carried out a study that assessed changes to QoL and testosterone dependent physiology in men receiving IAD. In this study, QoL data showed that testosterone suppression resulted in decreased global QoL, and deterioration in most function and symptom scales. However, a trend of gradual improvement in QoL was observed during the off-treatment period, and this correlated with testosterone recovery. Recovery of QoL was slower than the rate of deterioration during the treatment phase. Furthermore, testosterone and QoL recovery were slower in older men.43 Two years later, a randomized controlled study by Verhagen et al. reported that EORTC scores on two out of five domains showed that PCa patients receiving IAD had significantly better QoL scores for physical and emotional function (P < 0.05), whereas scores for role and social function were equal in both treatment arms. Cognitive function scores were reduced in the IAD, but not in the CAD treatment group.38 Similar observations were reported by Tunn et al., who also carried out a randomized controlled study. In this study, preliminary results showed a clear benefit of IAD over CAD on QoL, especially during the off-treatment period, and no differences in time to progression of cancer growth.28 These observations were confirmed by Bruchovsky et al. These data suggest that QoL was significantly improved during the off-treatment periods for the scales that included physical (P = 0.04) and work functions (P = 0.0001), hot flushes (P = 0.0003) and global QoL (P = 0.004).42

In contrast, Calais da Silva et al. reported that although hot flushes, gynecomastia, headaches and skin complaints were more frequent in patients receiving CAD, overall, there was no difference in QoL between the IAD and CAD arms. Patients receiving CAD did report better emotional responses, and the severity of nausea and vomiting was less in this arm (P = 0.01). Despite observed differences between IAD and CAD, the authors concluded that these QoL differences were small and not clinically meaningful.31 Miller et al. noted that patients' self assessments of their overall health appeared to be favorable with IAD compared with CAD; although there were no significant differences in the incidence of side-effects in patients treated with IAD compared with those receiving CAD.35 Whereas, Bouchot et al. investigated QoL in patients with metastatic PCa receiving IAD, and found that there were no differences in QoL scored during on-treatment and off-treatment periods. However, a decrease in side-effects as a result of the hormonal deprivation was reported in all cases during the off-therapy period; hot flushes, an early androgen deprivation side-effect, was statistically significant improved in the off-treatment period.41 Finally, no differences in QoL in IAD and CAD treatment groups have been observed in studies carried out by Mottet et al.39 and Langenhuijsen et al.36


The aim of the present review was to carry out a literature review of 13 clinical studies that reported safety and tolerability data in PCa patients treated with IAD. For this analysis, we examined the data reported for: (i) hot flushes and sexual dysfunction as indicators of early AE; (ii) bone loss, anemia, obesity and cardiovascular disease as long-term consequences; and finally (iii) QoL assessments in this patient population.

Our analysis suggests that there might be an improvement in the early side-effects and QoL of ADT in PCa patients receiving IAD. We found that over 42% of the studies reported decreased occurrence of hot flushes28,31,32,41–44 and increased sexual function31,32,35,40,42–44 (activity and libido), either as an end-point in itself, or as a part of QoL surveys. Of the 13 studies examined, nine reported QoL as a measureable end-point, and from these studies, 44% showed favorable QoL scores in PCa patients either on IAD or during their off-treatment periods.28,38,42,43 In particular, Spry et al. reported that PCa patients experiencing testosterone recovery during their off-treatment IAD period had significantly improved emotional well-being (P < 0.05), were less fatigued (P < 0.05), had less vomiting and nausea (P < 0.05), slept better (P < 0.01), and had less weight gain (P < 0.05), when compared with those receiving maximal androgen blockage.43

When examining the long-term AE, we determined that the data were inconclusive. Anemia was not a major focus in the studies reviewed, with only one study reporting that PCa patients had a lower incidence of anemia when treated with IAD than those on CAD.45 Based on the 13 studies reviewed, there was no evidence of IAD reducing the risk of anemia induced by CAD. Bone loss and BMI were examined in three studies, and of these, only one showed that patients on the IAD regimen had a decreased rate of bone loss during the off-treatment period; however, levels did not return to baseline.9 The remaining two studies examining BMI and osteoporosis reported that patients on IAD experienced either no benefit or only a small positive effect, which was not statistically significant.34,40 Hence, these current studies show that the effect of IAD on bone density improvement is only minimal, and the small positive effect of IAD compared with CAD on BMD is not clinically meaningful.34 The risk of bone loss during IAD and CAD warrants additional screening of BMD at baseline and/or monitoring of bone loss in order to treat osteopenia/osteoporosis in patients with PCa receiving hormone treatment. Similar inconclusive observations were also noted for anemia,40,45 obesity34,40,43,45 and cardiovascular diseases,31,34,42,44 where no statistical evidence was available to support the hypothesis that IAD, or off-treatment periods of IAD, could provide enhanced benefit and decrease the incidence of these long-term consequences. A recent report showed a relationship between CAD and increased risk of cardiovascular disease in patients with PCa;46 however, these findings do not support the observations by Alibhai et al. and Levine et al., who reported no increase in cardiovascular effects in PCa patients treated with CAD.25,47 Nevertheless, the American Heart Association, American Cancer Society and AUA all advise clinicians to carry out regular cardiovascular-focused follow-up examinations on PCa patients receiving ADT. Levine et al. further suggested that there might be a need for these patients to receive antihypertensive, lipid-lowering or other therapies during ADT.47

Estimation and prevention of cardiovascular risks, and treatment of cardiovascular disease remain an important issue for all patients with PCa, not only those on GnRH agonist therapy. These sentiments were recently echoed by the FDA who issued a Drug Safety Communication in 2010 about the possible increased risk of diabetes and certain cardiovascular diseases when treated with GnRH agonists (; accessed on 10 January 2011). An interesting insight that might provide further understanding of the mortality causes of patients with PCa on ADT was recently provided by Punnen et al. during the recent AUA congress (Abstract#721 from AUA 2011).48 The investigators used a propensity-adjusted multivariate method as an analysis tool to analyze 8095 patients from the CaPSURE registry. Punnen et al. suggested that this statistical methodology helps to provide unbiased estimation of treatment effects when data from retrospective and non-randomized experiments are compared. Based on this propensity-matched subset of patients, it was determined that ADT was not associated with an increased rate of cardiovascular mortality.48

One of the reasons for the inconclusiveness of the data highlighted in the present review might be that heterogeneous patients groups with PCa were treated with IAD, and the results of the analysis of tolerability and safety of the IAD treatment might depend on the disease stage, age, general health status and comorbidities of these patients. For example, patients with stage IV PCa might respond differently to side-effects35 than those with PSA-recurrence, who have recently received their primary localized therapy.28,34,35 Because many current IAD study protocols allow PCa patients with different stages of disease (early and advanced) to enrol, Tunn et al. and Klotz et al. have proposed that IAD treatment might be of most benefit to patients who are treated early in their disease stage.49

The main limitation of our analysis was the small number of studies published with sufficient safety and tolerability data, as 36% were published in abstract format or were preliminary observations. Furthermore, there are limited publications on results from completed randomized trials that assess safety and tolerability of IAD, as many are still in progress (SWOG 9346, CTG-PR7, EC507, TULP). In addition, many of the studies evaluated are single-arm trials and do not compare patients receiving IAD with those on CAD, which limits the ability to carry out a full evaluation on differences between IAD and CAD.

Furthermore, the present review includes clinical trials that used different study designs (randomized vs non-randomized), QoL questionnaires, end-points (just 2/13 studies examined anemia), patient populations and PCa stages. In addition, based on the limited data available, we were unable to explore the effects of the patient's age, metabolic condition, PSA level (pre-, during and post-treatment)/testosterone levels and other indicators of disease progression on the safety and tolerability profiles of IAD. Other limitations included the inability to accurately draw any conclusions on the effects of IDT on metabolic changes, bodyweight and BMI alterations, as little data were reported in these studies. For example, IAD was shown to positively influence bodyweight and BMI changes caused by CAD in just two of the 13 studies reviewed.40,43,45 To date, no published data exists with regards to the effects of IAD on metabolic changes in PCa patients. It therefore remains unclear as to whether the metabolic changes and long-term effects, such as increased risk of diabetes (e.g. decreased insulin sensitivity and increased fasting insulin levels), reported in patients treated with CAD can also be diverted by the introduction of IAD.10,46,50

One review article of interest is that by Abrahamsson.51 Abrahamsson recently carried out a systematic review of the literature, and evaluated the available evidence from 19 phase 2 studies and eight phase 3 trials regarding efficacy and tolerability in PCa patients receiving IAD. This study concluded that IAD appears to be as effective as CAD; however, further studies are currently required to confirm these findings, as the data available are limited.51

Health-care professionals should balance benefits and possible risks of GnRH agonists before commencing treatment. When prescribed, patients on any GnRH agonist should be carefully monitored for development of diabetes, cardiovascular disease or other side-effects, even though intermittent, not continuous, androgen deprivation would be used and, if necessary, treated accordingly.

In conclusion, the data available on the differences between IAD and CAD, with respect to the short-term side-effects, showed a small benefit of IAD in hot flushes, sexual dysfunction and QoL – in particular during the off-treatment time. However to date, the data for the long-term consequences and risks, are limited or inconclusive. Additional information from future publications and ongoing studies, such as SWOG 9346, EC 507 and CTG-PR7, are required to elucidate the potential advantage of IAD in terms of the tolerability and safety profile.

It would be of interest to assess the tolerability and safety profile of other ADT methods in an intermittent setting; for example, anti-androgens or estrogens. These alternatives might lead to a reduction of hot flushes, but in particular, estrogens could increase the risk for cardiovascular side-effects, and would therefore require further clinical trials.

However to date, only drugs leading to medical castration are considered for IAD.7

Overall, this literature review revealed only some safety, tolerability and QoL benefits associated with IAD over CAD. Because IAD did not show higher frequencies of side-effects, nor an increase in their severity, IAD can be considered as an alternative treatment option for patients with advanced PCa if the efficacy is comparable to CAD.7


The authors would like to thank Ai Lockard PhD, an employee of Abbott, for her medical writing support to the authors in the development of this publication. Abbott had the opportunity to review and comment on the publication content.

Conflict of interest

Professor Ulf Tunn declares no conflict of interest. Dr Damian Gruca is employed by Abbott and owns Abbott stock, and Dr Peter Bacher is employed by Abbott and owns Abbott stock.