Most of the phase III trials tended to focus on advanced or metastatic disease rather than biochemical failure, except for the NCT3653 trial [29] and others not yet published [35].

Although there was apparent consensus in most trials on the PSA level designated for ADT discontinuation (<4 ng/ml), the criteria for resuming treatment were less uniform (>10 ng/ml or >20 ng/ml, depending on the stage and the presence of symptoms) [22].

Time-off therapy is variable in the different studies (from 0.7–4.9 months in the TULP study to 20–59.6 months in the NCT3653 study), and the risk is that in some studies, IAD patients are often on therapy rather than off. In particular, the duration of off-treatment periods decreased progressively during the following cycles of IAD. The conclusions of these trials support the hypothesis that IAD, mainly in metastatic cases, can produce oncologic results similar (not inferior, as defined by some trials) to those of continuous ADT.

Although many studies and publications evaluated the efficacy of IAD, the evaluation of the safety and tolerability of this regimen is often limited, and data are not complete in all respects. Early side effects as hot flushes and sexual dysfunction are the most common early side effects described during ADT and are able to affect patient quality of life (QoL) [36].

In the TAP22 trial [28], the incidence of adverse side effects was significantly higher in the continuous ADT group (93.6 %) than in the IAD group (84.4 %) (p = 0.042). In the TULP trial [32], a trend of more side effects like hot flushes was seen in the continuous ADT group compared with the IAD group.

In the NCT3653 trial [29], IAD was associated with significantly better scores (not specified; p < 0.001) for hot flushes and desire for sexual activity. Although only 35 % of patients in the IAD group had a return to pre-treatment testosterone levels within 2 years after completing the first period of treatment, 79 % had a level of at least 5 nmol/l.

In the study by de Leval et al. [31], hot flushes were reported as less frequent (not specified) and as mild to moderate during IAD treatment and were resolved or improved in the off-treatment periods. Regarding erectile dysfunction, lower incidence (not specified) was reported in the IAD group, with significant improvement during the off phases of treatment.

In the FinnProstate study [34], a non-significantly lower incidence (p = 0.44) of hot flushes was found for the IAD group. The only significant difference in early side effects emerged unexpectedly in the numbers of patients reporting erectile dysfunction and depression, which were more common in the IAD group (for IAD vs. continuous ADT, respectively, erectile dysfunction was reported by 15.7 % vs. 7.9 % and depression was reported by 2.2 % vs. 0 %; p < 0.05). In the IAD group, testosterone levels showed recovery at the end of each off-treatment phase but did not reach the same level as at the end of the previous off phases. At entry, 81.2 % of patients in the IAD arm had a testosterone level >10 nmol/l; however, only 47.4 % had a testosterone level >10 nmol/l at the end of the 10th off-treatment phase.

In the SEUG 9401 trial [33], patients on IAD experienced lower incidence of hot flushes and fewer problems related to sexual function and reported increased sexual activity (p < 0.01). In the first period after randomization, virtually all IAD patients were off therapy, and their levels of sexual activity were similar to pre-treatment levels (35 %). As the follow-up time increased, more IAD patients were on therapy and sexual activity decreased.

Long-term side effects of ADT include loss of bone mineral density, metabolic changes, and cardiovascular disease [36]. Publications related to phase III trials were not designed to examine the long-term consequences of ADT for these parameters. The SEUG 9401 trial [33] found an increased risk of dying from cardiovascular disease in the continuous ADT group (cardiovascular deaths: 41 [13.1 %] in the IAD group, 52 [16.7 %] in the continuous ADT group) but similar incidence of cardiac ischemia/infarction in both treatment arms (10 % for IAD, 11 % for continuous ADT).

In the FinnProstate study [34], no significant differences were found between the two treatment groups in terms of cardiovascular side effects (31.8 % for IAD vs. 33.9 % for continuous ADT; p = 0.59). Cardiovascular-related mortality was also similar (12.8 % for IAD vs. 15.4 % for continuous ADT; p = 0.38).

Only the Finn Prostate phase III trial [34] published a study specifically focused on the effect of IAD on QoL and adverse events. Based on available data, the frequency of early side effects such as hot flushes or sexual dysfunction significantly decreases in the IAD group when compared with the continuous-treatment group. The severity of early side effects was also improved in the IAD group.

Considering the QoL difference as a whole, the two treatment regimens seem to be very similar. Results are probably influenced by the duration of the off-treatment periods and by the rate of testosterone recovery.

A recent randomized phase III trial, the SEUG 9901 [37], demonstrated the non-inferiority of antiandrogen monotherapy in IAD compared with maximal androgen blockade. They enrolled 1,045 patients with locally advanced or metastatic prostate cancer, 918 responded to induction therapy and therefore were randomized. Overall survival (OS) was similar in the two groups, and non-inferiority of IAD was demonstrated with HR 0.90 (95 % CI 0.76–1.07). There was a trend for an interaction between treatment and PSA, favouring IAD over CAB in patients with PSA < 1 ng/ml. After randomization, 50 % of patients were off therapy for >2.5 years and 28 % were off therapy for >5 years. In IAD group were found better results on sexual function.

The two largest phase III trials to date are the NCT3653 study [29] and the SWOG 9346 [30]. Both were designed as non-inferiority studies. The results, which are somewhat contradictory, are intriguing. The NCT3653 trial enrolled 1,386 patients with a PSA level greater than 3 ng/ml more than 1 year after primary or salvage radiotherapy for localized PC. IAD was provided in 8-month cycles, with non-treatment periods determined according to the PSA level. The primary end point was overall survival. Secondary end points included quality of life, time to CRPC, and duration of non-treatment intervals. Median follow-up was 6.9 years. In the IAD group, full testosterone recovery occurred in 35 % of patients, and testosterone recovery to the trial-entry threshold occurred in 79 %. Patients on the intermittent arm were on treatment only 27 % of the time. Clear cut QOL benefits during the off-treatment interval were seen in the domains of erectile function, libido, hot flashes, physical function, fatigue, and urinary symptoms. There were 268 deaths in the intermittent-therapy group and 256 in the continuous-therapy group. Median overall survival was 8.8 years in the intermittent-therapy group versus 9.1 years in the continuous-therapy group (hazard ratio 1.02; 95 % CI 0.86–1.21). The 7-year cumulative rates of PC mortality were 18 % and 15 %, respectively (p = 0.24). This important study demonstrated that overall survival was equivalent in men with biochemical failure treated with intermittent therapy.

SWOG 9346 was designed to evaluate overall and disease-specific survival in IAD compared to continuous therapy, but in men with metastatic disease. A total of 1,535 patients with metastatic PC and a PSA > 5 ng/ml were submitted to 7 months of Goserelin and Bicalutamide. If the PSA was <4 ng/ml by month 6, they were then randomized between IAD and continuous therapy. Treatment was re-initiated when the PSA reached 20 ng/ml, and discontinued again after 7 months if the PSA was <4 ng/ml. Overall survival showed a non-significant trend to improvement in the continuous arm (HR 1.10, 95 % CI 0.99–1.23). However, considering HR values, the study was considered statistically inconclusive in terms to define a non-inferiority or an inferiority of IAD versus continuous CAD. Median survival was 5.8 vs. 5.1 years in IAD and continuous ADT, respectively. A sub-analysis stratifying by minimal disease (confined to axial skeleton and pelvis or lymph nodes) vs. extensive disease (ribs, long bones, skull, and/or viscera) was performed. This showed a benefit in the minimal disease group with continuous therapy, with a HR of 1.19, (95 % CI 0.98–1.43, p = 0.034) and none in the extensive disease group, with a HR of 1.02 (95 % CI 0.85–1.22). In the minimal disease group, median overall survival was 5.4 years in the IAD group vs. 6.9 years in the continuous group. The interpretation of this study is controversial. The stratification analysis was post hoc, and is therefore hypothesis generating, not proof, but also it is a challenge to make biological sense of the bracketing of the supposed inferiority of IAD in minimal metastatic disease, but not in non-metastatic or extensive metastatic disease.