Of the three RCTs that comprised most of the articles in our search, only the ARASENS study set out to compare triplet therapy and doublet therapy from the start (12). The PEACE-1 and ENZAMET studies began using ADT alone (as the standard of care [SOC]) for background therapy but later included ADT + docetaxel as an option once this combination became the SOC (11, 13, 14).

Patient inclusion/eligibility criteria were similar but the studies had different designs (Table 1). ARASENS had a straightforward randomized, double-blind, placebo-controlled design, with two treatment arms; patients in both the active treatment group and control group received ADT + docetaxel (12), but one received darolutamide and the other placebo, i.e., this was a comparison of doublet therapy with ADT + docetaxel versus triplet therapy.

PEACE-1 was open-label and included four treatment groups using a 2 × 2 factorial design: 1) SOC alone; 2) SOC + radiotherapy; 3) SOC + abiraterone (and prednisone); and 4) SOC + abiraterone + radiotherapy (13). At the start of the study, SOC was ADT, but a protocol modification allowed physicians to start using ADT + docetaxel as SOC. Therefore, not all patients received docetaxel as SOC; of the 583 patients assigned to abiraterone (with or without radiotherapy), 355 (60.9%) received the triplet regimen of ADT + docetaxel + abiraterone.

ENZAMET was also open-label, with patients randomized to SOC + enzalutamide or SOC + a first-generation nonsteroidal ARPI (bicalutamide, flutamide or nilutamide) (11, 14). After enrolment of the first 88 patients, investigators were given the option to add open-label docetaxel to ADT as SOC. As a result, some of the patients (n = 483; 43%) received triplet therapy, of whom 243 received enzalutamide and 240 received a nonsteroidal ARPI. The rest of the patients in this study (n = 642) received doublet therapy with either enzalutamide + SOC or a nonsteroidal ARPI + SOC. Therefore, while patients in the ENZAMET study received doublet or triplet therapy, the study was not designed to compare outcomes in these two groups; rather, it was designed to compare outcomes between patients receiving enzalutamide and those receiving a first-generation ARPI. Moreover, patients receiving doublet therapy in this study received ADT with either enzalutamide or nonsteroidal ARPI; none received ADT + docetaxel.

All three studies had overall survival (OS) as a primary endpoint (11-13), but the PEACE-1 study had a co-primary endpoint of radiographic progression-free survival (PFS) (13).

Regarding patient characteristics, the population of the ARASENS trial had high-risk disease (78.2% Gleason score ≥8, 86.1% synchronous disease and 77.0% high-volume disease) (12, 40). All patients included in the PEACE-1 trial had synchronous disease and 57%–65% also had high-burden disease, so therefore were also at high risk overall (13). In ENZAMET, 60.6% of patients had synchronous disease, 53.5% had high-volume disease and 57%–60% had a Gleason score ≥8; 61% of patients with high-volume disease received docetaxel versus 27% of low-volume disease patients (11).

In all three studies, the treatment groups that included a next-generation ARPI had significantly better OS than the groups without this therapy (11-14).

In ARASENS, the only study specifically designed to compare doublet and triplet therapy as a predefined objective, triplet therapy containing darolutamide reduced the risk of death by 32.5% compared with ADT + docetaxel (hazard ratio [HR] 0.68, 95% confidence interval [CI] 0.57–0.80; p < 0.001; Table 2), even though 76% of patients in the doublet therapy group received subsequent systemic therapies (12). Triplet therapy was also associated with a significantly reduced risk of most secondary endpoints including time to castration-resistant prostate cancer (CRPC; HR 0.36, 95% CI 0.30–0.42; p < 0.001; Table 2), time to pain progression (HR 0.79, 95% CI 0.66–0.95; p = 0.01; Table 2), symptomatic skeletal event-free survival (HR 0.61, 95% CI 0.52–0.72; p < 0.001), time to first symptomatic skeletal event (HR 0.71, 95% CI 0.54–0.94; p = 0.02), and time to initiation of subsequent antineoplastic therapy (HR 0.39, 95% CI 0.33–0.46; p < 0.001). There was no significant difference between triplet and ADT + docetaxel doublet therapy in the time to worsening of disease-related symptoms (12). Median time to initiation of opioid use for ≥7 days was not reached in either group, but there was a trend towards earlier initiation of opioids in the triplet versus the doublet therapy group (HR 0.69, 95% CI 0.52–0.91).

ENZAMET, which compared enzalutamide with first-generation ARPIs in combination with ADT ± docetaxel, found that the enzalutamide-containing regimen reduced the risk of death by 30% (HR 0.70, 95% CI 0.58–0.84; p < 0.0001) in the overall study cohort at final analysis (14). Enzalutamide-containing treatment was also associated with a significantly reduced risk of most secondary endpoints including time to prostate-specific antigen (PSA) progression (HR 0.44, 95% CI 0.38–0.52), time to clinical progression (HR 0.45, 95% CI 0.39–0.53), and prostate cancer-specific survival (HR 0.67, 95% CI 0.54–0.82) (14). However, in a pre-specified subgroup analysis of patients with early planned use of docetaxel (the triplet therapy group), the difference in OS between the enzalutamide group and control group did not reach statistical significance (HR 0.82, 95% CI 0.63–1.06). On the one hand, the use of docetaxel was at the investigator’s discretion, and patients who were selected for triplet therapy in this study had worse prognostic features than those who received doublet therapy, specifically a higher proportion of patients receiving triplet therapy had synchronous and/or high-volume disease. On the other hand, in the high-risk subgroup (patients with synchronous high-volume disease), older patients and those with comorbidities were less likely to receive docetaxel (14). A pre-specified subgroup analysis showed that triplet therapy containing enzalutamide significantly improved OS in patients with synchronous metastases compared with triplet therapy containing a first-generation ARPI, but not in those with metachronous disease (Figure 2) (14).

The subgroup findings from ENZAMET are consistent with results in the ARASENS study, in which the effect of triplet versus ADT + docetaxel doublet therapy on OS remained significant in the group with synchronous metastases but not those with metachronous metastases (9). In ARASENS, triplet therapy significantly improved OS in patients with high-volume disease (using the definition from the CHAARTED study, i.e., visceral metastases and/or ≥4 bone metastases with ≥1 beyond the vertebral column and pelvis (62)) and in patients with low- or high-risk disease (high-risk defined as two of the following criteria: Gleason score ≥8, ≥3 bone metastases and presence of measurable visceral metastasis (63)), but not those with low-volume disease (Figure 2) (9).

In the PEACE-1 study, 60.5% of patients received docetaxel as SOC, so the results in this subgroup allowed a comparison of triplet therapy with ADT + abiraterone + docetaxel (±radiotherapy; n = 355) and doublet therapy with ADT + docetaxel (±radiotherapy; n = 355) (13). Triplet therapy significantly improved the two co-primary endpoints of OS and radiographic PFS compared with ADT + docetaxel. The adjusted HR for OS was 0.75 (95.1% CI 0.59–0.95; p = 0.017) and the adjusted HR for radiographic PFS was 0.50 (99.9% CI 0.34–0.71; p < 0.0001).

Secondary endpoints of CRPC-free survival and prostate cancer-specific survival were also significantly better in the triplet than doublet therapy groups, with median CRPC-free survival of 3.21 versus 1.45 years in the triplet versus ADT + docetaxel doublet therapy groups, respectively (HR 0.38, 95% CI 0.31–0.47; p < 0.0001). Median prostate cancer-specific survival was not reached in the triplet therapy group but was 4.72 years in the doublet therapy group (HR 0.69, 95% CI 0.53–0.90; p = 0.0062) (13).

Consistent with the results of ARASENS, triplet therapy in PEACE-1 was associated with a significant improvement in OS in the subgroup of patients with high-volume metastatic disease but not in those with low-volume disease (Figure 3) (13). However, the other co-primary endpoint of radiographic PFS showed a significant improvement with triplet versus ADT + docetaxel doublet therapy in patients with low- or high-volume metastatic disease, with an HR of 0.58 (99.9% CI 0.29–1.15; p = 0.0061) in the low-volume group and 0.47 (99.9% CI 0.30–0.72; p < 0.0001) in the high-volume group (13).

Overall, triplet therapy regimens demonstrated superior survival benefits in the key trials versus doublet therapy in terms of radiographic PFS, time to CRPC, pain progression and remaining free from skeletal events, and the need for subsequent antineoplastic therapy.

The preliminary results of a PEACE-1 subanalysis suggested that the magnitude of the benefit of adding abiraterone to SOC decreased with age (44). However, the magnitude of the improvement in radiographic PFS with triplet versus ADT + docetaxel doublet therapy was similar in men aged ≥70 years (HR 0.55, 95% CI 0.29–1.04) and those aged <70 years (HR 0.5, 95% CI 0.33–0.78), whereas younger men tended to derive a greater OS benefit (HR 0.71, 95% CI 0.52–0.95) than men aged ≥70 years (HR 0.80, 95% CI 0.53–1.2) (44). This was not the case in the ENZAMET trial, in which patients aged <70 and ≥70 years both derived a similar benefit from triplet therapy (43). The magnitude of the OS benefit associated with enzalutamide-containing triplet therapy relative to a nonsteroidal antiandrogen-containing triplet therapy in these age subgroups did not reach statistical significance (43). Both the PEACE-1 and ENZAMET analyses in elderly patients have been presented at conferences but data have not yet been published in full.

Subgroup analysis of the ARASENS data showed that the benefits of triplet versus ADT + docetaxel are directionally consistent in all age groups and ethnic/geographic/racial groups (12), including Chinese patients (56), North American patients (20), and Black/African-American patients (20, 21). Ongoing trials such as the PANTHER study are evaluating the effect of the combination of apalutamide + abiraterone + prednisone in clinical efficacy in Black men with mCRPC who are typically under-represented in clinical trials (64). Interim results indicate that Black participants had greater radiographic PFS, time to PSA progression, and overall survival than White participants (64). ARACOG is an ongoing, prospective, randomized, open-label phase II study comparing cognitive outcomes between men with metastatic and non-metastatic CRPC or mHSPC in the United States (65). Patients will be randomized (1:1) to receive treatment with enzalutamide 160 mg orally daily or darolutamide 600 mg orally twice daily, in combination with standard LHRH agonist-based treatment. Cognitive assessments will be to assess cognitive function and impairment (65).

To date, only the ARASENS study has reported patient-reported outcomes (PROs) data and only as conference abstracts (15, 18). These data suggest that the addition of an ARPI does not negatively impact on patient quality of life, with similar quality of life in the triplet and ARPI + docetaxel doublet therapy arms (15, 18).

In the ARASENS study, the overall incidences of adverse events (AEs), serious AEs, grade ≥3 AEs and fatal AEs were similar in patients receiving ADT + docetaxel doublet versus triplet therapy; 98.5% and 99.5% of patients in these groups, respectively, developed any AEs, 42.3% and 44.8% experienced serious AEs, 63.5% and 66.1% had grade 3 or 4 AEs, and 4.0% and 4.1% died as a result of AEs (12). The highest incidence of AEs was seen with docetaxel and the most common AEs were those related to docetaxel in both groups, i.e., alopecia, fatigue, anemia and neutropenia (which included the preferred terms of leukopenia, neutropenia, decreased neutrophil count and decreased white blood cell count). Neutropenia occurred in 38.8% of patients in the doublet therapy arm and 39.3% in the triplet therapy arm, alopecia in 40.6% and 40.5%, respectively, fatigue in 32.9% and 33.1%, respectively, and anemia in 25.1% and 27.8%, respectively. AEs that occurred at a higher incidence in the group receiving darolutamide versus placebo were rash (16.6% vs. 13.5%) and hypertension (13.7% vs. 9.2%), although the rate of these events was similar in the two groups when adjusted for exposure time (rash: 6.2 vs. 7.3 per 100 patient-years [PY], respectively; hypertension: 4.9 per 100 PY in both groups) (12). AEs led to treatment discontinuation in 13.5% of patients in the darolutamide group and 10.6% of patients in the placebo group.

In the PEACE-1 study, 100% of patients receiving abiraterone-containing triplet therapy and 100% receiving ADT + docetaxel doublet therapy developed an AE of any severity, 63% versus 52% of patients who received triplet versus doublet therapy developed at least one grade ≥3 AE, and 2% versus 1%, respectively, developed a fatal AE (13). However, the incidence of severe AEs per 100 PY of treatment was 49 in the group receiving triplet therapy with abiraterone versus 55 in the group receiving doublet therapy. Abiraterone was stopped because of toxicity in 29 patients (21.0%) receiving triplet therapy and in one patient (<1.0%) receiving doublet therapy. AEs occurring at a higher incidence in the triplet versus ADT + docetaxel doublet arms of the PEACE-1 study, respectively, were hypertension (22% vs. 13%) and hepatotoxicity with elevated aminotransferase levels (6% vs. 1%) (13). Grade ≥3 neutropenia occurred at a similar rate in patients receiving triplet therapy and those receiving ADT + docetaxel (10% and 9%, respectively).

In the ENZAMET study, the overall incidence of AEs was similar in patients receiving enzalutamide versus a first-generation ARPI as part of triple therapy, but more patients in the first-generation ARPI group had grade 1–2 AEs (51% vs. 31%) and more patients in the enzalutamide group had grade 3 AEs (58% vs. 37%) (14). The most common grade 3–4 AEs were febrile neutropenia during docetaxel, which developed in 6% of patients in both treatment groups, fatigue (occurring in 1% of the control group vs. 5% of the enzalutamide group) and hypertension (occurring in 6% of the control group vs. 10% of the enzalutamide group) (14). Deaths due to serious AEs occurred in 10 patients (2%) in the control group and 13 patients (2%) in the enzalutamide group; no deaths were considered to be caused by enzalutamide.

It is expected that triplet therapy regimes would precipitate more AEs than doublet regimens. Overall, AE occurrence has been consistent among the three main RCTs and no new safety concerns have been identified.

The five meta-analyses that compared triplet with ADT + docetaxel doublet therapy reported a consistent overall improvement in OS with triplet therapy of about 25% (Table 3) (48, 50, 56, 58, 59).

The 11 NMAs included between seven and 28 studies of triplet and doublet combinations conducted in between 5,804 and 12,298 patients (49, 51, 52, 53, 54, 55, 56, 57, 59, 60, 61). Because the NMAs included doublet therapy studies, the analyses also included apalutamide, which has not been studied as part of triplet therapy in RCTs. The survival analyses showed that darolutamide triplet therapy was consistently associated with a significant improvement in OS compared with ADT + docetaxel (Supplementary Figure S1A) (49, 52, 54, 55, 60). Abiraterone-containing triplet therapy was associated with a significant improvement in OS compared with ADT + docetaxel in three out of four NMAs analyzing this association (Supplementary Figure S1B) (49, 52, 55, 60), but neither enzalutamide- nor apalutamide-containing triplet regimens were associated with a significant improvement in OS compared with ADT + docetaxel (Supplementary Figures S1C, S1D) (49, 52, 54, 60, 61). The NMAs also evaluated indirect comparisons between triplet therapy and ADT + ARPI doublet therapy. In all cases, triplet therapy was not significantly more effective than ADT + ARPI doublet at improving OS (Supplementary Figures S2A–C) (53, 54, 55, 57, 60, 61), and in one case, abiraterone + ADT doublet therapy was found to be significantly better than abiraterone-based triplet therapy (Supplementary Figure S2C) (53).

The NMA data regarding PFS were generally similar to the OS data, with triplet therapy containing abiraterone or darolutamide being significantly better than SOC (ADT, ADT + ARPI or ADT + docetaxel) in one NMA (Supplementary Table S1) (53). Abiraterone-containing triplet therapy was significantly superior to ADT + docetaxel in two NMAs (55, 60). Enzalutamide-containing triplet therapy was superior to ADT + docetaxel in one NMA (60) but not in another (61). In most comparisons of triplet therapy with ARPI + ADT doublets, there was no significant differences in PFS (Supplementary Table S1), but Wang and colleagues found a significant improvement in PFS with abiraterone- or enzalutamide-containing triplet therapy compared with a doublet regimen of abiraterone + ADT (60). Triplet combinations were significantly superior to ADT + docetaxel for improving radiographic PFS and time to CRPC in an analysis by Jian and colleagues (Supplementary Table S2) (52).

When the data were analyzed by subgroup, none were significant in patients with low-volume disease (Supplementary Table S3), but some of the NMA comparisons of OS were statistically significant in patients with high-volume disease (Supplementary Table S4). With regard to PFS, none of the comparisons of abiraterone triplet therapy with any comparator (ADT alone, ADT + docetaxel or ADT + ARPI) were statistically significant in the low-volume disease groups (Supplementary Table S5) (55), whereas abiraterone was associated with a significant improvement in PFS in patients with high-volume disease compared with ADT alone (51, 54, 55), ADT + docetaxel (55), and ADT + apalutamide (55), but not compared with ADT + enzalutamide (55).

In the subgroup of patients with synchronous disease, significant improvements in OS were seen with triplet therapy compared with ADT alone or ADT + docetaxel, but not with ADT + ARPI (Supplementary Table S6) (55). No significant improvement was seen with triplet therapy vs. any comparator in patients with metachronous disease (Supplementary Table S6) (55).

Eight meta-analyses evaluated the effects of triplet therapy on the incidence of AEs (49, 50, 52, 54, 56, 58, 60, 61), but the comparisons were mostly with ADT alone (54, 60, 61), ADT ± standard nonsteroidal antiandrogen (56) or docetaxel alone (49). The meta-analysis by Jian and colleagues was the only one to specifically compare AE rates during individual triplet regimens (abiraterone + ADT + docetaxel, and darolutamide + ADT + docetaxel) versus doublet therapy (52). In this analysis, both abiraterone- and darolutamide-containing triplet therapy were associated with a higher risk of hypertension compared with ADT + docetaxel, but only abiraterone + ADT + docetaxel was associated with a higher risk of grade ≥3 AEs (Supplementary Figure S3) (52). The risks of neutropenia and febrile neutropenia were similar with triplet regimens and ADT + docetaxel (52).

Menges and colleagues evaluated the net clinical benefit of doublet and triplet regimens, as the number of deaths avoided per 1,000 patients compared with ADT alone, weighted by 0.18 for incident grade 1–2 AEs and by 0.53 for incident grade 3–4 AEs (54). In this analysis, the probability of having a net clinical benefit over 24 months compared with ADT alone was 63.3%–78.7% for ARPI + ADT doublet regimens and 20.0% for darolutamide + ADT + docetaxel, and over 5 years was 66.7%–83.2% with ARPI + ADT doublet regimens and 23.5% for darolutamide + ADT + docetaxel. A sensitivity analysis using lower preference weights for AEs increased the probabilities, but did not change the relative rankings (54). No analysis was conducted comparing the net clinical benefit of triplet therapy with that of ARPI + ADT doublet therapy.

AE results data for the meta-analyses were similar to those obtained for RCTs.