We report an updated meta-analysis and NMA analyzing and comparing the effect of triplet therapy on OS in patients with mHSPC stratified by disease volume. There are several key findings to our study. First, triplet therapy outperformed docetaxel plus ADT in terms of OS in patients with both high- and low-volume disease. Second, treatment ranking analysis confirmed that triplet therapy had the highest likelihood of improved OS among currently available treatment regimens in patients with high-volume disease. Conversely, in low-volume mHSPC, doublet therapy using ARSI (enzalutamide) plus ADT had the highest likelihood of improved OS.

Our meta-analysis demonstrated the OS benefit from triplet therapy even in mHSPC patients with low-volume disease compared to docetaxel plus ADT. Indeed, HRs for OS from each RCT in patients with low-volume disease were all comparable with those with high-volume, while no study showed statistical significant difference potentially due to the low number of patients and events for low-volume disease resulting in a low statistical power (Fig. 4) [3, 8, 19]. Based on this and a previous meta-analysis without the results from a subgroup analysis of the ARASENS trial, ARSI-based doublet therapy is currently recommended for low-volume disease [2, 28]. However, when synthesizing the results from three RCTs, we demonstrated significant OS benefit from triplet therapy compared to docetaxel plus ADT, even in patients with low-volume disease. On the contrary, when focusing on patients with de novo low-volume disease, we found no significant OS benefit from triplet therapy compared to docetaxel plus ADT. A possible explanation is low statistical power based on the low population of this group, highlighting the need for further studies with larger sample sizes to confirm these findings.

Low-volume mHSPC generally has a more favorable disease trajectory and relatively good prognosis when treated with combination therapies; therefore, long-term follow-up with larger sample size is needed to elucidate the survival outcomes. A recent meta-analysis with long-term follow-up (median 6 years) using individual participant data from the GETUG-15, CHAARTED, and STAMPEDE trials showed OS benefit from adding docetaxel to ADT in mHSPC patients, irrespective of disease volume, while the absolute value was more prominent in patients with high- (HR 0.60, 95%CI 0.52–0.68) compared to those with low- (HR 0.78, 95%CI 0.64–0.94) volume disease [29]. Notably, despite including updated results from the ENZAMET trial, which has a median follow-up of 68 months, the follow-up periods in the ARASENS and PEACE-1 trials at 45.7 and 43 months respectively, were not long enough to fully evaluate the OS outcomes in patients with low-volume disease [3, 4, 19]. Extended follow-up periods are needed to confirm a robust OS benefit from triplet therapy in patients with low-volume disease and to identify the best combination for each individual patient.

Our treatment rankings revealed that triplet therapy, darolutamide plus docetaxel plus ADT, had the highest likelihood of improved OS in patients with high-volume disease. This updated result, including subgroup analyses from the ENZAMET and ARASENS trials, was in line with the results from a previous NMA [2]. High-volume disease is generally associated with biologically and clinically aggressive disease bearing a high probability of harboring androgen receptor-independent cells; suggesting a possible rationale for the efficacy of adding cytotoxic chemotherapy to ADT + ARSI for high-volume disease [30]. In agreement with this concept, the OS benefit from adding docetaxel to ADT was most prominent in patients with high-volume disease in the CHAARTED trial [21]. Taken together, the state of current evidence suggests that triplet therapy with docetaxel plus ASRI plus ADT should be considered in patients with high-volume disease who can tolerate it.

In contrast, enzalutamide plus ADT not only had the highest likelihood of improved OS in patients with low-volume mHSPC but also significantly improved OS when compared to docetaxel plus ADT alone. This suggests a limited role for triplet therapy in patients with low-volume disease and it supports individual shared decision-making. However, our analysis should be interpreted with caution due to possible selection bias. In this NMA, we extracted the data on subgroups treated with or without docetaxel from the PEACE-1 and ENZAMET trials. In these trials, the decision of docetaxel application was based on the physician’s discretion. This might have led to selection bias; younger, healthier patients with more aggressive diseases are more likely to receive docetaxel. For example, in the ENZAMET trial, patients who did not receive docetaxel were more likely to harbor low-volume disease; 71% of patients treated with docetaxel harbored high-volume disease, while this rate was only 37% in patients who did not receive docetaxel [19].

Since the publication of the CHAARTED trial results in 2015, docetaxel plus ADT has gained acceptance as a standard of care (SOC) in patients with mHSPC. As mentioned before, this greatly affected the study design/eligibility of RCTs [18, 26, 31]. Unlike the ENZAMET trial, the ARCHES and TITAN trials also allowed the use of docetaxel as SOC; patients treated with docetaxel were included in both treatment arms. This made reliable NMAs including “true” ARSI-based doublet regimens difficult. Therefore, our treatment rankings lack the apalutamide plus ADT with or without docetaxel, not reflecting all the currently available regimens.

In the argument of whether triplet therapy outperforms ARSI-based doublet therapy in terms of survival outcomes, our previous NMA showed that darolutamide plus docetaxel plus ADT outperformed abiraterone plus ADT in terms of OS in the overall cohort[2]. However, present NMAs stratified by disease volume failed to show the statistical OS superiority of triplet therapy over ARSI-based doublet therapy in both patients with high- and low volume disease. A possible explanation is that these results were based on subgroup analyses from each RCT, limiting the statistical power. Only a head-to-head RCT comparing triplet therapy versus ARSI-based doublet therapy will clarify the potential impact on survival outcomes.

Last but not least, the benefit-harm balance is, indeed, an important factor for clinical decision-making. A recent meta-analysis showed that ARSI-based doublet therapy had high probabilities for a net clinical benefit; however, docetaxel-based doublet as well as triplet therapy appeared unlikely to be beneficial [32]. Docetaxel is known to increase the risk of severe adverse events and can reduce the patient quality of life [2, 33]. Our previous meta-analysis confirmed that docetaxel-based combination had a higher likelihood of severe adverse events, with triplet therapies being the highest adverse event rates [2, 33]. In the subgroup analysis of the ARASENS trial, 74% of patients with low-volume disease who received triplet therapy experienced severe adverse events [8]. Although our study showed the oncologic utility of triplet therapy, shared decision-making considering the benefit-harm balance is essential, specifically in such patients with a long life expectancy [8]. Further investigation is needed to select the optimal candidates who are most likely to benefit from triplet therapy.

Besides the abovementioned issues, the current study has several limitations that need to be considered. First, despite careful data extraction from RCTs, each study differed in patient proportion, such as the rates of de novo/metachronous and high/low-volume disease. Our analyses based on subgroup analyses of each RCT might reduce the patient heterogeneity; however, this differential proportion could potentially affect the outcomes. Second, in the ARCHES, ENZAMET, and TITAN trials assessing the efficacy of ARSIs plus ADT combinations compared to ADT alone, patients were allowed to use docetaxel at the time of or after randomization [18, 26, 27]. However, only updated results from the ENZAMET trial provided the OS data on patients treated with or without docetaxel, stratified by disease volume [19]. Therefore, we excluded the ARCHES and TITAN trials. Third, NMAs have a limited role in facilitating patient selection; this approach cannot substitute for a direct comparison of each treatment and is mostly hypothesis-generating. Additionally, while patients with low-volume disease are generally less likely to experience mortality in the short-term follow-up, the effectiveness of triplet therapy may not be apparent in our analysis. Our findings need to be validated in head-to-head, well-designed RCTs. Fourth, our analyses are limited in assessing which regimens are the best combination for each clinical setting. Due to the nature of subgroup analyses, a limited number of studies assessing the outcomes stratified by disease volume. Furthermore, there was a paucity of studies reporting outcomes stratified by metachronous high- or low-volume separately. Fifth, although we investigated OS stratified by disease volume, other factors, such as Gleason pattern 5, TP53 mutations, also affect mortality even for patients with low-volume disease [34, 35]. Therefore, patients with these factors may benefit from triplet therapy even in the context of low-volume disease. Finally, the ENZAMET trial included the use of non-steroidal antiandrogen therapy with ADT in the control arm. This might provide a differential survival benefit in the control arm, therefore, weighing against the survival outcomes of enzalutamide. Again, we need further RCTs, especially for RCTs directly comparing triplet therapy versus ARSI-based doublet therapy. In addition, the comparative efficacy of darolutamide with other ARSIs as ARSI-based doublet combinations from the ongoing ARANOTE trial (NCT04736199) will enrich the treatment option for mHSPC. Based on this analysis and a previous meta-analysis, which did not include results from a subgroup analysis of the AR therapy is currently recommended for patients with low-volume disease [36].