Comparative survival in metastatic hormone sensitive prostate cancer by volume of disease and timing of metastasis: a living network meta-analysis

Abstract

Background and Objective:

We aimed to assess the comparative effectiveness of contemporary systemic treatment options across patients with mHSPC across clinically relevant prognostic subgroups (synchronous high [SHV] and low volume [SLV] and metachronous high [MHV] and low volume [MLV]).

Methods:

This living network meta-analysis was conducted using the living interactive evidence (LIvE) synthesis framework. Phase III randomized controlled trials (RCTs) assessing treatment intensification with ARPI, D or both were included. Mixed treatment comparisons were conducted for overall population and for each prognostic subgroup SHV, SL, MHV and MLV. Overall survival (OS) and progression-free survival (PFS) were assessed.

Key findings and limitations:

Current report of living systematic review includes a total of 11 trials (12668 patients and 12 unique treatments). In the overall population, the results were consistent with the previous report. Analysis of OS by pre-specified subgroups included nine clinical trials (8990 patients and eight unique treatments). In SHV (N=5171; 57%), ARPI+D+ADT led to statistically significant improvement in OS compared to D+ADT (HR: 0.72; 95% CI: 0.62–0.83) and ARPI+ADT (0.71; 0.53–0.97). In SLV (N=2455; 27%), ARPI+ADT led to statistically significant improvement compared to ADT (0.65; 0.52–0.80). There was no statistically significant difference between ARPI+D+ADT and ARPI+ADT (1.08; 0.65–1.79). In MHV (N=589; 6.5%), no statistically significant improvement was observed with ARPI+D+ADT compared to ARPI+ADT (0.89; 0.43–1.85), and D+ADT (0.90; 0.60–1.36). There was no statistically significant difference between ARPI+ADT and D+ADT (1.02; 0.45–2.28). In MLV (N=775; 8.5%), ARPI+ADT led to statistically significant improvement compared to ADT (0.43;0.29–0.64) and D+ADT (0.41; 0.24–0.70). There was no statistically significant difference between ARPI+D+ADT and ARPI+ADT (1.56; 0.40–6.25). Inherent limitations of this analysis include the inability to account for all relevant variables such as the patient and cancer related factors which likely influenced the decision for physicians to offer docetaxel to patients.

Conclusions and clinical implications:

Current evidence suggests that triplet systemic therapy is preferred for patients with SHV mHSPC who are fit for docetaxel. Androgen receptor pathway doublet therapy is preferred for all other patient subgroups compared with ADT alone. There is no role of docetaxel doublet in patients with access to ARPI therapy and if they are able to receive it.

1. Introduction

The treatment landscape for patients diagnosed with metastatic hormone-sensitive prostate cancer (mHSPC) has rapidly evolved over the last five years with the emergence of triplet systemic regimens and disease prognostication by volume of disease [1]. Now, patients have several life-prolonging options such as novel androgen receptor pathway inhibitor (ARPI) doublet therapy [2–7], docetaxel doublet therapy, and triplet systemic therapy [8–10] with the addition of both ARPI and docetaxel to androgen deprivation therapy (ADT), as well as consideration of radiation to the prostate[11–13].

Results from the CHAARTED and STAMPEDE trials[14–16] demonstrated that patients with mHSPC could derive survival benefits by adding docetaxel to ADT. Subgroup analysis in the CHAARTED and GETUG-15 trials, but not STAMPEDE, suggested that this was driven by benefit in the synchronous high volume (SHV) subgroup with smaller benefits in patients with synchronous low volume (SLV) disease. Quantitative evidence from a recent individual-patient data meta-analysis of the three main docetaxel trials (GETUG 15, CHAARTED, and STAMPEDE arm C) confirmed the differential efficacy of docetaxel by timing of metastases and disease volume[17]. Similarly, in early 2023, the initial report from our living interactive systematic review (LISR) which was based on primary analysis of the PEACE-1 and ARASENS trials [8, 9] suggested an overall survival benefit for triplet therapy over docetaxel plus ADT (D+ADT) and ARPI plus ADT (ARPI+ADT) only in patients with SHV disease[1].

Since our first report, the data for triplet therapy with darolutamide and docetaxel from the ARASENS trial [18] has been updated with subset analyses according to risk/volume of disease and timing of metastases and has shown consistent benefit with triplet therapy in all prognostic subgroups with the most significant benefit in SHV disease. Likewise, the ENZAMET trial [6] showed that enzalutamide with concurrent docetaxel and ADT (E+D+ADT) appeared to be more beneficial than D+ADT with standard non-steroidal anti-androgen (NSAA) in SLV and SHV disease while E+ADT was more beneficial than ADT with NSAA in all subgroups when concurrent docetaxel was not suitable. Results from the trials - ARCHES (enzalutamide) [3, 19] and TITAN (apalutamide) [20] - further confirmed the consistent OS benefits regardless of disease volume of and timing of metastases. Recently, results from the ARANOTE trial showed significant progression-free survival improvement with the addition of darolutamide to ADT[21]. Hence, it has now become crucial to reassess how these systemic therapies compare with each other and which patients may benefit from these options using recently available data. The need for robust indirect analyses is further apparent in the absence of any trials directly comparing ARPI plus docetaxel and ADT (ARPI+D+ADT) to ARPI+ADT and the need for data to guide patients and physicians.

Collating disease volume and metastasis timing creates four prognostic groups that may help guide systemic therapy in the absence of validated biomarkers. Therefore, in this report, we present the updated results from our LISR and network meta-analysis assessing the comparative effectiveness of contemporary systemic treatment options by disease volume and timing of metastatic presentation in patients diagnosed with mHSPC. Future reports will incorporate the published data from PEACE-1 (NCT01957436) of radiation to the prostate.

2. Methods

This LISR is conducted using the living interactive evidence (LIvE) synthesis framework [22]. Detailed methods were published with the initial report and provided in Supplementary Methods 1[1]. It is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension statement for systematic reviews, incorporating network meta-analyses for healthcare interventions (Supplementary Methods 2) [23]. The original living systematic review was registered at the open science framework (OSF: https://osf.io/e2q3w) and is live on (living evidence)

2.1. Literature search and study selection

The ‘Watcher’ module performs automated weekly searches using EMBASE and Ovid MEDLINE^® to identify new or updated reports of eligible criteria. The detailed ‘living’ search strategy and methods are provided in Supplementary Methods 1.

All phase III randomized controlled trials assessing contemporary systemic treatment options in patients diagnosed with mHSPC were eligible. Pre-specified subgroups of interest (a priori for this report) included the following categories: synchronous (de novo) high volume (SHV), synchronous (de novo) low volume (SLV), metachronous (recurrent) high volume (MHV), and metachronous (recurrent) low volume disease (MLV), using CHAARTED criteria for volume assessment[14, 15].

2.2. Data extraction and quality assessment

The main outcome of interest included overall survival (OS) and progression-free survival (PFS). When eligible trials had multiple reports, data from the most recent follow-up was included in the analysis. The quality of included trials was assessed using the Cochrane Risk of Bias tool version 2. This process of data extraction and quality assessment was carried out by two independent reviewers (SAAN and IBR). Any discrepancies were resolved by consensus and input from a senior reviewer (AHB).

2.3. Data analysis

Pre-computed hazard ratios (HR) with corresponding 95% confidence intervals (CI) were pooled using an inverse variance approach following logarithmic transformation. A DerSimonian and Laird random-effects meta-analysis was conducted to make updated direct (pairwise) comparisons. A p-value of interaction was computed to assess differences by choice of doublet therapy (ARPi vs. docetaxel) volume of disease and timing of metastatic presentation. A p-value <0.1 was considered statistically significant for subgroup differences. Cochran’s Q statistical test described statistically significant heterogeneity not explained by chance.

A trial-level network meta-analysis [24, 25] using both direct and indirect evidence was conducted within the frequentist framework. Mixed treatment comparisons were updated and were made for pre-specified subgroups of interest (SHV, MHV, SLV, MLV). Analyses which grouped each treatment regimen into triplet therapy, ARPI doublet or docetaxel doublet therapy were conducted to assess comparative effectiveness among treatment class groups (ARPI+D+ADT, ARPI+ADT, D+ADT).

Both fixed-effect and random-effects models were fitted, and the final choice of model was made based on the network geometry and sparsity of direct evidence. The fixed effect model was used if the network was open and sparse, given that the common between-study heterogeneity cannot be estimated reliably in such networks [26]. P-scores were computed to assess relative treatment rankings for overall survival. A higher P-score indicated potentially better survival for a given treatment compared with other treatment options. Treatment rankings were interpreted in congruency with the pairwise estimates.

All statistical analyses were conducted using the ‘Analyzer’ module enabled on the R project for statistical computing (version 4.1.1). Pairwise and network meta-analyses were carried out using meta (version 5.1–1), and netmeta (version 2.0–1), respectively.

2.4. Summary of findings

The relative effect estimates along with their 95% confidence intervals pushed from the analysis module to the ‘Tabulator’ module are translated into intervention risk, and absolute risk differences using relative estimates and assumed baseline event risk. The absolute risk difference per 1000 patients using HR is calculated as:

$$RR=\frac{(1-e^{HR\times \ln \left(1-baselinerisk\right)})}{baselinerisk}$$

$$ARD=1000\times baselineeventrisk\times (RR-1)$$

3. Results

The updated report of living systematic review includes data from 11 clinical trials published as of February 1^st, 2025. The study selection process is shown in the PRISMA flowchart (Supplementary Figure 1). Results are available on an online interactive website (living evidence)[27]

3.1. Baseline trial and population characteristics

3.1.1. Overall population

The current report of this LISR includes a total of 11 trials with 12668 patients and twelve unique treatments. All eligible trials followed a phase III randomized design, six (PEACE-1, STAMPEDE, CHAARTED, GETUG-AFU15, ENZAMET, SWOG 1216) were open-label while five (ARASENS, LATITUDE, TITAN, ARCHES, ARANOTE) were double-blinded. Triplet regimens, including the combination of enzalutamide with docetaxel plus ADT (E+D+ADT), darolutamide with docetaxel plus ADT (DARO+D+ADT), abiraterone acetate-prednisone with docetaxel plus ADT (AAP+D+ADT), enzalutamide with abiraterone plus ADT (E+AAP+ADT) were assessed in one trial each. Among doublet regimens – in terms of chemotherapy doublet - docetaxel plus ADT was assessed in three trials. In terms of ARPI doublet therapy, abiraterone acetate-prednisone plus ADT (AAP+ADT) in two, enzalutamide plus ADT in two, apalutamide plus ADT (APA+ADT) in one, and darolutamide plus ADT (DARO+ADT) in one. The overall risk of bias for OS was low with some concerns for PFS due to open label trial in some trials (Supplementary Table 1). Detailed baseline characteristics are available as Supplementary Table 2 and are also available on an online interactive website (living evidence)

3.1.2. Pre-specified subgroups by volume of disease and timing of metastatic presentation

Analysis by pre-specified subgroups for this report included a total of 8990 patients and eight unique treatment options from nine clinical trials [3, 9, 15, 18, 20, 28–31]. Three trials (STAMPEDE arm C, SWOG 1216, and ARANOTE) did not report analyzable estimates for any of four pre-specified subgroups and hence were excluded from the respective analyses. PEACE-1[9] reported data for both the overall population and patients who received docetaxel; however, only data for patients who received docetaxel were used for analysis since data by subgroups for patients who did not receive docetaxel was inconsistently reported.

Of the total population in the trials eligible for analysis by prognostic subgroups, approximately 5,171 (57%) patients had SHV, 2,455 (27%) SLV, 589 (6.5%) MHV, and 775 (8.5%) patients had MLV. Detailed results are shown in Table 1 and Supplementary Tables 3–4

Table 1.

Baseline trial and population characteristics of the studies included in the analysis for pre-specified subgroup of interests

Study	Treatment arms		Total participants	Median age (years)	Median follow up (months)	Prognostic Subgroups - N (%)				Docetaxel (%)	Median OS -(months) c
Year	Experimental	Control				SHV	SLV	MHV	MLV
CHAARTED 2018	D+ADT	ADT	790	Rx: 64; Control: 63	54	421 (53)	154 (20)	91 (12)	123 (16)	100	Rx: 57.6; Control: 47.2
GETUG-AFU15 2018	D+ADT	ADT	385	Rx: 63; Control: 64	84	153 (40)	119 (31)	29 (7.6)	79 (21)	100	Rx: 62.1; Control: 48.6
LATITUDE 2019	AAP+ADT	ADT	1199	Rx: 68; Control: 67	52	955 (80) a	243 (20) a	0 (0)		0	Rx: 53.3; Control: 36.5
STAMPEDE ARM-G 2019	AAP+ADT	ADT	901	Rx: 67; Control: 67	96	486 (54)	373 (41)	13 (1.4)	29 (3.2)	0	Rx: 79.0 d; Control: 46.0 d
ARCHES 2022	E+ADT	ADT	1150	Rx: 70; Control: 70	45	606 (53) b	284 (25) b	116 (10) b	130 (11) b	18	Rx: NR; Control: NR
PEACE-1 2022	AAP+D+ADT	D+ADT	1172	Rx: 67; Control: 66	46	667 (57) a	505 (43) a	0 (0)		61	Rx: 68.4; Control: 56.4
ARASENS 2023	DARO+D+ADT	D+ADT	1305	Rx: 67; Control: 67	~44	877 (68) b	247 (19) b	117 (9.1) b	51 (3.9) b	100	Rx: NR; Control: 48.9
ENZAMET 2023	E+ADT	NSAA + ADT	622	Rx: 69; Control: 69	68	169 (27)	153 (25)	74 (12)	227 (36)	0	Rx: NR; Control: NR
	E+D+ADT	NSAA+D+ADT	503			270 (54)	92 (18)	89 (18)	52 (10)	100	Rx: NR; Control: ~62e
TITAN 2023	APA+ADT	ADT	1052	Rx: 69; Control; 68	44	567 (54) b	285 (27) b	60 (5.7) b	84 (7.9) b	11	Rx: NR; Control: 52.2

Abbreviations: Rx: Treatment; ~: Approximate value; 95% CI: 95% Confidence interval; ADT: Androgen deprivation therapy; AAP: Abiraterone acetate and Prednisone; APA: Apalutamide; CHAARTED: Chemo Hormonal Therapy vs Androgen Ablation Randomized Trial for Extensive Disease in Prostate Cancer; D: Docetaxel; DARO: Darolutamide; E: Enzalutamide; ENZAMET: Enzalutamide in First Line Androgen Deprivation Therapy for Metastatic Prostate Cancer; NR: Not reached; HV: High volume disease; LV: Low volume disease; NA: Not available; NSAA: Non-steroidal anti-androgen; STAMPEDE: Systemic Therapy in Advancing or Metastatic Prostate Cancer: Evaluation of Drug Efficacy; TITAN; Targeted Investigational Treatment Analysis of Novel Anti-androgen

^aLATITUDE, and PEACE-1 did not include patients with metachronous metastases and GETUG-AGU15 and ARASENS did not provide analyzable estimates for timing of metastasis in high volume patients and metachronous low volume disease patients respectively.

^bThe relative percentages of patients in prognostic subgroups in ARASENS, ARCHES and TITAN do not add up to 100. A total of 13 patients in ARASENS, 14 patients in ARCHES and 56 patients in TITAN had missing scans, unknown or undetermined status of metastatic presentation at initial diagnosis. Therefore, a total of 8990 patients were included in the current analysis.

^cOutlines the median overall survival in the overall patient population.

^dEstimates for median overall survival were extracted from the 5-year follow up analysis of STAMPEDE ARM-G.

^eApproximated using Kaplan-Meier curves.

3.2. Direct pairwise comparisons

Pairwise comparisons, using new data from ARANOTE trial, to assess differences by choice of doublet therapy (ARPi vs. docetaxel), volume of disease (high vs. low) and timing of metastatic presentation (synchronous vs. metachronous) were consistent with previous report [1]. Detailed pairwise results are available on an online interactive website (living evidence).

3.3. Mixed treatment comparisons

A total of 11 trials contributed to the network in the overall population (Supplementary Figure 2). Eight trials contributed to the network for the SHV subgroup, nine to SLV, and five trials contributed to both the MHV and MLV subgroups, respectively (Supplementary Figure 3). Results of the fixed-effect NMA are reported here and available in Figures 1–3, and Supplementary Figures 4–7. Random-effects NMA showed consistent results and are reported in Supplementary Figures 8–12. Updated analyses by clinically relevant subgroups are available on an online interactive website (living evidence).

Figure 1. Mixed treatment comparisons for OS and PFS in overall population by treatment class.

Abbreviations: ADT: androgen deprivation therapy; ARPI: androgen receptor pathway inhibitors; D: docetaxel; HR: hazard ratio; CI: confidence intervals.

^a League table (left), forest plot (right). In league tables, the values in each cell represent the relative treatment effect (and 95% CI) of the treatment on the top, compared to the treatment on the left. Green color suggests relative treatment benefit. Light green suggests non-significant benefit and dark green suggests significant benefit. Red color suggests relative treatment harm. Light red suggests non-significant harm and dark red suggests significant harm. The "ARPI+D+ADT" triplet refers specifically to regimens with concurrent administration of androgen deprivation therapy, androgen receptor pathway inhibitor, and docetaxel. At this time, apalutamide is not included in this category due to the lack of published data on concurrent dosing with ADT and docetaxel.

Figure 3. Mixed treatment comparisons by class across prognostic subgroups.

Abbreviations: ADT: androgen deprivation therapy; ARPI: androgen receptor pathway inhibitors; D: docetaxel; HR: hazard ratio; CI: confidence intervals.

^a League table (left), forest plot (right). In league tables, the values in each cell represent the relative treatment effect (and 95% CI) of the treatment on the top, compared to the treatment on the left. Green color suggests relative treatment benefit. Light green suggests non-significant benefit and dark green suggests significant benefit. Red color suggests relative treatment harm. Light red suggests non-significant harm and dark red suggests significant harm. The "ARPI+D+ADT" triplet refers specifically to regimens with concurrent administration of androgen deprivation therapy, androgen receptor pathway inhibitor, and docetaxel. At this time, apalutamide is not included in this category due to the lack of published data on concurrent dosing with ADT and docetaxel.

3.3.1. Overall population

When analyzed by treatment class, the combination of ARPI+D+ADT led to statistically significant improvement in OS when compared to ARPI+ADT (HR: 0.81; 95% CI: 0.68–0.96), D+ADT (HR: 0.73; 95% CI: 0.64–0.82), and ADT (HR: 0.56; 95% CI: 0.48–0.66). The combination of ARPI+ARPI+ADT led to statistically significant improvement in OS only when compared to ADT (HR: 0.65; 95% CI: 0.55–0.77). There were no statistically significant differences between ARPI+D+ADT and ARPI+ARPI+ADT, and between ARPI+ADT and D+ADT as shown in Figure 1.

When analyzed by treatment regimen, the pattern of results for OS was consistent with analysis by treatment class (Figure 2). Similar results were observed for PFS.

Figure 2. Mixed treatment comparisons for OS and PFS in overall population by treatment regimen.

Abbreviations: AAP: abiraterone acetate and prednisone; ADT: androgen deprivation therapy; APA: apalutamide; DARO: darolutamide; D: docetaxel; E: enzalutamide; NSAA: non-steroidal anti androgen HR: hazard ratio; CI: confidence intervals.

^a League table (left), forest plot (right). In league tables, the values in each cell represent the relative treatment effect (and 95% CI) of the treatment on the top, compared to the treatment on the left. Green color suggests relative treatment benefit. Light green suggests non-significant benefit and dark green suggests significant benefit. Red color suggests relative treatment harm. Light red suggests non-significant harm and dark red suggests significant harm. Sensitivity analyses limited to trials with strict radiographic progression free-survival were also conducted and showed a consistent pattern of results as shown in Supplementary Figures 15–18.

3.3.2. Analyses by pre-specified prognostic subgroups

Overall survival was analyzed and is reported here; PFS was not consistently reported across the eligible trials by pre-specified prognostic subgroups. Three trials (STAMPEDE arm C, SWOG 1216, and ARANOTE) did not report analyzable estimates for OS any of four pre-specified subgroups and hence were excluded from the following subset analyses.

3.3.2.1. Synchronous high-volume disease

When analyzed by treatment class, ARPI+D+ADT (P-score: 0.99) was ranked as the potentially most efficacious treatment with regards to OS improvement (Figure 3 and Table 2). The combination of ARPI+D+ADT led to statistically significant improvement in OS when compared to ARPI+ADT (HR: 0.71; 95% CI: 0.53–0.97), and D+ADT (HR: 0.72; 95% CI: 0.62–0.83). There was no statistically significant difference between ARPI+ADT and D+ADT (HR: 1.01; 95% CI: 0.76–1.32).

Table 2.

Summary of findings outlining absolute risk differences for overall survival by class ^a.

Comparators	Triplet therapy
	Synchronous High Volume	Synchronous Low Volume	Metachronous High Volume	Metachronous Low Volume
Androgen receptor pathway inhibitor doublet	101 fewer deaths per 1000	18 more deaths per 1000	33 fewer deaths per 1000	78 more deaths per 1000
SHV: 463 deaths per 1000	(from 172 fewer to 10 fewer)	(from 83 fewer to 158 more)	(from 222 fewer to 274 more)	(from 93 fewer to 504 more)
SLV: 225 deaths per 1000	HR: 0.71 (0.53–0.97)	HR: 1.08 (0.65–1.79)	HR: 0.89 (0.36–2.22)	HR: 1.56 (0.40–6.25)
MHV: 317 deaths per 1000	2164 patients (7 RCTs)	1055 patients (7 RCTs)	357 patients (4 RCTs)	236 patients (3 RCTs)
MLV: 209 deaths per 1000	Rank 3	Rank 1	Rank 2	Rank 1
Docetaxel doublet	97 fewer deaths per 1000	61 fewer deaths per 1000	30 fewer deaths per 1000	54 fewer deaths per 1000
SHV: 531 deaths per 1000	(from 135 fewer to 57 fewer)	(from 114 fewer to 2 more)	(from 131 fewer to 98 more)	(from 129 fewer to 168 more)
SLV: 340 deaths per 1000	HR: 0.72 (0.62–0.83)	HR: 0.74 (0.53–1.01)	HR: 0.90 (0.60–1.36)	HR: 0.64 (0.18–2.28)
MHV: 468 deaths per 1000	1817 patients (3 RCTs)	757 patients (3 RCTs)	352 patients (2 RCTs)	111 patients (1 RCTs)
MLV: 279 deaths per 1000	Rank 2	Rank 3	Rank 3	Rank 4
Androgen deprivation therapy	206 fewer deaths per 1000	73 fewer deaths per 1000	113 fewer deaths per 1000	48 fewer deaths per 1000
SHV: 569 deaths per 1000	(from 255 fewer to 139 fewer)	(from 138 fewer to 20 more)	(from 251 fewer to 125 more)	(from 129 fewer to 200 more)
SLV: 334 deaths per 1000	HR: 0.45 (0.34–0.61)	HR: 0.69 (0.44–1.09)	HR: 0.65 (0.29–1.47)	HR: 0.68 (0.18–2.56)
MHV: 412 deaths per 1000	2404 patients (8 RCTs)	1139 patients (9 RCTs)	403 patients (5 RCTs)	295 patients (3 RCTs)
MLV: 307 deaths per 1000	Rank 4	Rank 4	Rank 4	Rank 3
	Androgen receptor pathway inhibitor doublet
Triplet therapy	118 more deaths per 1000	14 fewer deaths per 1000	31 more deaths per 1000	70 fewer deaths per 1000
SHV: 428 deaths per 1000	(from 10 more to 230 more)	(from 92 fewer to 100 more)	(from 159 fewer to 337 more)	(from 172 fewer to 238 more)
SLV: 264 deaths per 1000	HR: 1.40 (1.03–1.90)	HR: 0.93 (0.56–1.54)	HR: 1.12 (0.45–2.78)	HR: 0.64 (0.16–2.53)
MHV: 380 deaths per 1000	2164 patients (7 RCTs)	1055 patients (7 RCTs)	352 patients (4 RCTs)	236 patients (3 RCTs)
MLV: 160 deaths per 1000	Rank 1	Rank 2	Rank 1	Rank 2
Docetaxel doublet	3 more deaths per 1000	64 fewer deaths per 1000	5 more deaths per 1000	117 fewer deaths per 1000
SHV: 531 deaths per 1000	(from 86 fewer to 97 more)	(from 112 fewer to 2 more)	(from 159 fewer to 264 more)	(from 154 fewer to 58 fewer)
SLV: 340 deaths per 1000	HR: 1.01 (0.76–1.32)	HR: 0.69 (0.47–1.01)	HR: 1.02 (0.45–2.28)	HR: 0.41 (0.24–0.70)
MHV: 468 deaths per 1000	2403 patients (6 RCTs)	1108 patients (7 RCTs)	393 patients (4 RCTs)	297 patients (5 RCTs)
MLV: 279 deaths per 1000	Rank 2	Rank 3	Rank 3	Rank 4
Androgen deprivation therapy	139 fewer deaths per 1000	72 fewer deaths per 1000	74 fewer deaths per 1000	113 fewer deaths per 1000
SHV: 569 deaths per 1000	(from 165 fewer to 110 fewer)	(from 101 fewer to 41 fewer)	(from 147 fewer to 23 more)	(from 143 fewer to 70 fewer)
SLV: 334 deaths per 1000	HR: 0.63 (0.57–0.70)	HR: 0.65 (0.52–0.80)	HR: 0.73 (0.49–1.09)	HR: 0.43 (0.29–0.64)
MHV: 412 deaths per 1000	2990 patients (5 RCTs)	1490 patients (5 RCTs)	444 patients (3 RCTs)	481 patients (3 RCTs)
MLV: 307 deaths per 1000	Rank 4	Rank 4	Rank 4	Rank 3

Abbreviations: SHV: synchronous high volume; SLV: synchronous low volume; MHV: metachronous high volume; MLV: metachronous low volume; HR: hazard ratio; RCT: randomized controlled trials.

^aThis table outlines relative and absolute risk estimates of death with triplet therapy and androgen receptor pathway inhibitors. The comparison is directed as treatment (in column) versus comparators (in rows). Each sub column represents the subgroup (synchronous high volume; synchronous low volume; metachronous high volume; metachronous low volume). The relative effect estimates along with their 95% confidence intervals are translated into corresponding intervention risk. Assumed baseline risk of death for each comparator by subgroups is computed as mean of risk of death within the study-follow-up period from the included trials. Absolute risk difference is calculated as the absolute risk difference between corresponding intervention risk and assumed comparator risk. Treatment ranks provided in the cells reflect the rankings of each comparator in a given comparison for a given subgroup. For example, in comparison between triplet therapy and androgen receptor pathway inhibitor doublet therapy in synchronous high-volume subgroup, rank 3 indicates the relative treatment ranking for androgen receptor pathway inhibitor doublet. The "ARPI+D+ADT" triplet refers specifically to regimens with concurrent administration of androgen deprivation therapy, androgen receptor pathway inhibitor, and docetaxel. At this time, apalutamide is not included in this category due to the lack of published data on concurrent dosing with ADT and docetaxel.

When analyzed by treatment regimen, DARO+D+ADT led to a statistically significant improvement in OS when compared to D+ADT (HR: 0.69; 0.57–0.85), E+ADT (HR: 0.67; 95% CI: 0.46–0.99), and APA+ADT (HR: 0.65; 95% CI: 0.43–0.96). There was no statistically significant difference between DARO+D+ADT and AAP+ADT (HR: 0.72; 95% CI: 0.51–1.02). The combination of AAP+D+ADT led to a statistically significant improvement in OS when compared to D+ADT (HR: 0.72; 95% CI: 0.55–0.95), and ADT (HR: 0.45; 95% CI: 0.31–0.66) but not compared to ARPI doublet regimens. The combination of E+D+ADT led to a statistically significant improvement in OS compared to ADT (HR: 0.50; 95% CI: 0.33–0.75) but no statistically significant differences were observed between E+D+ADT and doublet regimens. There were no statistically significant differences when triplet regimens were compared. Ranking analysis showed that triplet regimens which included AAP+D+ADT (P-score: 0.85), DARO+D+ADT (P-score: 0.89), E+D+ADT (P-score: 0.73) were potentially more efficacious treatment options in terms of OS improvement when compared to doublet regimens in patients with SHV mHSPC. Ranking analysis and detailed results are provided in Supplementary Figure 4 and Supplementary Table 7–8.

3.3.2.3. Synchronous low-volume disease

When analyzed by treatment class (Figure 3 and Table 2), ARPI+ADT led to statistically significant improvement in OS compared to ADT (HR: 0.65; 95% CI: 0.52–0.80) but not compared to D+ADT (HR: 0.69; 95% CI: 0.47–1.01) in patients with SLV mHSPC. There was no statistically significant difference between ARPI+D+ADT and ARPI+ADT (HR: 1.08; 95% CI: 0.65–1.79).

When analyzed by treatment regimen, E+ADT (HR: 0.62; 95% CI: 0.42–0.91) and AAP+ADT (HR: 0.66; 95% CI: 0.49–0.89) led to statistically significant improvement in OS compared to ADT. No statistically significant improvement in OS was observed for APA+ADT (HR: 0.65; 95% CI: 0.40–1.05), AAP+D+ADT (HR: 0.78; 95% CI: 0.43–1.43), DARO+D+ADT (HR: 0.71; 95% CI: 0.38–1.32), E+D+ADT (HR: 0.53; 95% CI: 0.25–1.14) when compared to ADT. Ranking analysis and detailed results are provided in Supplementary Figure 5; and Supplementary Table 7–8.

3.3.2.3. Metachronous high-volume disease

The MHV population was the smallest of the four groups (589 patients) representing 6.5% of the total.

When analyzed by treatment class (Figure 3 and Table 2), no statistically significant improvement in OS was observed with ARPI+D+ADT when compared to ARPI+ADT (HR: 0.89; 95% CI: 0.36–2.22) and D+ADT (HR: 0.90; 95% CI: 0.60–1.36) in patients with MHV mHSPC.

When analyzed by treatment regimen, there was no statistically significant improvement in OS with DARO+D+ADT compared to APA+ADT (HR: 0.72; 95% CI: 0.23–2.33), E+ADT (HR: 0.67; 95% CI: 0.24–1.86), E+D+ADT (HR: 0.59; 95% CI: 0.26–1.34), and D+ADT (HR: 0.69; 95% CI: 0.39–1.23). No statistically significant differences were observed among comparisons of other treatment regimens.

The PEACE-1 and LATITUDE trials did not include patients with metachronous metastases, and STAMPEDE arm G did not report analyzable estimates. Hence, these were excluded from the network for this subgroup. Ranking analysis and detailed results are provided in Supplementary Figure 6 and Supplementary Table 7–8.

3.3.2.4. Metachronous low-volume disease

When analyzed by treatment class, ARPI+ADT (P-score: 0.91) was ranked potentially as more efficacious treatment option than D+ADT and ADT with regards to OS improvement in patients with MLV mHSPC (Figure 3 and Table 2). The combination of ARPI and ADT led to statistically significant improvement in OS when compared to ADT (HR: 0.43; 95% CI: 0.29–0.64) and D+ADT (HR: 0.41; 95% CI: 0.24–0.70). However, there was no statistically significant difference between ARPI+D+ADT and ARPI+ADT (HR: 1.56; 95% CI: 0.40–6.25)

When analyzed by treatment regimen, E+ADT led to statistically significant improvement in OS when compared to ADT (HR: 0.51; 95% CI: 0.32–0.79), and D+ADT (HR: 0.48; 95% CI: 0.27–0.85). The combination of APA+ADT led to statistically significant improvement in OS when compared to ADT (HR: 0.22; 95% CI: 0.09–0.54), and D+ADT (HR: 0.21; 95% CI: 0.08–0.55). No statistically significant differences were observed among comparisons of other treatments. Ranking analysis showed that ARPI doublet regimens were potentially more efficacious than docetaxel doublet and ADT. Detailed results are provided in Supplementary Figure 7 and Supplementary Table 7–8.

As indicated above, the PEACE-1 and LATITUDE trials did not include patients with metachronous metastases, while the STAMPEDE arm G and ARASENS trials did not report analyzable estimates, hence, these trials were excluded from network for this subgroup.

Results for trial assessing triplet therapy and summary of findings across the four prognostic subgroup by treatment regimen are available in Supplementary Tables 9–10

4. Discussion

This LISR provides a comprehensive and up-to-date assessment of the comparative effectiveness of contemporary systemic therapies in patients diagnosed with mHSPC including the new data from the recently published ARANOTE trial[21]. It outlines relative and absolute benefits in overall population and in the context of prognostic groups defined by volume of disease and timing of metastasis[27]. Prior evidence has suggested that patients across these prognostic subgroups have variable survival on ADT alone with the longest median survival of ~3 years in SHV, ~4.5 years in patients with either SLV or MHV, and ~8 years in patients with MLV disease[17, 32]

In patients with SHV mHSPC, our analysis indicates evidence of OS benefit with triplet therapy with the combination of ARPI, docetaxel, and ADT over doublets with either docetaxel or ARPI therapy. There were no statistically significant differences among different triplet regimens with respect to OS improvement, indicating that the findings are a “class-effect” with drugs with similar mechanisms of action. These findings are reassuring and reinforce the efficacy of triplet therapy in patients with the worst prognosis, having high volume and initial presentation with distant metastases and who are fit for docetaxel[9, 18, 33]. It should also be noted that the improved efficacy observed with triplet therapy occurs at the expense of increased toxicity; and global health-related quality of life is expected to decrease during active docetaxel treatment [1, 34, 35]. However, overall quality of life tends to improve with longer follow up and after completion of chemotherapy[34]. Patient-specific factors such as age, fitness for chemotherapy[33], overall general physical health, and patient preferences should also be weighed in when making an informed decision about triplet therapy[36]. However, it should be noted that there appears to be no clear definition for docetaxel (chemotherapy) fitness[37]. Assessing patient-important outcomes, such as net clinical benefit in future clinical trials, may provide a clear picture of tradeoffs between treatment benefit and harm[38]. As such, current evidence suggests that docetaxel doublet therapy may be considered in SHV mHSPC only when access to an androgen receptor pathway inhibitor is not possible or absolute contraindications to androgen receptor pathway inhibitor doublet therapies exist[9, 18]. If a patient is not fit for docetaxel[33], an androgen receptor pathway inhibitor doublet regimen may be preferred [1] as suggested by the improved OS for patients with SHV in the ENZAMET study who were not treated with docetaxel and had a median age of 74 years versus 66 years for those treated with docetaxel[28].

In patients with SLV mHSPC, ARPI doublet or triplet therapy, may be effective treatment options. However, it is important to note that most mixed treatment comparisons for this subgroup precluded statistical significance due to the limited sample size. As a result, the absence of a clear survival benefit for triplet therapy over ARPI doublets raise questions about its utility in this setting, particularly given the added toxicity burden associated with triplet therapy. In the ENZAMET trial, docetaxel use was at the physician’s discretion before randomization, with stratification based on planned docetaxel use. Consequently, patients with SLV mHSPC receiving concurrent docetaxel may have had more aggressive disease, as reflected by worse PSA progression and prostate cancer-specific survival compared to those not receiving docetaxel. Longer-term follow-up of the PEACE-1 trial is expected to clarify whether the combination of abiraterone acetate, docetaxel and ADT improves survival in men with SLV mHSPC.

In patients with MHV mHSPC, no statistically significant differences were observed among comparisons of treatments. Metachronous high volume mHSPC is a rare situation, given that by far most patients relapsing after local treatments have oligo-metastatic low volume disease. Therefore, the proportion of patients with MHV mHSPC was small (589 patients; 6.5%) which could be a plausible explanation for lack of statistically significant differences for treatment comparisons. The effect of docetaxel may be greater in high volume disease. However, it is possible that the effect may also be increased with the bulkiness of primary disease, as defined by increasing T-stage which could explain the stronger evidence of potential benefit for patients with SHV over MHV[17]. Furthermore, it would be plausible to consider that SLV and MHV disease may represent a broader mix of patients in terms of their clinical characteristics and underlying biology as they have similar median OS of ~4.5 years on ADT alone[17, 32]. Some MHV patients may have been SHV if their diagnosis was delayed or if they had an indolent course with rising PSA post-prostatectomy before mHSPC therapy. Similarly, some SLV patients may have been SHV due to delayed initial diagnosis.

In patients with MLV mHSPC, current evidence supports the use of ARPI plus ADT over ADT alone with clear clinically and statistically significant OS benefit compared to ADT alone and D+ADT. Our results are also consistent with the STOPCAP meta-analysis[17], which showed no clinical benefit with docetaxel in patients with MLV mHSPC. This suggests that ARPI doublets are preferable treatment options in this patient population. However, it is important to note that patients with MLV disease were also rarely included in trials, which were enriched with patients with synchronous (de novo) disease or even excluded (PEACE-1 and LATITUDE). Hence, while there was no statistically significant difference between ARPI doublet and triplet therapy, the overall sample size in this subgroup was very small (~750 patients), which could have precluded statistical significance. Overall, these findings highlight the absence of clinical evidence for meaningful benefit with docetaxel and guide the clinical management of patients with MLV mHSPC to treatment with ARPI plus ADT.

While these prognostic subgroups guide systemic treatment but may not fully capture biological heterogeneity in mHSPC. Hence, biologically driven biomarkers are needed to better predict treatment response and prognosis. Consequently, efforts are underway to determine if predictive molecular biomarkers such as SPOP mutations, which render superior responses to ARPI therapy[39, 40], or PTEN/Tp53/RB1[41] mutations could potentially be used to guide treatment selection in mHSPC patients. Developing predictive biomarkers for treatment intensification could personalize care, especially for intermediate-risk disease states where the benefit of therapies like docetaxel may be more variable.

It is also noteworthy that no significant differences were detected in efficacy between ARPi agents highlighting the need for a nuanced, patient-centered approach when choosing ARPi in doublet or triplet regimens in real world settings. Cost-effectiveness remains a critical determinant, especially in regions with variable access (Supplementary Table 11). Accessibility, including drug availability and insurance coverage, can profoundly influence regimen selection in real-world practice [42]. Furthermore, the potential for drug-drug interactions, particularly in patients with polypharmacy, and the distinct toxicity profiles of ARPis, such as cardiovascular, hepatic, or fatigue-related adverse events, necessitate individualized risk-benefit assessments.

This iteration of the living network meta-analysis has several strengths. Compared to the prior literature, this report of the living network meta-analysis incorporates data from the ARANOTE trial evaluating comparative effectiveness of DARO+ADT, updated subgroup analyses from ARASENS by volume of disease, and extends the analysis to four clinically relevant prognostic subgroups (SHV, SLV, MHV, MLV) as shown in Supplementary Figures 13–14. Unlike prior meta-analyses, which are static, cross-sectional, and unidimensional this work is part of an ongoing living evidence program using a novel living interactive evidence synthesis framework[1, 22, 43]. This not only allow us to update evidence in a timely manner (as soon as new data becomes available) but also enables us to present evidence in a way that is accessible and meaningful to clinicians. These results are presented on a living interactive website (living evidence).

However, these results should be interpreted in the context of several limitations. First, this is a trial-level meta-analysis with sparse direct evidence and an open network, which did not allow us to formally assess incoherence for mixed treatment comparisons [24, 25, 44]. Network meta-regression using trial-level data was also not feasible due to the limited number of included trials at the level of each mixed treatment comparison. In contrast, an individual patient data (IPD) network meta-analysis could offer additional insights by adjusting for potential covariates across studies (Supplementary Table 11). We initially attempted to conduct an IPD network meta-analysis by gathering data from individual trials. However, restrictive data-sharing agreements limited us to perform a formal IPD network meta-analysis. Second, there are variable follow-up durations across the included trials, which could have potentially over- or under-estimated relative treatment effects in mixed treatment comparisons. Third, none of the included trials were originally designed to capture treatment efficacy by volume of disease and timing of metastatic presentation. Most trials reported these subgroups as post hoc analyses without stratification and adjustment for potential confounders. Fourth, TITAN[20] and ARCHES[3] included patients who received prior docetaxel therapy and did not report survival by receipt of docetaxel in each prognostic subgroup. Although the relative proportions of such patients in these trials were small, they could overestimate the efficacy of androgen receptor pathway inhibitor doublet regimens in the analysis. However, numerous studies[1, 45, 46] have demonstrated a consistent effect even after the exclusion of patients who received prior docetaxel in these trials (Supplementary Table 12). Fifth, rankings alone may not be a reliable indicator of treatment benefit or harm and should be carefully interpreted in congruence with pairwise estimates. Sixth there was unequal distribution of the four prognostic subgroups between trials, and patients with higher risk disease were preferentially included in the design of some trials (Table 1). Limited sample size in other prognostic subgroups could have precluded statistical significance at the level of each pairwise comparison. Seventh, it is also important to note that volume definition in the pivotal trials was purely based on conventional imaging (CT or MRI and bone scintigraphy) and the use of PSMA PET imaging may lead to a higher proportion of patients with high volume mHSPC with a risk of potential overtreatment when using triplet therapy. Eighth, access to mCRPC therapies varied by region and over time, potentially influencing comparator arm outcomes and likely treatment rankings. For example, the median OS for SHV on D+ADT (with NSAA) in ENZAMET trial was 62 months[28] compared with 40 to 49 months on D+ADT (without NSAA) for all the other trials[9, 14, 18, 28, 31, 47]. Additionally, salvage therapy use differed (Supplementary Table 13)—85% in PEACE-1 and ENZAMET vs. ~76% in ARASENS—potentially underestimating OS in some control arms. Ninth, the definition of PFS varied across the included trials. We assumed that radiographic progression typically precedes symptomatic progression, initiation of new anticancer therapies, and death from other causes. Nevertheless, a sensitivity analysis restricted to trials using radiographic PFS definitions demonstrated a consistent pattern of results (Supplementary Figures 15–18 and Supplementary Tables 14–15). Finally, this analysis did not adjust for patient and cancer-related factors influencing docetaxel use in certain trials. In ENZAMET, for instance, SHV patients treated with docetaxel were younger (median age 66 vs. 74) and had worse prognoses, while those not given docetaxel had higher non-cancer mortality, reflected in better prostate cancer–specific survival.

5. Conclusions

Current evidence from LIvE framework suggests that triplet systemic therapy is preferred for patients fit for docetaxel with SHV mHSPC. Androgen receptor pathway doublet therapy is preferred compared with ADT alone for all other subgroups with mHSPC. There is no role of docetaxel doublet in patients with access to an ARPI therapy if they are able to receive it.

Advancing practice.

What does this study add:

This living network meta-analysis investigates the role of systemic treatment intensification in metastatic hormone-sensitive prostate cancer (mHSPC) by stratifying outcomes by disease volume and timing of metastasis. Findings support the use of triplet therapy exclusively in fit patients with synchronous high-volume disease, where it yields significant survival benefit compared to other treatment options. In all other subgroups, the combination of androgen receptor pathway inhibitors with androgen deprivation therapy remains the preferred option, with no added value from docetaxel. The use of a living interactive evidence platform (living evidence) allows clinicians to continuously access the latest treatment comparisons, supporting dynamic, evidence-based decision-making for mHSPC patients as new data emerges.

Patient Summary:

For patients with metastatic hormone-sensitive prostate cancer, treatment depends on disease severity and overall health. For patients with high-volume and advanced disease at presentation who can tolerate chemotherapy, a combination of hormonal therapy (androgen receptor pathway inhibitors [ARPI]) and chemotherapy (docetaxel), in addition to ADT, offers the best survival benefit. For other patients, the combination of ARPI and ADT is the preferred treatment. Docetaxel is not beneficial if ARPI is an option.

Biographies

Daniel S. Childs (DSC): DSC has received honoraria from Targeted Oncology, IntrinsiQ, MJH Life Sciences, and the International Centers for Precision Oncology Foundation. He has served as a consultant or advisor to Janssen Biotech (institution) and Novartis (institution) and received research funding to his institution from Janssen Biotech. His travel, accommodations, and expenses have been supported by the Prostate Cancer Foundation.

Syed A. Hussain (SAH): SAH has received fees for consulting or honoraria from Janssen, Roche, Merck, Bristol-Myers Squibb (BMS), AstraZeneca, Pfizer, Astellas, GlaxoSmithKline, and Eisai. He has also received grants from Boehringer Ingelheim, Roche, Janssen, and AstraZeneca and support for attending meetings or travel from Janssen, Boehringer Ingelheim, Pfizer, Roche, BMS, AstraZeneca, and Merck Sharp & Dohme (MSD) Oncology.

Kim Chi (KC): KC has received honoraria from Amgen, Astellas Pharma, AstraZeneca, Bayer, Bristol Myers Squibb, Janssen, Merck, Novartis, Pfizer, POINT Biopharma, and Roche. He has served as a consultant or advisor to Amgen, Astellas Pharma, AstraZeneca, Bayer, Janssen, Merck, POINT Biopharma, and Roche. He has also received institutional research funding from Astellas Pharma, AstraZeneca, Bayer, Bristol Myers Squibb, ESSA Pharma, Janssen, Merck, Novartis, Pfizer, Roche, and Sanofi.

Neeraj Agarwal (NA): NA received honorarium before May 2021 and during his lifetime for consulting to Astellas, AstraZeneca, Aveo, Bayer, Bristol Myers Squibb, Calithera, Clovis, Eisai, Eli Lilly, EMD Serono, Exelixis, Foundation Medicine, Genentech, Gilead, Janssen, Merck, MEI Pharma, Nektar, Novartis, Pfizer, Pharmacyclics, and Seattle Genetics. He has also received research funding during his lifetime (to NA’s institution) from Arnivas, Astellas, AstraZeneca, Bavarian Nordic, Bayer, Bristol Meyers Squibb, Calithera, Celldex, Clovis, CRISPR Therapeutics, Eisai, Eli Lilly, EMD Serono, Exelixis, Genentech, Gilead, Glaxo Smith Kline, Immunomedics, Janssen, Lava, Medivation, Merck, Nektar, Neoleukin, New Link Genetics, Novartis, Oric, Pfizer, Prometheus, Rexahn, Roche, Sanofi, Seattle Genetics, Takeda, and Tracon.

Axel S. Merseburger (ASM): ASM has received lectures/speaker/honoraria from Ambu, Amgen, Apogepha, AstraZeneca, Astellas, Bayer, Bristol-Myers Squibb, Eisai, EUSAPharma, Farco, Ferring Ipsen, Hexal, Sandoz, MedUpdate, MSD, Merck Serono, Novartis, Janssen, Pfizer, Takeda, Novartis, Recordati, and Roche. He has also served as a consultant for Ambu, Amgen, Apogepha, AstraZeneca, Astellas, Bayer, Bristol-Myers Squibb, Eisai, EUSAPharma, Farco, Ferring Ipsen, Hexal, Sandoz, MedUpdate, MSD, Merck Serono, Novartis, Janssen, Pfizer, Takeda, Novartis, Recordati, and Roche. Moreover, he has been involved in research and clinical trials with AstraZeneca, Astellas, Bayer, Bristol-Myers Squibb, Ipsen, Janssen, EUSAPharm, MSD, Merck Serono, Novartis, Takeda, Teva, Pfizer, and Roche.

Ian D. Davis (IDD): IDD has research support to his institution for clinical trials from Amgen, Astellas Pharma, AstraZeneca, Bayer, Bristol Myers Squibb, Eisai, Janssen Oncology, MSD Oncology Pfizer, and Roche/Genentech. He serves as the Chair of ANZUP Cancer Trials Group (unremunerated). Additionally, he holds International Patent Application No: PCT/US2004/032147 (NY-ESO-1) through Ludwig Institute for Cancer Research.

Andrew Armstrong (AA): AA has received research support (to Duke) from the NIH/NCI, PCF/Movember, DOD, Astellas, Pfizer, Bayer, Janssen, Dendreon, BMS, AstraZeneca, Merck, Forma, Celgene, Amgen, and Novartis. He has consulting or advising relationships with Astellas, Pfizer, Bayer, Janssen, BMS, AstraZeneca, Merck, Forma, Celgene, Myovant, Exelixis, GoodRx, Novartis, Medscape, MJH, Z Alpha, and Telix.

Maha H. Hussain (MHH): MHH has received grant support, advisory board fees, and travel support from Pfizer, Bayer, and Genentech/Roche. She has also received grant support paid to Northwestern University, from Arvinas, lecture fees and travel from Astellas, lecture fees from Physician’s Education Resource and Sanofi/Genzyme, advisory board fees, and lecture fees from AstraZeneca. Additionally, she has received honoraria from Research to Practice and OncLive, advisory board fees from Bristol-Myers Squibb and Daiichi Sankyo, and fees for conducting an interview with UroToday. Moreover, she holds patents UM-14437/US-1/PRO 60/923,385 and UM-14437/US-2/ORD12/101,753 on systems and methods for tissue imaging, Patent 224990/10-016P2/31173361/481/671 on a method of treating cancer, and Patent 11764665.4-1464 on dual inhibition of MET and vascular endothelial growth factor for the treatment of castration-resistant prostate cancer and osteoblastic bone metastases.

Mathew Smith (MS): MS has served in a consulting or advisory role for Bayer, Janssen Oncology, Amgen, Pfizer, Lilly, and Novartis. He has also received research funding to his institution from Janssen Oncology, Bayer, Lilly, ESSA, and ORIC Pharmaceuticals.

Gerhardt Attard (GA): GA has received personal fees, grants, and travel support from Janssen and Astellas. Additionally, he has received personal fees or travel support from Pfizer, Ipsen, Novartis (Advanced Accelerator Applications), Abbott Laboratories, Ferring, ESSA Pharmaceuticals, Bayer Healthcare Pharmaceuticals, BeiGene, Takeda, AstraZeneca, and Sanofi Aventis. Moreover, he has received grant support from AstraZeneca, Innocrin Pharma, and Arno Therapeutics. He receives a share of the royalty income from The Institute of Cancer Research Rewards to Discoverers Scheme for abiraterone and holds a patent on plasma methylation signatures as biomarkers for prostate cancer (GB1915469.9).

Bertrand Tombal (BT): BT is a paid consultant and/or an investigator for Accor, Amgen, Astellas, Bayer, Myovant, Ferring, Janssens, and MSD.

Karim Fizazi (KF): KF has received consulting fees from Amgen, AstraZeneca, Astellas, Bayer, CureVac, Janssen, Novartis, Orion, Pfizer, and Sanofi. He has also received honoraria from AstraZeneca, Astellas, Bayer, Janssen, Novartis, and Sanofi. Additionally, he participates on a Data Safety Monitoring Board for Lilly.

Nick James (NJ): NJ has received grant support, drug supplies and distribution, lecture fees, and advisory board fees from Astellas Pharma and Novartis. He has also received grant support, drug supplies and distribution, and lecture fees from Pfizer, grant support and drug supplies and distribution from Clovis Oncology, and grant support, discounted drug supplies, lecture fees, advisory board fees, and travel assistance from Sanofi-Aventis.

Aurelius Omlin (AO): AO reports other support from AstraZeneca, Astellas, Bayer, Janssen, Molecular Partners, MSD, Myriad, Pfizer, Roche, and Sanofi. He has also received personal fees from Novartis, Merck, Astellas, MSD, Bayer, and Janssen outside the submitted work, research support (institutional) from Teva and Janssen, travel support from Astellas, Bayer, Janssen, and Sanofi Aventis, and compensated speakers bureau service (institutional) for Astellas, Bayer, and Janssen.

Silke Gillessen (SG): SG has served as a consultant/advisor for Active Biotech, Advanced Accelerator Applications, Amgen, Astellas Pharma, Bayer, Bristol-Myers Squibb, CellSearch, Clovis Oncology, CureVac, Dendreon, ESSA Pharmaceuticals, Ferring, Innocrin, Janssen, MaxiVAX, Millennium, Nectar, Novartis, Pfizer, and Orion. She has patents, royalties, or other intellectual property for a biomarker method (WO 2009138392 A1).

Eliezer M. Van Allen (EMVA): EMVA has reported Tango Therapeutics, Genome Medical, Genomic Life, Monte Rosa Therapeutics, Manifold Bio, Illumina, Enara Bio, Forley & Hoag, and Riva Therapeutics as well as grants from Novartis, BMS, and Janssen outside the submitted work. He has a patent filed on chromatin mutations and immunotherapy response and methods for clinical interpretation pending.

Christopher J. Sweeney (CJS): CJS has research funding paid to his institution by Janssen, Astellas, Pfizer, Sanofi, and Bayer. He has been involved in patents, consulting, or advisory roles with Sanofi, Johnson and Johnson, Astellas, Bayer, Genentech/Roche, Pfizer, Lilly; CellCentric, PointBiopharma; Amphista, QEDDI, BMS. Additionally, he has royalties and other Intellectual Property with Parthenolide (Indiana University), dimethylamino parthenolide (Leuchemix), Exelixis: Abiraterone plus cabozantinib combination, FRAS1 SNP, and tristetraprolin as biomarkers of lethal prostate cancer. He also has stock or other ownership in Leuchemix.

Alan H Bryce (AHB): AHB has received grants from Janssen and funding to his institution from Janssen, AstraZeneca, and Gilead. Additionally, he has received personal fees from AstraZeneca, Merck, Bayer, Elsevier, Fallon Medica, Horizon CME, PRIME Education, MJH Life Sciences, and Novartis outside the submitted work. He also holds a patent for therapeutic targeting of cancer patients with NRG1 rearrangements.