ABSTRACT
Background
Radiographic progression (RP) without prostate‐specific antigen (PSA) elevation has emerged as a clinically important pattern in the androgen receptor pathway inhibitors (ARPI) era, but its prevalence and prognostic significance in real‐world metastatic hormone‐sensitive prostate cancer (mHSPC) cohorts remain unclear.
Objective
To assess the prevalence and prognostic implications of RP alone progression in mHSPC patients treated with first‐line ARPIs.
Methods
We retrospectively analyzed 518 mHSPC patients treated with abiraterone acetate plus prednisolone, enzalutamide, or apalutamide. PSA progression was defined using PCWG3 criteria. Patients were categorized into PSA progression or RP alone progression groups, and clinical outcomes were compared.
Results
Among 169 patients who developed castration‐resistant prostate cancer (CRPC), 34 (20.1%) experienced RP alone. RP alone patients had significantly worse overall survival compared to those with PSA progression (median 21 vs. 39 months, p = 0.007), despite exhibiting deeper biochemical responses—PSA nadir (0.055 vs. 0.86 ng/mL, p < 0.001) and a higher rate of PSA nadir < 0.2 ng/mL (61.8% vs. 27.4%, p < 0.001). The baseline characteristic differing between groups was CHAARTED volume classification; high volume disease was markedly more common in RP alone patients (94.1% vs. 77.8%). In univariate and multivariate analyses of the overall cohort, CHAARTED high volume disease was identified as an independent risk factor for RP alone (HR 10.3, 95% CI 1.20–89.3; p = 0.034).
Conclusion
RP alone progression occurred in approximately one‐fifth of patients and was associated with significantly poorer survival despite excellent PSA responses, underscoring the limitation of PSA‐based monitoring and supporting routine imaging for early detection of PSA‐incongruent disease progression in mHSPC.
Keywords: ARPI, prostate cancer, radiographic progression
1. Introduction
Prostate cancer is one of the most prevalent malignancies among men worldwide and remains a leading cause of cancer‐related mortality. For patients with metastatic hormone‐sensitive prostate cancer (mHSPC), androgen deprivation therapy (ADT) had long been the mainstay of treatment. In recent years, several landmark trials have demonstrated the survival benefit of combining ADT with androgen receptor pathway inhibitors (ARPIs), leading to a paradigm shift in the management of mHSPC. The LATITUDE trial showed that abiraterone acetate plus prednisolone (AA + P) significantly improved overall survival (OS) and radiographic progression‐free survival (rPFS) in high‐risk mHSPC patients compared to ADT alone [1]. Similarly, the ENZAMET and ARCHES trials demonstrated that the addition of enzalutamide (Enz) to ADT conferred survival benefits in both high‐ and low‐volume disease [2, 3]. The TITAN trial further confirmed the efficacy of apalutamide (Apa) in combination with ADT across diverse mHSPC populations, regardless of disease volume or prior treatment [4]. Prostate‐specific antigen (PSA) kinetics have historically been used as surrogate markers for treatment efficacy. However, the increasing use of ARPIs has revealed a novel clinical challenge ‐radiographic progression (RP) that occurs independently of PSA progression. This discordant progression pattern suggests that some tumors may evolve to become androgen receptor (AR) independent or neuroendocrine‐like [5], leading to disease progression that is not captured by PSA monitoring alone. In a recent secondary analysis of the TITAN trial, among patients who developed RP while receiving Apa, 52.2% experienced progression in the absence of PSA progression [6]. These patients were more likely to have visceral metastases and soft‐tissue progression and exhibited worse radiographic outcomes despite favorable PSA kinetics. Similar findings were reported in post hoc analyses of the ARCHES and CHAARTED trials [7, 8]. Although this discordance has gained increasing recognition, its clinical significance in real‐world cohorts of patients with mHSPC remains inadequately characterized. In particular, the impact of ARPI treatment on the prognosis of RP progression alone remains unclear. This multicenter retrospective study was conducted to evaluate the prevalence, prognostic implications, and treatment outcomes of RP alone progression in mHSPC patients receiving first‐line ARPIs. Our goal was to determine whether PSA‐based monitoring remains adequate in current practice and to identify patients who may benefit from more vigilant radiographic surveillance.
2. Materials and Methods
2.1. Study Population
Between 2011 and 2024, a total of 662 patients with mHSPC were treated in our multicenter cohort. Data were collected from five leading university hospitals and their affiliated hospitals, that is, Kindai University Hospital (Osaka, Japan), Jikei University School of Medicine (Tokyo, Japan), Osaka Medical and Pharmaceutical University (Osaka, Japan), Tokyo Medical University Hospital (Tokyo, Japan), Kawasaki Medical School (Okayama, Japan), Hamamatsu University School of Medicine (Shizuoka, Japan) and Fujita Health University Hospital (Aichi, Japan). Study participants were required to have a confirmed diagnosis of adenocarcinoma of the prostate and to have distant metastasis on bone, lymph node, or viscera. Of these, 518 mHSPC patients treated with ARPIs as first‐line treatment were included in this study. Patients who did not treated with ARPIs or had any missing clinicopathological/laboratory information were excluded from this study. The RP in each patient was evaluated by computed tomography (CT), which was typically performed every 3–6 months as part of routine follow‐up at each institution. New or enlarging soft‐tissue metastases (e.g., lymph nodes or visceral lesions) were evaluated on CT, whereas new bone lesions were assessed by either CT findings consistent with metastasis and/or confirmatory bone scintigraphy when clinically indicated. The PSA progression group includes patients with PSA progression, based on the Prostate Cancer Working Group 2 (PCWG2) criteria [9], both with and without concomitant RP. The RP alone group includes patients with RP without PSA progression.
2.2. Ethical Considerations
This study design was approved by the Institutional Review Board of Osaka Medical and Pharmaceutical University (IRB approved number: 2571‐7) and performed in accordance with the ethical standards of the World Medical Association Declaration of Helsinki [10].
2.3. Statistical Analysis
PSA progression was defined based on the PCWG2 criteria [9], which require a confirmed rise in PSA of ≥ 25% over the nadir and an absolute increase of at least 2 ng/mL, confirmed by a second measurement obtained at least 3 weeks later. Baseline clinical and pathological characteristics were summarized using descriptive statistics. Continuous variables were expressed as medians with interquartile ranges (IQRs) and compared using analysis of variance (ANOVA). Categorical variables were presented as counts and percentages, and comparisons between groups were performed using two‐sided Fisher's exact tests. Overall survival (OS) was defined as the time from the initiation of first‐line ARPI therapy to death from any cause. Time to CRPC was defined as the duration from treatment initiation to radiographic or PSA progression during continuous ADT. OS from CRPC was calculated from the date of CRPC diagnosis to death from any cause. Survival analyses were conducted using the Kaplan–Meier method, and differences between survival curves were assessed using the log‐rank test. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards regression models. Subgroup analyses were conducted to further explore the association between progression pattern (PSA progression vs. RP alone) and survival outcomes across different clinical strata, including type of ARPI used (abiraterone, Enz, Apa), baseline PSA level, metastatic burden (low vs. high volume), LATITUDE risk classification, and CHAARTED volume classification. Interaction terms were not formally tested due to limited statistical power. All statistical analyses were performed using GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA) and JMP Pro (version 18.2.1, SAS Institute Inc., Cary, NC, USA). A p < 0.05 was considered statistically significant.
3. Results
Table 1 summarizes the clinicopathologic characteristics of 518 patients according to first‐line ARPI treatment. The majority received AA + P (62.9%), while Enz and Apa were given to 18.7% and 18.3%, respectively. The median age was 73 years (IQR, 69–78), with 42.9% of patients aged ≥ 75 years. The median follow‐up period for the overall cohort was 21 months (IQR, 12–34). By treatment group, the median follow‐up was 23 months (IQR, 14–39) for AA + P, 22 months (IQR, 13–32) for Enz, and 16 months (IQR, 10–25) for Apa. Most patients presented with high risk (77.6%) and high volume (74.5%). Gleason pattern 5 was present in 305 patients (58.9%). Bone metastases were present in 86.7%, with visceral involvement in 27.2%. Significant between‐group differences were observed in the presence of bone metastases and LATITUDE risk and CHAARTED volume classifications.
Table 1.
Clinical variables treated by first‐line doublet therapy using ARPIs among 518 mHSPC patients.
| Variables | Type of the first‐line ARPI | |||||
|---|---|---|---|---|---|---|
| Total (n = 518) | AA + P (n = 326) | ENZ (n = 97) | APA (n = 95) | |||
| n (%) | n (%) | n (%) | n (%) | p‐value | ||
| Age (years) | Median (IQR) | 73 (69–78) | 73 (68–78) | 73 (70–77) | 75 (71–80) | 0.020* |
| < 75 years | 296 (57.1) | 190 (58.3) | 62 (63.9) | 44 (46.3) | 0.039* | |
| ≥ 75 years | 222 (42.9) | 136 (41.7) | 35 (36.1) | 51 (53.7) | ||
| Gleason pattern | ||||||
| < 5 | 193 (37.2) | 116 (35.6) | 39 (40.2) | 38 (40.0) | 0.296 | |
| 5 | 305 (58.9) | 204 (62.6) | 51 (52.6) | 50 (52.6) | ||
| Unknown | 20 (3.9) | 6 (1.8) | 7 (7.2) | 7 (7.4) | ||
| Clinical T stage | ||||||
| < T4 | 328 (63.3) | 208 (63.8) | 56 (57.7) | 64 (67.4) | 0.280 | |
| T4 | 184 (35.5) | 116 (35.6) | 40 (41.3) | 28 (29.4) | ||
| Unknown | 6 (1.2) | 2 (0.6) | 1 (1.0) | 3 (3.2) | ||
| PSA levels (ng/mL) | ||||||
| Median (IQR) | 261 (68.0–1000) | 359 (98.5–1307) | 185 (55.3–828) | 91.0 (18.0–500) | 0.270 | |
| ECOG‐PS | ||||||
| 0 | 343 (66.2) | 217 (66.6) | 65 (67.50 | 61 (64.2) | 0.903 | |
| ≥ 1 | 175 (33.8) | 109 (33.4) | 32 (33.0) | 34 (35.8) | ||
| de novo bone metastasis | ||||||
| No | 69 (13.3) | 25 (7.7) | 23 (23.7) | 21 (22.1) | < 0.001* | |
| Yes | 448 (86.7) | 300 (92.3) | 74 (76.3) | 74 (77.9) | ||
| de novo Visceral metastasis | ||||||
| No | 377 (72.8) | 229 (70.2) | 77 (79.4) | 71 (75.8) | 0.194 | |
| Yes | 141 (27.2) | 97 (29.8) | 20 (20.6) | 24 (25.3) | ||
| Sites of metastasis | ||||||
| Lung | 128 (24.7) | 89 (27.3) | 17 (17.5) | 22 (23.2) | 0.139 | |
| Liver | 15 (2.9) | 12 (3.7) | 1 (1.0) | 2 (2.1) | 0.417 | |
| LATITUDE criteria | ||||||
| Low risk | 116 (22.4) | 36 (11.0) | 36 (37.1) | 44 (46.3) | < 0.001* | |
| High risk | 402 (77.6) | 290 (89.0) | 61 (62.9) | 51 (53.7) | ||
| CHAARTED criteria | ||||||
| Low volume | 132 (25.5) | 56 (17.2) | 33 (34.0) | 43 (45.3) | < 0.001* | |
| High volume | 386 (74.5) | 270 (82.8) | 64 (66.0) | 52 (54.7) | ||
Note: LATITUDE high risk had least two of the three following high‐risk factors: a Gleason sum score of 8 or more, at least three bone lesions, and the presence of measurable visceral metastasis. CHAARTED high volume had the presence of visceral metastases or > 4 bone lesions with one or more outside the vertebral bodies and pelvis.
Abbreviations: AA + P, abiraterone acetate + prednisolone; APA, apalutamide; ARPIs, androgen receptor pathway inhibitors; ECOG‐PS, Eastern Cooperative Oncology Group performance status; ENZ, enzalutamide; IQR, interquartile range; mHSPC, metastatic hormone‐sensitive prostate cancer; PSA, prostate‐specific antigen.
Denotes p < 0.05.
Among all 518 patients, 169 patients (32.6%) developed CRPC, including 135 (26.0%) with PSA progression and 34 (6.6%) with RP alone. All cases of RP were diagnosed based on these scheduled CT examinations rather than symptom‐driven imaging. By first‐line ARPI treatments, PSA progression occurred in 28.8% with AA + P, 17.5% with Enz, and 25.3% with Apa, while RP alone occurred in 7.4%, 5.2%, and 5.3%, respectively (Figure 1). Kaplan–Meier curves showed significantly shorter OS in patients with RP alone compared to PSA progression (median OS, 21 vs. 39 months; HR 0.50, 95% CI 0.26–0.96, p = 0.007; Figure 2A). Time to CRPC was also assessed. Contrary to expectations, the median time to CRPC was similar between RP alone and PSA progression (median time 12 vs. 11 months, HR 0.74, 95% CI 0.49–1.12, p = 0.100; Figure 2B). In contrast, OS after CRPC was significantly shorter in patients with RP alone (median 7 months) compared to those with PSA progression (19 months), although (HR 0.51, 95% CI 0.26–0.98, p = 0.007; Figure 2C). Consistent with previous reports [11], Kaplan–Meier curves showed significantly shorter OS in patients with combined PSA and RP (PSA + RP) compared with those with PSA progression alone (PSA alone) (median OS, 29 vs. 49 months; HR 0.57, 95% CI 0.33–0.98; p = 0.023). In contrast, no significant difference in OS was observed between the RP alone and the PSA + RP (median OS, 21 vs. 29 months; HR 1.36, 95% CI 0.73–2.53; p = 0.300) (Supporting Information: Figure S1).
Figure 1.
Patterns of disease progression in patients with mHSPC treated with first‐line ARPIs. Distribution of progression patterns among the total cohort (n = 518) and stratified by type of ARPI treatments: AA + P, Enz, and Apa. The proportions of patients showing no progression, PSA progression, and RP alone are illustrated. 169 patients (32.6%) developed CRPC, including 135 (26.0%) with PSA progression and 34 (6.6%) with PR alone. PSA progression occurred in 28.8% with AA + P, 17.5% with Enz, and 25.3% with Apa, while RP alone occurred in 7.4%, 5.2%, and 5.3%, respectively. [Color figure can be viewed at wileyonlinelibrary.com]
Figure 2.
Kaplan–Meier survival analyses for CRPC patients according to progression pattern. Kaplan–Meier curves comparing outcomes between patients with PSA progression and those with radiographic progression (RP) alone. (A) Overall survival (OS) from initiation of first‐line ARPI therapy. (B) Time to development of CRPC. (C) OS measured from the onset of CRPC. [Color figure can be viewed at wileyonlinelibrary.com]
To identify poor prognostic factors following progression to CRPC with RP alone, we further analyzed prognostic factors in patients with PSA progression (n = 135) and RP alone (n = 34), based on the result that post‐CRPC survival influence OS. Table 2 summarizes the baseline clinicopathologic characteristics of these 169 patients. The median age at primary diagnosis was 73 and 75 years, respectively (p = 0.357), and the distributions of age categories (< 75 vs. ≥ 75 years), Gleason pattern, and clinical T stage were comparable. Median baseline PSA was relatively lower in the RP alone group (180 vs. 307 ng/mL, p = 0.099). Similarly, no significant differences were found in Gleason pattern, clinical T stage, ECOG performance status, EOD score, or sites of metastasis. Histologically confirmed neuroendocrine prostate cancer was observed in only one patient in the RP alone group. In the PSA progression group, 14 patients (10.4%) received radiotherapy to the primary prostate lesion and 21 patients (15.6%) to metastatic sites during the disease course. In the RP alone group, 6 patients (17.7%) received radiotherapy to metastatic lesions, whereas no patients received radiotherapy to the primary tumor.
Table 2.
Clinical variables of 169 CRPC patients progressing after first‐line doublet therapy using ARPIs.
| CRPC patients (n = 169) | ||||
|---|---|---|---|---|
| Variables | PSA progression (n = 135) | RP alone (n = 34) | p‐value | |
| n (%) | n (%) | |||
| Baseline characteristics | ||||
| Age (years) | ||||
| Median (IQR) | 73 (67–78) | 75 (71–79) | 0.357 | |
| < 75 years | 80 (59.3) | 17 (50.0) | 0.340 | |
| ≥ 75 years | 55 (40.7) | 17 (50.0) | ||
| Gleason pattern | ||||
| < 5 | 31 (23.9) | 9 (27.3) | 0.658 | |
| 5 | 99 (76.2) | 24 (72.7) | ||
| Clinical T stage | ||||
| < T4 | 79 (59.0) | 19 (55.9) | 0.846 | |
| T4 | 55 (41.9) | 15 (44.1) | ||
| PSA levels (ng/mL) | ||||
| Median (IQR) | 307 (75–1191) | 180 (52–1259) | 0.099 | |
| ECOG‐PS | ||||
| 0 | 83 (61.5) | 20 (58.8) | 0.845 | |
| ≥ 1 | 52 (38.5) | 14 (41.2) | ||
| de novo bone metastasis | ||||
| No | 11 (8.2) | 1 (2.9) | 0.463 | |
| Yes | 124 (91.8) | 33 (97.1) | ||
| EDO score | ||||
| 0–1 | 45 (33.3) | 90 (66.7) | 0.539 | |
| ≥ 2 | 9 (26.5) | 25 (73.5) | ||
| de novo visceral metastasis | ||||
| No | 99 (73.3) | 23 (67.6) | 0.525 | |
| Yes | 36 (26.7) | 11 (32.4) | ||
| Sites of metastasis | ||||
| Lung | 31 (23.0) | 10 (29.4) | 0.502 | |
| Liver | 5 (3.7) | 3 (8.8) | 0.202 | |
| LATITUDE criteria | ||||
| Low risk | 18 (13.3) | 4 (11.8) | 1.000 | |
| High risk | 117 (86.7) | 30 (88.2) | ||
| CHAARTED criteria | ||||
| Low volume | 30 (22.2) | 2 (5.9) | 0.029* | |
| High volume | 105 (77.8) | 32 (94.1) | ||
| Type of first‐line ARPI | ||||
| Abiraterone | 92 (68.2) | 24 (70.6) | 1.000 | |
| Enzalutamide | 20 (14.8) | 5 (14.7) | ||
| Apalutamide | 23 (17.0) | 5 (14.7) | ||
| Radiation therapy | ||||
| Prostate | 14 (10.4) | 0 (0) | 0.075 | |
| Bone metastases | 21 (15.6) | 6 (17.7) | 0.795 | |
| Characteristics at CRPC diagnosis | ||||
| Sites of radiographic progression at CRPC | ||||
| Bone | 46 (34.1) | 22 (64.7) | 0.002* | |
| Visceral | 13 (9.6) | 8 (23.5) | 0.040* | |
| docetaxel treatment after CRPC | ||||
| No | 72 (54.1) | 24 (70.6) | 0.119 | |
| Yes | 61 (45.9) | 10 (42.1) | ||
Note: LATITUDE high‐risk had least two of the three following high‐risk factors: a Gleason sum score of 8 or more, at least three bone lesions, and the presence of measurable visceral metastasis. CHAARTED high volume had the presence of visceral metastases or > 4 bone lesions with one or more outside the vertebral bodies and pelvis.
Abbreviations: AA + P, abiraterone acetate + predonisone; Apa, apalutamide; ARPIs, androgen receptor pathways inhibitors; CRPC, castration‐resistant prostate cancer; ECOG‐PS, Eastern Cooperative Oncology Group performance status; Enz, enzalutamide; EOD score, extend of disease score; IQR, interquartile range; PSA, prostate‐specific antigen; RP, radiographic progression.
Denotes p < 0.05.
Notably, while the LATITUDE risk classification showed no significant difference, when patients were classified according to the CHAARTED volume criteria, the RP alone group had a significantly higher proportion of patients with high volume compared with those with low volume; in the PSA progression group, 30 patients (22.2%) had low volume and 105 patients (77.8%) had high volume, whereas in the RP alone group, 2 patients (5.9%) had low volume disease and 32 patients (94.1%) had high volume disease (p = 0.029).
To identify risk factors for RP alone, univariate and multivariate analyses were performed in the overall cohort of 518 patients (Table 3). In univariate analyses, a higher EOD score (≥ 2) and high‐volume disease according to the CHAARTED criteria were significantly associated with RP alone. In multivariate analysis, high volume disease remained an independent risk factor for RP alone (HR 10.3, 95% CI 1.20–89.3; p = 0.034), whereas other clinical variables, including age, Gleason pattern, clinical T stage, baseline PSA level, visceral metastases, and LATITUDE risk classification, were not independently associated with RP alone.
Table 3.
Univariate and multivariate analyses of risk factors for RP alone in the overall cohort (n = 518).
| Variables | Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|---|
| HR | (95%CI) | p‐value | HR | (95%CI) | p‐value | ||
| Age (years) | < 75 | ref | ref | ||||
| ≥ 75 | 1.36 | (0.67–2.75) | 0.386 | 1.17 | (0.56–2.43) | 0.68 | |
| Gleason pattern | < 5 | ref | ref | ||||
| 5 | 1.75 | (0.82–4.05) | 0.152 | 1.78 | (0.76–4.13) | 0.182 | |
| Clinical T stage | < T4 | ref | ref | ||||
| T4 | 1.44 | (0.70–2.91) | 0.310 | 1.48 | (0.69–3.16) | 0.313 | |
| log2 PSA (continuous) | 0.96 | (0.84–1.09) | 0.518 | 0.86 | (0.74–1.00) | 0.051 | |
| EOD score | 0–1 | ref | ref | ||||
| ≥ 2 | 2.28 | (1.08–5.25) | 0.031* | 1.54 | (0.57–4.14) | 0.394 | |
| de novo visceral metastasis | No | ref | ref | ||||
| Yes | 1.30 | (0.60–2.69) | 0.494 | 0.44 | (0.04–5.05) | 0.513 | |
| de novo Lung mets | No | ref | ref | ||||
| Yes | 1.29 | (0.58–2.71) | 0.519 | 1.80 | (0.18–18.3) | 0.619 | |
| de novo Liver mets | No | ref | ref | ||||
| Yes | 3.81 | (0.84–12.8) | 0.079 | 4.21 | (0.89–20.0) | 0.071 | |
| LATITUDE criteria | Low risk | ref | ref | ||||
| High risk | 2.26 | (0.87–7.72) | 0.100 | 0.38 | (0.06–2.30) | 0.293 | |
| CHAARTED criteria | Low volume | ref | ref | ||||
| High volume | 5.88 | (1.75–36.6) | 0.002* | 10.30 | (1.20–89.3) | 0.034* | |
Note: LATITUDE high risk had least two of the three following high‐risk factors: a Gleason sum score of 8 or more, at least three bone lesions, and the presence of measurable visceral metastasis. CHAARTED high volume had the presence of visceral metastases or > 4 bone lesions with one or more outside the vertebral bodies and pelvis. EOD score, extend of disease score; PSA, prostate specific antigen; RP, radiographic progression. *Denotes p < 0.05.
We further compared the patterns of metastatic spread between patients with RP alone and those with PSA alone progression and found no significant differences in metastatic sites (Table S1). Similarly, no significant difference in OS was observed according to the receipt of radiotherapy to either the primary tumor or metastatic lesions (Supporting Information: Figure S2).
These findings indicate that tumor volume, rather than risk factors such as a high Gleason Score, may be more relevant to RP alone progression. We had hypothesis that the treatment after CRPC may affect the prognosis, however, the proportion of taxane used after second‐line treatment after CRPC was 45.9% for PSA progression and 42.1% for RP alone, with no significance (p = 0.119; Table 2).
We next examined whether biochemical responses differed according to the pattern of progression (Table 4). Notably, patients who later developed RP alone exhibited deeper PSA suppression during initial ARPI therapy. The PSA nadir was significantly lower in the RP alone group compared with the PSA‐progression group (median 0.055 vs. 0.86 ng/mL, p < 0.001), and the proportion achieving a PSA nadir < 0.2 ng/mL was more than doubled (61.8% vs. 27.4%, p < 0.001). Despite these deeper PSA responses, the rate of PSA decline ≥ 90% was similarly high in both groups (94.1% vs. 88.9%, p = 0.529). Taken together, these findings demonstrate that PSA kinetics did not discriminate between patients who experienced PSA progression and those who developed RP, indicating that excellent biochemical responses do not preclude the emergence of PSA‐incongruent RP under ARPI therapy.
Table 4.
PSA kinetics and biochemical response parameters in patients with PSA progression versus RP alone progression.
| PSA kinetics | CRPC patients (n = 169) | p value | |
|---|---|---|---|
| PSA progression (n = 135) | RP alone (n = 34) | ||
| PSA nadir (ng/mL), median (IQR) | 0.86 (0.18–6.09) | 0.055 (0.01–1.77) | < 0.001* |
| PSA nadir < 0.2, n (%) | 37 (27.4) | 21 (61.8) | < 0.001* |
| PSA decline > 90%, n (%) | 120 (88.9) | 32 (94.1) | 0.529 |
Abbreviations: ARPIs, androgen receptor pathways inhibitors; CRPC, castration‐resistant prostate cancer; IQR, interquartile range; PSA, prostate‐specific antigen; RP, radiographic progression.
Denotes p < 0.05.
4. Discussion
In this retrospective multicenter study, we evaluated the prognostic significance of RP without concurrent PSA progression in patients with mHSPC treated with first‐line androgen receptor signaling inhibitors (ARPIs). Our findings demonstrate that RP alone patients had significantly shorter OS compared to those with PSA progression (median OS: 21 vs. 39 months, p = 0.007). Importantly, at the time of mHSPC diagnosis, differences between the two groups were observed only in the CHAARTED volume classification, suggesting that tumor volume may be relevant to the distinct disease biology of RP alone. Supplementary analyses further clarified the prognostic hierarchy among progression patterns. Patients with PSA + RP had significantly shorter survival than those with PSA alone, whereas survival did not significantly differ between RP alone and PSA + RP. These findings collectively indicate that RP—whether occurring alone or together with PSA progression—is consistently associated with poor prognosis, while PSA alone progression appears to confer comparatively more favorable outcomes.
Our observations are consistent with recent post hoc analyses of major randomized trials [6]. The discordant progression cases were more likely to present with visceral metastases, soft‐tissue disease, and neuroendocrine‐like features, all of which are associated with poor prognosis [6]. Similar patterns were also observed in the ARCHES and CHAARTED studies [7, 8], suggesting that RP alone is a reproducible and clinically meaningful phenomenon across various ARPI‐based regimens.
This discordance is likely rooted in tumor heterogeneity. Advanced prostate cancer consists of a complex mixture of AR‐dependent and AR‐independent subclones. Potent ARPI therapy may rapidly suppress PSA‐producing tumor populations that are dependent on androgen signaling, while inadvertently allowing androgen‐independent clones to expand. One well‐characterized escape mechanism is neuroendocrine differentiation, which is associated with low PSA levels, aggressive progression, and poor prognosis [5]. Emerging evidence implicates genomic alterations such as TP53, RB1, and PTEN loss in promoting lineage plasticity and resistance to AR‐targeted therapies [12]. Prior studies have suggested that alterations such as TP53 and RB1 loss have been implicated in neuroendocrine‐like transformation and may contribute to RP without PSA elevation [5]. Therefore, RP alone may reflect an evolutionary shift toward AR‐independence and PSA‐silent disease, necessitating alternate monitoring and treatment strategies. We believe that future studies incorporating molecular analyses will be valuable to address these important questions.
From a clinical perspective, these findings emphasize the limitations of PSA‐based monitoring, particularly in the era of potent ARPIs. Although PSA kinetics remain a useful surrogate for treatment response, they fail to capture all instances of disease progression. The Prostate Cancer Working Group 3 (PCWG3) recommends radiographic assessments every 8–12 weeks in clinical trials, especially in the presence of symptoms or PSA‐incongruent progression [13]. However, in real‐world settings, imaging is often delayed or omitted in asymptomatic patients with low PSA levels. Our data support incorporating routine imaging into follow‐up protocols for mHSPC patients, even when PSA responses appear favorable.
This study has several limitations. First, its retrospective design and lack of standardized imaging intervals may have introduced bias in progression assessment. Second, radiographic criteria were not centrally reviewed, and imaging modalities differed across participating institutions. Nonetheless, the multicenter design and relatively large cohort strengthen the generalizability of our findings. Future prospective studies are needed to validate RP alone as a clinically meaningful subtype. Third, in the present dataset, genetic testing was not performed before disease progression because, under current Japanese insurance guidelines, genetic testing, including comprehensive genomic profiling (CGP), is not approved for patients with mHSPC in the first‐line setting. Consequently, the information on somatic or germline genetic alterations was not available for prediction of any progression. Integration of biomarker platforms, including circulating tumor DNA (ctDNA), PSMA PET imaging, and biopsy‐based genomic profiling, may aid in early detection of patients at risk for discordant progression.
Conclusion
RP without PSA elevation is associated with worse survival despite favorable PSA kinetics in mHSPC patients receiving first‐line ARPI therapy. Our cohort showed tumor volume may be relevant to RP alone progression. These findings underscore the limitations of PSA‐based monitoring and highlight the need for routine imaging and further investigation into PSA‐incongruent disease biology.
Author Contributions
Takuya Tsujino, Shogo Yamazaki, Takuya Matsuda, and Kazumasa Komura conceptualized, designed the study. Masanobu Saruta, Takafumi Yanagisawa, Keiichiro Mori, Wataru Fukuokaya, Fumihiko Urabe, Atsuhiko Yoshizawa, Saizo Fujimoto, Teruo Inamoto, Tatsuo Fukushima, Taizo Uchimoto, Ko Nakamura, Kyosuke Nishio, Keita Nakamori, Kazuki Nishimura, Tomohisa Matsunaga, Ryoichi Maenosono and Yuki Yoshikawa acquired data. Takuya Tsujino, Moritoshi Sakamoto, and Shingo Toyoda performed the majority of analyses. Takuya Tsujino, Takeshi Tsutsumi, and Takuya Matsuda wrote the manuscript and prepared figures. Kiyoshi Takahara, Kazutoshi Fujita, Teruo Inamoto, Takahiro Kimura, and Haruhito Azuma revised and supervised. Takuya Tsujino acquired funding for the present study.
Ethics Statement
This study was approved by the IRB from Osaka Medical and Pharmaceutical University (approved number: 2571‐7). This observational study adhered to ethical principles aligned with the Declaration of Helsinki, the International Conference on Harmonization's Good Clinical Practice guidelines, Good Pharmacoepidemiology Practices, and relevant laws for noninterventional and observational studies.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supplementary Figure S1: Kaplan–Meier survival analyses for CRPC patients according to radiographic progression pattern. Supplementary Figure S2: Kaplan–Meier survival analyses for CRPC patients according to receipt of radiotherapy to the primary prostate lesion or metastatic sites.
Supplemental Table 1: Site of metastasis of 117 CRPC patients who had PSA only or RP alone progressing after first‐line doublet therapy using ARPIs.
Acknowledgments
This research was partially funded by the Grant‐in‐Aid No. 25K12625 (Japan Society for the Promotion of Science: JSPS) and Grant‐in‐Aid for Research Promotion Project (Osaka Medical and Pharmaceutical University).
Tsutsumi T., Yamazaki S., Matsuda T., et al., “The Prognostic Significance of Radiographic Progression Without PSA Progression in Patients With Metastatic Hormone‐Sensitive Prostate Cancer Treated With First‐Line ARPI Therapy,” The Prostate 86 (2026): 882–890. 10.1002/pros.70157.
Takeshi Tsutsumi, Shogo Yamazaki, Takuya Matsuda and Takuya Tsujino contributed equally to this work.
Contributor Information
Takuya Tsujino, Email: takuya.tsujino@ompu.ac.jp.
Kazumasa Komura, Email: k.komura@med.kawasaki-m.ac.jp.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Fizazi K., Tran N., Fein L., et al., “Abiraterone Plus Prednisone in Metastatic, Castration‐Sensitive Prostate Cancer,” New England Journal of Medicine 377, no. 4 (2017): 352–360. [DOI] [PubMed] [Google Scholar]
- 2. Davis I. D., Martin A. J., Stockler M. R., et al., “Enzalutamide With Standard First‐Line Therapy in Metastatic Prostate Cancer,” New England Journal of Medicine 381, no. 2 (2019): 121–131. [DOI] [PubMed] [Google Scholar]
- 3. Armstrong A. J., Szmulewitz R. Z., Petrylak D. P., et al., “ARCHES: A Randomized, Phase III Study of Androgen Deprivation Therapy With Enzalutamide or Placebo in Men With Metastatic Hormone‐Sensitive Prostate Cancer,” Journal of Clinical Oncology 37, no. 32 (2019): 2974–2986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Chi K. N., Agarwal N., Bjartell A., et al., “Apalutamide for Metastatic, Castration‐Sensitive Prostate Cancer,” New England Journal of Medicine 381, no. 1 (2019): 13–24. [DOI] [PubMed] [Google Scholar]
- 5. Yamada Y. and Beltran H., “Clinical and Biological Features of Neuroendocrine Prostate Cancer,” Current Oncology Reports 23, no. 2 (2021): 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Fukuokaya W., Yanagisawa T., Mori K., et al., “Radiographic Progression Without Corresponding Prostate‐Specific Antigen Progression in Patients With Metastatic Castration‐Sensitive Prostate Cancer Receiving Apalutamide: Secondary Analysis of the TITAN Trial,” European Urology Oncology 8, no. 2 (2025): 263–269. [DOI] [PubMed] [Google Scholar]
- 7. Armstrong A. J., Mottet N., Iguchi T., et al., “Radiographic Progression in the Absence of Prostate‐Specific Antigen (PSA) Progression in Patients With Metastatic Hormone‐Sensitive Prostate Cancer (mHSPC): Post Hoc Analysis of ARCHES,” supplement, Journal of Clinical Oncology 40, no. 16_suppl (2022): 5072. [Google Scholar]
- 8. Bryce A. H., Chen Y. H., Liu G., et al., “Patterns of Cancer Progression of Metastatic Hormone‐Sensitive Prostate Cancer in the ECOG3805 CHAARTED Trial,” European Urology Oncology 3, no. 6 (2020): 717–724. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Scher H. I., Halabi S., Tannock I., et al., “Design and End Points of Clinical Trials for Patients With Progressive Prostate Cancer and Castrate Levels of Testosterone: Recommendations of the Prostate Cancer Clinical Trials Working Group,” Journal of Clinical Oncology 26, no. 7 (2008): 1148–1159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. World Medical Association m, “Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects,” Journal of the American Medical Association 310, no. 20 (2013): 2191–2194. [DOI] [PubMed] [Google Scholar]
- 11. Hussain M., Goldman B., Tangen C., et al., “Prostate‐Specific Antigen Progression Predicts Overall Survival in Patients With Metastatic Prostate Cancer: Data From Southwest Oncology Group Trials 9346 (Intergroup Study 0162) and 9916,” Journal of Clinical Oncology 27, no. 15 (2009): 2450–2456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Chen J., Shi M., Chuen Choi S. Y., et al., “Genomic Alterations in Neuroendocrine Prostate Cancer: A Systematic Review and Meta‐Analysis,” BJUI Compass 4, no. 3 (2023): 256–265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Scher H. I., Morris M. J., Stadler W. M., et al., “Trial Design and Objectives for Castration‐Resistant Prostate Cancer: Updated Recommendations From the Prostate Cancer Clinical Trials Working Group 3,” Journal of Clinical Oncology 34, no. 12 (2016): 1402–1418. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure S1: Kaplan–Meier survival analyses for CRPC patients according to radiographic progression pattern. Supplementary Figure S2: Kaplan–Meier survival analyses for CRPC patients according to receipt of radiotherapy to the primary prostate lesion or metastatic sites.
Supplemental Table 1: Site of metastasis of 117 CRPC patients who had PSA only or RP alone progressing after first‐line doublet therapy using ARPIs.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.