Androgen receptor signaling inhibitors have improved outcomes for patients with metastatic castration-sensitive prostate cancer (mCSPC), but progression to castration-resistant prostate cancer (CRPC) does not always follow the same pattern. In particular, radiographic progression may occur without a corresponding rise in prostate-specific antigen (PSA), raising questions about the prognostic implications of different progression patterns and the limitations of PSA-based monitoring.
A post hoc analysis of the phase 3 TITAN trial (NCT02489318) evaluated patterns of progression to CRPC in patients with mCSPC treated with apalutamide plus androgen deprivation therapy (ADT) or placebo plus ADT. Investigators compared PSA-defined and radiographic CRPC, assessed their association with overall survival after castration resistance, and examined baseline factors associated with each progression pattern.
The study, “Association Between Castration-resistant Prostate Cancer Progression Patterns and Survival Outcomes in Metastatic Castration-sensitive Prostate Cancer: A Post Hoc Analysis of the TITAN Trial,” was published as an article in press in European Urology Focus on September 11, 2026.
Authors: Tomohiro Hori, Hiroaki Iwamoto, Takahiro Inaba, Ryunosuke Nakagawa, Taiki Kamijima, Hiroshi Kano, Tomoyuki Makino, Renato Naito, Hiroshi Yaegashi, Kazuyoshi Shigehara, Takahiro Nohara, Kouji Izumi, and Atsushi Mizokami.
How Was the Analysis Conducted?
TITAN was a randomized, double-blind, placebo-controlled, multinational phase 3 trial evaluating apalutamide plus androgen deprivation therapy (ADT) versus placebo plus ADT in patients with mCSPC.
For the present post hoc analysis, investigators used the TITAN intention-to-treat population of 1,052 patients, including 525 assigned to apalutamide and 527 to placebo. The first interim analysis, with a data cutoff of November 23, 2018 and a median follow-up of 22.7 months, was used to define patterns of progression to CRPC. Overall survival was assessed using the final analysis dataset, with a September 7, 2020 cutoff and median follow-up of 44.0 months.
CRPC was classified into two progression patterns. PSA-CRPC was defined as PSA progression occurring at least 1 day before radiographic progression. Radiographic CRPC was defined as radiographic progression occurring concurrently with or before PSA progression. Concurrent cases were assigned to the radiographic CRPC group to account for the longer intervals between imaging assessments compared with PSA monitoring.
By the interim-analysis cutoff used to classify progression patterns, 499 patients had developed CRPC, including 162 in the apalutamide group and 337 in the placebo group. The placebo group included 61 patients who later crossed over to apalutamide. Thirty-three patients died before a CRPC diagnosis and were not included among patients classified according to CRPC progression subtype.
Apalutamide Reduced PSA-Defined, but Not Radiographic, CRPC
A marked difference emerged between the two treatment groups in the cumulative incidence of PSA-defined progression. At 24 months, the cumulative incidence of PSA-CRPC was 19% with apalutamide plus ADT compared with 52% with placebo plus ADT (95% CI, 16–23% vs 48–56%; p<0.001).
In contrast, the 24-month cumulative incidence of radiographic CRPC was similar between the two groups: 13% with apalutamide and 14% with placebo (95% CI, 11–17% in both groups; p=0.6). These findings indicate that although apalutamide substantially reduced progression defined initially by PSA, it was not associated with a statistically significant reduction in the cumulative incidence of radiographic CRPC in this analysis.
Survival Differed by Progression Pattern in the Placebo Group
The investigators next examined overall survival after the development of castration resistance. Because the proportional hazards assumption was not satisfied, survival differences were evaluated using inverse probability of treatment weighting-adjusted restricted mean survival time (RMST), with a common truncation time of 41.8 months.
Among patients in the placebo plus ADT group, radiographic CRPC was associated with significantly shorter survival than PSA-CRPC. The adjusted difference in RMST was –6.1 months (95% CI, –9.9 to –2.3; p=0.002). This difference was not observed among patients treated with apalutamide plus ADT. In this group, the adjusted RMST difference between the two progression patterns was 3.4 months (95% CI, –1.9 to 8.7; p=0.2).
The authors therefore concluded that the prognostic heterogeneity between PSA-defined and radiographic progression observed with placebo plus ADT was less evident among patients who received apalutamide.
Radiographic Progression Could Occur at Low PSA Levels
Another important finding involved PSA levels at the time CRPC developed. At the onset of PSA-CRPC, median PSA was 4.38 ng/ml in the apalutamide group compared with 5.26 ng/ml in the placebo group (p=0.044).
The difference was greater among patients developing radiographic CRPC. Median PSA at radiographic progression was 0.27 ng/ml with apalutamide, compared with 3.50 ng/ml with placebo (p<0.001). These findings show that radiographic progression can occur despite low PSA levels, particularly among patients receiving upfront apalutamide.
Visceral Metastases Were Associated With Radiographic CRPC
The investigators also assessed baseline clinical factors associated with subsequent CRPC progression patterns. Baseline visceral metastasis was consistently associated with progression to radiographic CRPC in both treatment groups. In the placebo group, visceral metastasis was associated with a higher cause-specific hazard of radiographic CRPC:
- HR 2.22; 95% CI, 1.25–3.94; p=0.007.
A similar association was observed in the apalutamide group:
- HR 2.73; 95% CI, 1.49–4.98; p=0.001.
In contrast, no baseline clinical factor was consistently associated with PSA-CRPC across both treatment groups. Complete-case sensitivity analyses produced results consistent with those from the primary analyses using imputed data. The investigators also performed a sensitivity analysis using a 28-day window to define concurrent PSA and radiographic progression. The survival findings remained consistent with those of the primary analysis.
What Do the Findings Mean?
According to the authors, the analysis highlights differences in how mCSPC may progress after upfront androgen receptor signaling inhibitor therapy.
Apalutamide plus ADT markedly reduced PSA-defined progression compared with placebo plus ADT, but was not associated with a statistically significant reduction in radiographic CRPC. Moreover, patients receiving apalutamide who developed radiographic CRPC frequently did so while PSA levels remained low.
The authors noted that PSA monitoring alone may therefore be insufficient to fully capture disease progression during upfront androgen receptor signaling inhibitor treatment. Because radiographic progression may occur despite suppressed PSA levels, regular imaging follow-up may be considered, particularly in patients with high-risk features such as baseline visceral metastases.
The association between baseline visceral metastases and radiographic CRPC suggests that patients with visceral disease may be at higher risk of radiographic progression. However, the authors emphasized that any risk-adapted treatment strategies based on these observations require prospective validation.
Study Limitations
Several limitations should be considered when interpreting the findings. This was an exploratory post hoc analysis, and no adjustment was made for multiple testing. The classification of PSA and radiographic progression as mutually exclusive categories may also have been affected by differences in the timing of PSA measurements and imaging assessments, although the sensitivity analysis using a 28-day window produced results consistent with the primary analysis.
In addition, only patients who had developed CRPC by the TITAN interim analysis were included in the progression-pattern analysis, which may limit the generalizability of the findings to patients with more indolent mCSPC.
Although inverse probability of treatment weighting was used to balance baseline characteristics, residual confounding from unmeasured variables cannot be excluded. The weighting also relied on baseline characteristics rather than clinical features present at the time CRPC developed, which may have influenced the survival comparisons.
Takeaway
In this post hoc analysis of TITAN, apalutamide plus ADT significantly reduced the 24-month cumulative incidence of PSA-defined CRPC compared with placebo plus ADT, but the cumulative incidence of radiographic CRPC remained similar between treatment groups.
Radiographic CRPC was associated with poorer survival than PSA-CRPC in patients receiving placebo plus ADT, whereas this prognostic difference was not statistically significant among patients treated with apalutamide. Baseline visceral metastases were associated with radiographic CRPC in both treatment groups, and radiographic progression during apalutamide treatment could occur while PSA levels remained low.
The findings support the authors’ conclusion that PSA monitoring alone may not capture all disease progression during upfront androgen receptor signaling inhibitor therapy and that regular imaging follow-up may be considered, particularly in patients with high-risk features such as visceral metastases. As an exploratory post hoc analysis, the findings require prospective validation.
