Abstract
Background/Aim: We examined the prognostic use of the 3-month prostate-specific antigen (PSA3m) level after androgen-deprivation therapy in patients with de novo metastatic hormone-sensitive prostate cancer (mHSPC). Patients and Methods: The present study included 145 patients with mHSPC who received primary androgen-deprivation therapy. Results: The optimal cutoff PSA3m value for prediction of 5-year overall survival was 2.56 ng/ml (area under the receiver operating characteristics curve=0.67) using a time-dependent receiver operating characteristic (survival ROC) curve. In patients with CHAARTED low-volume and LATITUDE low-risk disease, the median overall survival was longer for patients with low PSA3m than that for those with high PSA3m. Multivariate analysis revealed PSA3m (hazard ratio=1.99; p=0.006) and age ≥80 years as independent risk factors for overall survival in patients with mHSPC. Conclusion: PSA3m can be a useful prognostic biomarker to avoid excessive upfront combination therapy, particularly in elderly patients with low-volume and low-risk mHSPC.
Prostate cancer (PC) is the second most common malignancy among males worldwide (1). Approximately 10-20% of patients in Japan are diagnosed with metastatic hormone-sensitive PC (mHSPC) (2). Until recently, androgen-deprivation therapy (ADT) has been the standard treatment for men with mHSPC. Although PC is highly androgen-dependent and sensitive to primary ADT, patients with mHSPC eventually develop metastatic castration-resistant PC (CRPC) that is associated with a high rate of mortality (1).
Recent developments in novel therapies have dramatically changed the treatment strategies for patients with mHSPC. Some therapeutic options for CRPC have been combined with ADT in the setting of mHSPC, leading to upfront combination therapy. Large-scale phase 3 trials have demonstrated the clinical benefits of upfront combination therapy, such as ADT plus docetaxel (3, 4), abiraterone acetate (5), enzalutamide (6, 7), or apalutamide (8). The European Association of Urology (9) and American Urological Association/American Society for radiation Oncology/Society of Urologic Oncology (10) guidelines recommend the use of upfront combination therapy as first-line therapy for all patients with mHSPC, reflecting the current standard of care practices.
However, the clinical benefits for patients with low-volume (low-volume; CHAARTED criteria) (4, 11)/low-risk (low-risk; LATITUDE criteria) or aged ≥75 years (5, 12) remain unclear. Moreover, upfront agents can cause agent-specific adverse events (AEs). Notably, severe (grade ≥3) AEs were observed in 30-60% of patients receiving upfront combination therapies (6-8, 12, 13). Compared with younger patients, patients aged ≥75 years showed higher rates of cardiac disorders, fluid retention, and fatigue in response to abiraterone plus prednisone, and higher rates of grade ≥3 AEs, falls, fractures, appetite loss, and asthenia in response to enzalutamide (14, 15). On the other hand, some patients with mHSPC achieved long-term survival with ADT alone. In the CHAARTED trial, the median overall survival (OS) of patients with low-volume de novo mHSPC receiving ADT alone was approximately 5 years (16). The latest European Association of Urology guidelines strongly recommend not offering ADT monotherapy to patients with de novo mHSPC if they have a sufficient life expectancy to benefit from upfront combination therapy and are willing to accept the increased risk of side-effects (9). Thus, an ‘all-comers’ upfront therapeutic approach may be excessive, particularly for elderly or frail patients with long-term survival, with ADT alone due to the trade-offs between cancer morbidity and treatment morbidity. Therefore, it is necessary to predict the precise oncological outcomes to avoid overtreatment.
Prognostic biomarkers are required to identify which patients might achieve long-term survival using primary ADT since the population of patients with mHSPC is heterogeneous. The American Urological Association/American Society for radiation Oncology/Society of Urologic Oncology guidelines moderately recommend the prognostic value of the CHAARTED criteria to assess the extent of metastatic disease (10). Several studies have also shown that the CHAARTED/LATITUDE criteria can predict survival in patients with mHSPC (4, 17-19). Identifying prognostic biomarkers that can be measured early in the treatment process, which is known as response-guided treatment strategy, might help to avoid treatment delays in men who require upfront combination therapy with ADT. Prostate-specific antigen (PSA) kinetics, such as nadir PSA level and time to nadir (TTN), during ADT are recognized as strong prognostic factors in patients with mHSPC (20, 21). A 3-month delay may be acceptable to start upfront combination therapy, as demonstrated in the CHAARTED (4), LATITUDE (5), and STAMPEDE (3) trials; therefore, the present study examined the association between survival and 3-month PSA (PSA3m) levels after the initiation of primary ADT combined with the CHAARTED/LATITUDE criteria in patients with mHSPC.
Patients and Methods
Patients. This retrospective study was conducted in accordance with the ethical standards of the Declaration of Helsinki and was approved by the Institutional Review Board of the University of Occupational and Environmental Health, Japan (UOEHCRB21-074). We applied an opt-out methodology to provide accessible information to all patients to facilitate informed consent without interfering with the medical consultation and the patients were informed of their inclusion in the study and were provided information on the institution’s website. Consecutive patients with newly diagnosed mHSPC who received combined androgen blockade (CAB) as primary ADT at the University of Occupational and Environmental Health Hospital in Japan between January 2000 and December 2019 were considered for analysis. Patients for whom CAB was initiated at other hospitals and who were then referred to our hospital were also included. All patients received CAB comprising castration (either via surgery, luteinizing hormone-releasing hormone agonist, or antagonist) plus a non-steroidal antiandrogen (80 mg/day bicalutamide or 375 mg/day flutamide). The serum PSA levels were measured every month for several months after the initiation of CAB. PSA levels were measured every 3 months in patients with low and stable PSA levels. Imaging studies were conducted at the discretion of each physician. Patients who received docetaxel or next-generation anti-androgens as upfront combination therapy prior to CRPC were excluded. CRPC was defined according to the recommendations of the Prostate Cancer Clinical Trials Working Group 2 (22).
Patient background assessments included age, Eastern Cooperative Oncology Group performance status (ECOG-PS), initial PSA levels, PSA3m, International Society of Urological Pathology grade group, primary tumor stage, regional lymph node stage, visceral metastases, extent of disease, CHAARTED volume criteria (4), and LATITUDE risk criteria (5). Metastatic status was determined using chest and body computed tomography and bone scintigraphy prior to initiating ADT. OS, prostate cancer-specific survival (PCSS), and CRPC-free survival (CRPC-FS) were defined as the interval from initiation of ADT to death from any cause, PC-related death, and progression to CRPC, respectively. Patients were censored when the event was not observed until the end of follow-up.
Treatment protocol. All patients with mHSPC were initially treated with CAB. After CRPC diagnosis, patients received treatments comprising systemic chemotherapy (docetaxel and/or cabazitaxel) and/or next-generation anti-androgens (abiraterone and/or enzalutamide). Patients who chose not to receive chemotherapy or next-generation anti-androgens were treated with alternative antiandrogen therapy, anti-androgen withdrawal therapy, estramustine, or low-dose oral steroid therapy. Cabazitaxel was approved as second-line treatment after docetaxel. No companion diagnostics were used to determine treatment options.
Statistical analysis. Data were analyzed statistically using EZR (Easy R, Vienna, Austria), which is a graphical user interface for R (The R Foundation for Statistical Computing). All p-values less than 0.05 were considered statistically significant. Survival rates were calculated using the Kaplan–Meier method and compared using the log-rank test. The optimal cut-off value for PSA3m was defined as the point closest to the upper left corner of the receiver operating characteristics curve for survival. Multivariate Cox regression analysis was performed to evaluate the impact of PSA3m on OS. Variables with p-values less than 0.2 were selected in the univariate models to avoid overfitting the regression model.
Results
Patients. A total of 158 patients with de novo mHSPC treated with CAB were identified during the study period. Among these patients, 13 were excluded due to poor Eastern Cooperative Oncology Group performance status (ECOG-PS) values (≥3) or a lack of detailed clinical and follow-up data. The remaining 145 patients were eligible for analysis and the baseline characteristics are shown in Table I. The time-dependent receiver operating characteristic (survival ROC) curve analysis for 5-year OS revealed that the optimal cutoff value of PSA3m was 2.56 ng/ml and the area under the receiver operating characteristics curve was 0.67 (Figure 1).
Patient characteristics (n=145).
Survival receiver operating characteristics curve analysis of 3-month prostate-specific antigen levels for predicting 5-year overall survival. The area under the curve (AUC), sensitivity, and specificity are shown. The red circle indicates the optimal point.
Kaplan–Meier analyses. The median follow-up time from CAB initiation was 43 (interquartile range=22-72) months. The median OS was significantly longer in patients with low-volume and low-risk disease than in those with high-volume or high-risk disease, or both [92.0 vs. 45.0 months; hazard ratio (HR)=0.56, 95% confidence interval (CI)=0.34-0.91; p=0.021] (Figure 2A). The median CRPC-FS was significantly longer in men with low-volume and low-risk disease compared with men with high-volume or high-risk disease, or both (48.0 vs. 11.0 months; HR=0.31, 95% CI=0.19-0.50; p<0.001) (Figure 2B).
Kaplan–Meier curves for overall survival (A) and castration-resistant prostate cancer-free survival (CRPC-FS) (B) in patients with metastatic hormone-sensitive prostate cancer who were initially treated with combined androgen blockade. Overall survival and CRPC-FS were stratified by CHAARTED and LATITUDE criteria. p-Values were calculated using log-rank test.
Heterogeneity in survival was investigated using a cut-off value for PSA3m of 2.56 ng/ml. The median OS was significantly longer for patients with low PSA3m compared with those with high PSA3m (89.0 vs. 43.0 months; HR=0.39, 95% CI=0.25-0.61; p<0.001) (Figure 3A). The median CRPC-FS was significantly longer for men with low PSA3m than those with high PSA3m (29.0 vs. 10.0 months; HR=0.29, 95% CI=0.20-0.44; p<0.001) (Figure 3B).
Kaplan–Meier curves for overall survival (A) and castration-resistant prostate cancer-free survival (CRPC-FS) (B) in patients with metastatic hormone-sensitive prostate cancer who were initially treated with combined androgen blockade. Overall survival and CRPC-FS were stratified by 3-month prostate-specific antigen (PSA3m). p-Values were calculated using log-rank test.
We compared OS according to the CHAARTED volume/LATITUDE risk criteria and PSA3m levels (Figure 4A). In patients with low-volume and low-risk disease, the median OS was longer in patients with low PSA3m compared with those with high PSA3m (92 vs. 60 months; HR=0.38, 95% CI=0.15-0.96; p=0.040). The 5-year OS rates for patients low-volume and low-risk disease with low and high PSA3m were 75.9% and 38.1%, respectively. PCSS and CRPC-FS also showed similar patterns to those for OS (Figure 4B and C). In the low-volume and low-risk group, patients with low PSA3m had significantly longer PCSS (median: 112 vs. 60 months; HR=0.32, 95% CI=0.12-0.89; p=0.029) and CRPC-FS (median: 66 vs. 18 months; HR=0.29, 95% CI=0.12-0.74; p=0.009) than those with high PSA3m. Of note, 51% of patients with low-volume and low-risk disease whose PSA3m levels <2.56 ng/ml reached 5-year CRPC-FS.
Kaplan–Meier curves for overall survival (A), prostate cancer-specific survival (PCSS) (B), and castration-resistant prostate cancer-free survival (CRPC-FS) (C) in patients with metastatic hormone-sensitive prostate cancer who were initially treated with combined androgen blockade stratified by 3-month prostate-specific antigen (PSA3m) and low/high (L/H) volume/risk (V/R). p-Values were calculated using log-rank test.
Analyses of prognostic factors in mHSPC patients. Table II shows the results of the univariate and multivariate survival analyses using Cox regression model for OS. Univariate analysis revealed that PSA3m <2.56 ng/ml, low CHAARTED volume/LATITUDE risk criteria, age <80 years, and ECOG-PS <2 were significantly associated with long-term survival. Among the pretreatment variables (excluding PSA3m), multivariate analysis revealed that age ≥80 years (HR=2.33, 95% CI=1.39-3.92; p=0.001), and high-volume or high-risk disease, or both (HR=1.69, 95% CI=1.01-2.82; p=0.046) were significant prognostic factors. When PSA3m was included in the variables, PSA3m was significantly associated with OS (HR=1.99, 95% CI=1.22-3.23; p=0.006), whereas volume/risk was no longer associated with OS (p=0.193).
Univariate and multivariate analyses of overall survival.
Discussion
The findings of the present study demonstrate that PSA3m can independently predict long-term survival in patients with de novo mHSPC better than the CHAARTED/LATITUDE criteria. Firstly, we analyzed the optimal cut-off value of PSA3m which was significantly associated with OS after initiation of CAB. Next, we showed that patients in the low-volume and low-risk subgroup with low PSA3m had a favorable outcome. Finally, Cox hazard model analysis revealed that PSA3m was an independent prognostic factor for OS and had a higher HR than the other variables. Notably, the prognostic impact of PSA3m on OS was superior to that of the CHAARTED volume and LATITUDE risk criteria, which have been reported as prognostic factors for mHSPC (4, 17-19). Furthermore, age ≥80 years was independently significantly associated with poor OS. Therefore, PSA3m may be a useful prognostic biomarker, particularly for elderly patients with low-volume and low-risk mHSPC to avoid excessive upfront combination therapy in practice.
To our knowledge, this is the first study to show that PSA3m levels during CAB are a better indicator of prognosis in patients with mHSPC than the CHAARTED volume/LATITUDE risk criteria.
We assessed the utility of PSA3m for the following reasons. Firstly, PSA kinetics, such as nadir PSA and TTN, during primary ADT/CAB are recognized as prognostic factors in patients with mHSPC (21, 22). However, it takes approximately 6 to 12 months to detect the nadir PSA and TTN (23, 24), which means the disease state at nadir PSA/TTN is near CRPC. Decision-making is required prior to nadir PSA/TTN to determine administration of upfront combination therapy for mHSPC. Secondly, as demonstrated in the CHAARTED (4), LATITUDE (5), and STAMPEDE (3) trials, commencement of upfront combination therapy can be acceptable up to 3 months after ADT initiation. Therefore, PSA3m is a potential prognostic biomarker for mHSPC. We successfully demonstrated that PSA3m was an independent factor for OS. Since mHSPC with low PSA3m may have a higher rate of hormone-sensitive clones, PSA3m may directly reflect outcomes. Finally, several studies have reported the associations between survival outcomes and early changes in serum PSA level during primary ADT (17, 25, 26). One study showed that a log-transformed PSA change at 12 weeks was associated with OS in patients with mHSPC (25), whereas the study found no statistical association between the CHAARTED criteria and OS in univariate analysis using a log-rank test, presumably due to the small number of patients (n=60). It is important to note that the other studies included patients treated with ADT alone or CAB, which may have affected PSA kinetics (17, 26). For example, a large, multicenter study showed that the median nadir PSA was significantly lower in patients treated with CAB than that in those treated without CAB (0.3 vs. 1.0 ng/ml; p<0.001) and there was no statistical difference in median TTN between the CAB and non-CAB groups (7.0 vs. 7.3 months; p=0.967) (27). A phase 3 clinical trial showed that patients who received CAB with bicalutamide achieved lower nadir PSA levels than those who received ADT alone (28). Since PSA3m is most likely to be between the initial and nadir PSA levels, the PSA3m level after initiation of ADT alone may have a higher cutoff value to predict survival than that of nadir PSA with CAB, or no predictive power for survival.
Several prognostic factors for survival have been suggested, including the number and location of bone metastases, presence of visceral metastases, Gleason score, PS, alkaline phosphatase and lactate dehydrogenase (4, 29). The CHAARTED/LATITUDE criteria consist of these metastatic factors and Gleason score and have been reported as prognostic risk factors for mHSPC in a clinical trial as well as real-world settings (4, 17-19). The present study demonstrated that PSA3m was an independent prognostic factor and had a higher HR than the other variables, including the CHAARTED volume and LATITUDE risk criteria in multivariate analysis. Therefore, PSA3m may be a useful factor to establish more precise prognostic models or nomograms for mHSPC.
The ‘all-comers’ approach using upfront combination therapy is in principle useful for patients with mHSPC. It is likely that patients with mHSPC, particularly those with high-volume/high-risk disease, or both who are fit to receive an upfront agent in addition to ADT may benefit from such a regimen. However, the balance between the risks and benefits should be considered for each patient for a personalized approach. A PSA response-guided treatment strategy using PSA3m can be useful, particularly for elderly/frail patients with low CHAARTED volume/LATITUDE risk mHSPC in the rapidly aging population.
The present study has several limitations. Firstly, the retrospective design and limited sample size may have led to a selection bias and influenced confounding factors. Secondly, our database lacked some prognostic biomarkers for PC. Finally, we only showed the value of PSA3m as a prognostic factor in Japanese patients with mHSPC treated with CAB, not with ADT alone. Nevertheless, we do not consider these factors to have had a strong impact on our conclusions.
In conclusion, the PSA3m level after CAB may be a more accurate prognostic factor than the CHAARTED volume/LATITUDE risk criteria in patients with de novo mHSPC. Further large-scale, prospective studies are warranted.
Footnotes
Conflicts of Interest
The Authors declare that there are no financial disclosures or conflicts of interest regarding this article.
Authors’ Contributions
YN: Project development, article writing and editing, data analysis. TM: Data collection and data analysis. IT: Critical review and supervision. NF: Project development, critical review, and supervision. All Authors discussed, verified and approved the final version of the article.
- Received December 4, 2021.
- Revision received December 17, 2021.
- Accepted December 21, 2021.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
