Abstract
Background/Aim: The primary objective of this study was to identify predictors for biochemical recurrence (BCR) within 2 years following robot-assisted radical prostatectomy (RARP). Identifying predictors will enable insights that enhance personalized patient management and facilitate the ongoing refinement of postoperative therapy strategies. Patients and Methods: This retrospective study included patients undergoing RARP from September 2014 to January 2021. Exclusion criteria were preoperative endocrine therapy, BCR beyond 2 years post-surgery, and incomplete postoperative data. Multivariate analyses were conducted to evaluate predictors of BCR, focusing on preoperative prostate-specific antigen (PSA) level, pathological tumor (pT) stage, Gleason score (GS), extraprostatic extension (EPE), and surgical margin status. Results: Among 374 patients, 40 experienced BCR within 2 years. Significant predictors of early BCR included initial PSA level ≥10 ng/ml, pT3 or greater, GS ≥8, EPE, and positive surgical margins (RM1). Multivariate analysis identified pT3 or higher, GS ≥8, and RM1 as independent risk factors for early BCR. Conclusion: Early BCR after RARP is significantly associated with advanced pathological stage, high GS, and positive surgical margins. These findings emphasize the need for tailored postoperative management strategies and highlight the importance of precision in surgical technique to improve patient outcomes.
- Robot-assisted radical prostatectomy
- biochemical recurrence
- prostate-specific antigen
- Gleason score
- predictive factors
Prostate cancer is a prevalent malignancy, and radical prostatectomy (RP) has been adopted as a standard treatment option. Recently, robot-assisted radical prostatectomy (RARP) has become the predominant treatment modality due to its minimally invasive approach (1). Despite successful surgery, a considerable number of patients experience biochemical recurrence (BCR) based on a rising prostate-specific antigen (PSA) level. Recurrence rates, which vary from 10% to 50%, underscore the clinical significance of this phenomenon (2-7). Notably, the majority of BCRs manifest within the initial years post-RP (4), further emphasizing the need for a deeper understanding of early recurrence dynamics. The occurrence of high-risk BCR within 2 years following RP has been associated with an increased likelihood of developing metastasis (8). Additionally, patients with prognostically adverse clinicopathological factors before and after treatment, such as a high Gleason score (GS) and short PSA doubling time, are considered to have an increased risk of developing BCR (9).
While research has been devoted to understanding BCR post-RP, a crucial aspect remains unexplored: specific prognostic factors for BCR following RARP are currently lacking. The emergence of RARP as a minimally invasive alternative to traditional RP warrants a dedicated investigation into the predictors of BCR in this context. While prior studies offer insights into overall BCR rates following RP, the distinct factors influencing early recurrence after RARP remain unclear. This study seeks to bridge the existing knowledge gap regarding BCR following RARP. Recognizing these gaps in knowledge is crucial for improving clinical follow-up strategies and refining postoperative management. The primary aim of this study was to identify the specific factors that may influence the prognosis for early BCR after RARP. The findings from our study provide valuable insights that can inform clinical decision-making, facilitate personalized patient management, and contribute to the ongoing refinement of postoperative care strategies.
Patients and Methods
This retrospective study focused on patients who underwent RARP between September 2014 and January 2021 at our hospital. Data were collected retrospectively, and a comprehensive multivariate analysis was conducted to identify significant predictors of BCR within 2 years of RARP. Exclusion criteria encompassed patients who received preoperative endocrine therapy, experienced BCR beyond the 2-year mark post-surgery, and those with missing postoperative data.
In the preoperative evaluation, we assessed the clinical T stage (10) through a comprehensive assessment that incorporated digital rectal examination and pelvic magnetic resonance imaging. In addition, the preoperative evaluations included chest-to-pelvis computed tomography and whole-body bone scanning. These diagnostic procedures confirmed the absence of metastatic disease, ensuring that the cancer was localized and had not spread to other parts of the body.
Surgical procedures were performed via the transperitoneal approach using the da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA, USA). And anterograde surgery was performed. Pelvic lymph node dissection (PLND) was performed on the high-risk group according to the D’Amico risk classification (11).
In this study, the criteria for defining BCR post-RARP were established as follows: When the PSA level was less than 0.2 ng/ml more than 1 month after surgery, it was considered as no BCR (12). Subsequently, when the PSA level was greater than 0.2 ng/mL on two consecutive occasions, it was classified as BCR, with the date of the first elevated measurement being designated as the date of recurrence. In cases where the PSA level never fell below 0.2 ng/ml post-surgery, the date of the surgery was determined as the date of recurrence (12). Patients underwent postoperative PSA assessments at 1 month, followed by assessments every 3 months for the first 3 years, biannually from the second to the fourth years, and annually from the fifth year onward. Patients without confirmed BCR after surgery did not receive immediate adjuvant hormonal treatment or pelvic radiation therapy.
The statistical analysis employed the Kaplan–Meier method, evaluating predictors of BCR. Subsequently, Cox regression analysis was utilized to scrutinize the impact of various factors on BCR, including preoperative PSA (ng/ml) at diagnosis (less than 10 vs. 10 or greater), pathological tumor stage (pT2 or less vs. pT3 or greater), pathological GS (7 or less vs. 8 or greater), extraprostatic extension status (no vs. yes), surgical margin status (RM0 vs. RM1). To ensure clarity and facilitate interpretation, all data are presented as medians with corresponding interquartile ranges (IQR: first quartile–third quartile). All p-values were calculated using two-sided tests, and values less than 0.05 were considered statistically significant. All statistical analyses were performed with EZR for R (13).
Ethical considerations were upheld, and the study was conducted in compliance with the principles outlined in the Declaration of Helsinki. The study protocol received approval from the Institutional Review Board at Nagasaki University Hospital (Approval ID: 23061910). Written informed consent was obtained from all individual participants included in the study.
Results
Patient characteristics. Table I shows the clinical characteristics of the patients. A cohort of 374 eligible patients was enrolled with a median age of 69 (IQR=65-72) years. The initial median PSA level was 7.89 (IQR=5.74-11.23) ng/ml. Clinical stage at diagnosis was mainly cT1b-c (58.3%). Biopsy GS was ≤7 in 74.1%. Among the 40 patients with BCR, the median age was 68 (IQR=65-75) years, and the median initial PSA was 11 (IQR=7.37-14.51) ng/ml. PSA levels were ≥10 ng/ml in 55% of BCR cases. The clinical stage was generally cT2a-b or higher (57.5%). GS in the BCR group was ≥8 in 55%, suggesting a higher incidence of aggressive disease.
Patients’ background data.
Pathological characteristics. Table II shows the pathological characteristics of the patients. The BCR subgroup exhibited a higher incidence of EPE and RM1 than that in the total cohort. Additionally, the proportions of pT3a and pT3b stages were significantly elevated in the BCR subgroup relative to the total cohort, and the proportion of cases with a GS of 7 or higher was also greater. These findings indicate a higher incidence of advanced disease in patients with recurrence.
Pathological characteristics of the total cohort and patients with biochemical recurrence.
Extended PLND. In the total cohort, 107 (28.6%) underwent extended PLND. Among these patients, lymph node metastasis was identified in 25 (23.4%). Of the 40 patients who experienced BCR, 18 had undergone extended PLND, and of these, 12 (66.7%) were found to have lymph node metastasis.
Factors influencing early BCR. During a median follow-up of 51 months (range=38-67 months), 40 patients (10.7%) developed BCR.
Table III shows the results of univariate and multivariate analyses of the associations between BCR and perioperative factors. Kaplan–Meier log-rank testing highlighted significant associations between early BCR and several prognostic indicators. Specifically, an initial PSA level of 10 ng/ml or higher, pathological staging of pT3 or above, GS of 8 or higher, EPE, and RM1 were statistically significant, each with p-values below 0.01.
Factors predictive of biochemical recurrence.
Multivariate analysis identified three independent risk factors for early BCR. Pathological staging at pT3 or above correlated with a six-fold increased risk of BCR [odds ratio (OR)=6.31, 95% confidence interval (CI)=1.20-33.4, p=0.0299]. Higher pathological GS (≥8) doubled the risk (OR=2.10, 95% CI=1.05-4.19, p=0.0364), and RM1 demonstrated a three-fold risk increase (OR=3.02, 95% CI=1.44-6.32, p=0.0034).
Discussion
In this retrospective study, we explored the outcomes of patients undergoing RARP at our Institution, to identify significant predictors of BCR within 2 years post-surgery. Early BCR is associated with a high risk of developing metastasis (8), making the identification of its predictors critical for enabling early intervention in postoperative management. Our findings indicate that pathological GS, pathological T stage, and RM1 are significant predictors of early BCR. This study holds the potential for significant impact on the approach to follow-up care and treatment strategy adjustments following RARP, thereby contributing valuable insights into the management of patients with prostate cancer.
Additionally, nomograms combining various prognostic variables to more accurately predict BCR have been reported (14-17). The prognostic variables used in these nomograms largely concur with the predictive factors identified in this study. However, these nomograms were validated based on patients who underwent a variety of surgical approaches, including open surgery, laparoscopic surgery, or robotic surgery. While there are predictive factors independent of the surgical approach, such as GS and PSA, there are also factors like RM1 that reflect the techniques used in surgery.
Recent research indicates that RARP is associated with a reduced incidence of RM1 when compared to RP (18, 19). Consequently, our study, which focused exclusively on robotic surgery, is believed to have identified more accurate predictive factors in RARP. Similar to our study, there are studies focused exclusively on RARP, concluding that higher GS at RM1, greater size of RM1, and lymph node invasion were significant predictors of BCR following RARP (20).
The study of Liesenfeld et al. investigated factors contributing to late BCR, and, in addition to pathological T stage and pathological GS, a family history of prostate cancer and age at surgery were also identified as factors (4).
Pathological GS as a prognostic indicator. Our findings underscore the prognostic significance of the pathological GS in predicting early BCR after RARP. Specifically, GS of 8 or higher was associated with a doubled risk of recurrence. Considering that GS is indicative of malignancy severity, it follows that higher GS in resected specimens correlates with a heightened risk of BCR (20). This correlation is supported by findings from previous research (1, 2) and underscores the increased aggressiveness of cancers presenting with higher GS. Reports suggest that the inclusion of pathological parameters, such as the percentage of Gleason patterns 4 and/or 5, or the presence of intraductal carcinoma of the prostate, as other variables in nomograms for predicting BCR enhances their predictive accuracy. These findings underscore the significance of the GS (21).
In our study, we also demonstrate the significance of the presence of high Gleason patterns in predicting BCR. Our findings indicate that patients with GS of 8 or higher should be closely monitored after surgery, and consideration should be given to tailored treatment strategies. Incorporating GS into risk-stratification models can improve predictive accuracy and guide individualized therapeutic decisions (3).
Pathological stage pT3 or higher as a critical predictor of BCR. Pathological staging, particularly of pT3 or higher, emerged as a robust predictor of early BCR in our study. Patients with pT3 or more staging exhibited a six-fold increased risk of recurrence. This finding reinforces the importance of accurately assessing the extent of cancer invasion during pathological evaluation. Higher pathological stages are indicative of more aggressive disease and a greater likelihood of recurrence (4). Although pT has been reported as a prognostic factor for BCR, it is specifically associated with BCR occurring after 10 years post-surgery rather than early BCR (4). However, this report (4) suggests that pT might also be indicative of early BCR. Therefore, patients with pT3 or higher staging should undergo intensive postoperative monitoring and, when appropriate, more aggressive adjuvant therapies should be considered (8). Our study emphasizes the critical role of pathological staging in refining postoperative strategies and underscores its importance in clinical practice.
RM1 and recurrence risk. RM1 was identified as an independent risk factor for early BCR in our analysis. Patients with RM1 had a three-fold increased risk of recurrence. The presence of RM1 suggests residual neoplastic tissue postoperatively, emphasizing the need for meticulous surgical technique and comprehensive neoplastic excision during RARP. Our findings highlight the imperative for achieving negative surgical margins whenever possible to minimize recurrence risk (20, 22).
Previous reports have indicated that higher GS at RM1 and greater RM1 are significant predictors of early BCR following RARP. It was found that lymph node invasion, GS of ≥7 at RM1, and a maximum RM1 exceeding 6 mm significantly negatively influenced BCR-free survival, underscoring not only the presence but also the extent of disease and indicating GS at RM1 as critical factors for early BCR prediction (20).
While RARP initially presents a long learning curve, performance, including reduced RM1 rates and improved urinary function, increases significantly over time (23). The outcomes, particularly RM1 rates, are heavily influenced by the surgeon’s experience, skill and specific surgical techniques (24). Therefore, experienced surgeons should perform surgeries on patients where RM1 is a concern.
Limitations and considerations. Lymph node metastasis significantly impacts prognosis, conferring an increased risk of BCR. This emphasizes the importance of integrating lymph node status into risk-assessment models to enhance predictive accuracy. However, in our study, given that extended PLND was only conducted for patients classified at high-risk according to the D’Amico risk classification, it was not considered as a predictive factor in the evaluation of predictors of BCR. Additionally, the retrospective design of our study introduced potential selection bias, highlighting the need for prospective validation of our findings. Moreover, variations in surgical techniques and postoperative care across different centers might limit the generalizability of our results. Acknowledging these limitations is crucial.
In conclusion, our study emphasizes the critical role of pathological GS, pathological T stage, and RM1 in predicting early BCR following RARP. These findings pave the way for tailored treatment strategies and underscore the importance of precision in surgical practice. Together, these elements hold the potential to improve the prognosis and quality of life for patients with prostate cancer.
Acknowledgements
The Authors would like to thank Editage (www.editage.jp) for English language editing.
Footnotes
Authors’ Contributions
KM designed this study and contributed to writing the article. KM, YN, JH, HK and HF collected the data, analyzed the data, and assisted in article preparation. RI, KO, TM and YM reviewed the article.
Conflict of Interest
The Authors declare no conflicts of interest.
Funding
This study was conducted without any financial support.
- Received April 23, 2024.
- Revision received June 4, 2024.
- Accepted June 6, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
