Predictor Of Biochemical Recurrence Research Articles

Preoperative Predictors of Lymph Node Invasion and Biochemical Recurrence in High-risk Prostate Cancer.

To evaluate the preoperative predictors of pathological lymph node (LN) metastasis and prognostic factors for postoperative biochemical recurrence (BCR) in robot-assisted radical prostatectomy with extended pelvic LN dissection in patients with D'Amico high-risk prostate cancer (PCa). Overall, 107 patients with D'Amico high-risk PCa underwent robot-assisted radical prostatectomy with extended pelvic LN dissection without neoadjuvant or adjuvant therapy. BCR was defined as a prostate-specific antigen (PSA) level ≥0.2 ng/ml. Moreover, BCR-free survival rates were determined using Kaplan-Meier analysis. Logistic regression analysis was used to evaluate preoperative predictors of pathological LN metastasis. Cox regression analysis was used to evaluate the effects of preoperative and pathologic variables on BCR. The median follow-up was 21 months, and the 5-year BCR-free survival rate was 59.8%. The positive LN rate was 21.5%. In multivariate analysis, the percentage of positive cores was a significant preoperative predictor of positive LNs. Patients with >50% positive cores (p=0.004) and PSA density (PSAD) >0.5 ng/ml/cc (p=0.005) had a high risk of having ≥3 positive LNs. In multivariate analysis, PSAD >0.5% was a significant preoperative predictor of BCR. Among the postoperative predictors, the number of positive LNs was significantly associated with BCR. Patients with ≥3 positive LNs (n=7) had significantly lower BCR-free survival rates than patients with one or two positive LNs (n=16) (p<0.001). Patients with >50% positive cores and PSAD >0.5 ng/ml/cc had a risk for a high number of positive LNs (≥3) that was strongly associated with shorter BCR-free survival (p<0.001). The percentage of positive cores may be useful as a preoperative predictor of pathological LN metastasis in patients with high-risk PCa. Patients with >50% positive cores and PSAD >0.5 ng/ml/cc were found to have a high risk for ≥3 positive LNs and shorter BCR-free survival.

State of the art of radiomic analysis in the clinical management of prostate cancer: A systematic review.

We present the current clinical applications of radiomics in the context of prostate cancer (PCa) management. Several online databases for original articles using a combination of the following keywords: "(radiomic or radiomics) AND (prostate cancer or prostate tumour or prostate tumor or prostate neoplasia)" have been searched. The selected papers have been pooled as focus on (i) PCa detection, (ii) assessing the clinical significance of PCa, (iii) biochemical recurrence prediction, (iv) radiation-therapy outcome prediction and treatment efficacy monitoring, (v) metastases detection, (vi) metastases prediction, (vii) prediction of extra-prostatic extension. Seventy-six studies were included for qualitative analyses. Classifiers powered with radiomic features were able to discriminate between healthy tissue and PCa and between low- and high-risk PCa. However, before radiomics can be proposed for clinical use its methods have to be standardized, and these first encouraging results need to be robustly replicated in large and independent cohorts.

Prediction of Prostate Cancer Biochemical Recurrence by Using Discretization Supports the Critical Contribution of Extra-Cellular Matrix Genes

Prediction of Prostate Cancer Biochemical Recurrence by Using Discretization Supports the Critical Contribution of Extra-Cellular Matrix Genes

Independent validation of a pre-specified four-kallikrein marker model for prediction of adverse pathology and biochemical recurrence.

Accurate markers for prostate cancer (PC) risk stratification could aid decision-making for initial management strategies. The 4Kscore has an undefined role in predicting outcomes after radical prostatectomy (RP). We included 1476 patients with 4Kscore measured prior to RP at two institutions. The 4Kscore was assessed for prediction of adverse pathology at RP and biochemical recurrence (BCR) relative to a clinical model. We pre-specified that all analyses would be assessed in biopsy Grade Group 1 (GG1) or 2 (GG2) PC patients, separately. The 4Kscore increased discrimination for adverse pathology in all patients (delta area under the receiver operative curve (AUC) 0.009, 95% confidence interval (CI) 0.002, 0.016; clinical model AUC 0.767), driven by GG1 (delta AUC 0.040, 95% CI 0.006, 0.073) rather than GG2 patients (delta AUC 0.005, 95% CI -0.012, 0.021). Adding 4Kscore improved prediction of BCR in all patients (delta C-index 0.014, 95% CI 0.007, 0.021; preop-BCR nomogram C-index 0.738), again with larger changes in GG1 than in GG2. This study validates prior investigations on the use of 4Kscore in men with biopsy-confirmed PC. Men with GG1 PC and a high 4Kscore may benefit from additional testing to guide treatment selection. Further research is warranted regarding the value of the 4Kscore in men with biopsy GG2 PC.

Incorporating artificial intelligence in urology: Supervised machine learning algorithms demonstrate comparative advantage over nomograms in predicting biochemical recurrence after prostatectomy

After radical prostatectomy (RP), one-third of patients will experience biochemical recurrence (BCR), which is associated with subsequent metastasis and cancer-specific mortality. We employed machine learning (ML) algorithms to predict BCR after RP, and compare them with traditional regression models and nomograms. Utilizing a prospective Uro-oncology registry, 18 clinicopathological parameters of 1130 consecutive patients who underwent RP (2009-2018) were recorded, yielding over 20,000 data points for analysis. The data set was split into a 70:30 ratio for training and validation. Three ML models: Naïve Bayes (NB), random forest (RF), and support vector machine (SVM) were studied, and compared with traditional regression models and nomograms (Kattan, CAPSURE, John Hopkins [JHH]) to predict BCR at 1, 3, and 5 years. Over a median follow-up of 70.0 months, 176 (15.6%) developed BCR, at a median time of 16.0 months (interquartile range [IQR]: 11.0-26.0). Multivariate analyses demonstrated strongest association of BCR with prostate-specific antigen (PSA) (p: 0.015), positive surgical margins (p < 0.001), extraprostatic extension (p: 0.002), seminal vesicle invasion (p: 0.004), and grade group (p < 0.001). The 3 ML models demonstrated good prediction of BCR at 1, 3, and 5 years, with the area under curves (AUC) of NB at 0.894, 0.876, and 0.894, RF at 0.846, 0.875, and 0.888, and SVM at 0.835, 0.850, and 0.855, respectively. All models demonstrated (1) robust accuracy (>0.82), (2) good calibration with minimal overfitting, (3) longitudinal consistency across the three time points, and (4) inter-model validity. The ML models were comparable to traditional regression analyses (AUC: 0.797, 0.848, and 0.862) and outperformed the three nomograms: Kattan (AUC: 0.815, 0.798, and 0.799), JHH (AUC: 0.820, 0.757, and 0.750) and CAPSURE nomograms (AUC: 0.706, 0.720, and 0.749) (p < 0.001). Supervised ML algorithms can deliver accurate performances and outperform nomograms in predicting BCR after RP. This may facilitate tailored care provisions by identifying high-risk patients who will benefit from multimodal therapy.

Is tumour volume an independent predictor of outcome after radical prostatectomy for high-risk prostate cancer?

BackgroundPreoperative PSA, ISUP grade group (GG), prostate examination and multiparametric MRI (mpMRI) form the basis of prostate cancer staging. Unlike other solid organ tumours, tumour volume (TV) is not routinely used aside from crude estimates such as maximum cancer core length. The aim of this study is to assess the role of TV as a marker for oncological outcomes in high-risk non-metastatic prostate cancer.MethodsA prospectively maintained database of patients undergoing minimally invasive (laparoscopic or robot-assisted laparoscopic) radical prostatectomy at a UK centre between 2007 and 2019 were analysed. A total of 251 patients with NCCN high or very high-risk prostate cancer were identified. Primary outcome measure was time to biochemical recurrence (BCR) and the secondary outcome was time to treatment failure (TTF). TV was measured on the pathological specimen using the stacking method. Multivariable cox regression analysis was used to identify factors predicting BCR and TFF. TV as a predictor of BCR and TFF was further analysed through time-dependent receiver operating characteristic (ROC) curves. Kaplan–Meier survival estimates were used to evaluate TV cut-off scores.ResultsMedian follow up was 4.50 years. Four factors were associated with BCR and TFF on multivariable analysis (TV, pathological GG, pathological T stage, positive margin >3 mm). Area under the Curve (AUC) for TV as a predictor of BCR and TTF at 5 years was 0.71 and 0.75, respectively. Including all 4 variables in the model increased AUC to 0.84 and 0.85 for BCR and TFF. A 2.50 cm TV cut off demonstrated a significance difference in time to BCR, p < 0.001.ConclusionsPathological tumour volume is an independent predictor of oncological outcomes in high risk prostate cancer but does not add significant prognostic value when combined with established variables. However, the option of accurate TV measurement on mpMRI raises the possibility of using TV as useful marker for preoperative risk stratification.

Influence of the time interval between biopsy and surgery on the biochemical recurrence after robot-assisted radical prostatectomy in Japanese patients

ObjectiveWe evaluated the impact of the duration between the biopsy and surgery on the biochemical recurrence (BCR) after robot-assisted radical prostatectomy (RARP). MethodsWe retrospectively evaluated 302 patients who underwent RARP in our institution from April 2010 to December 2017. Patients were categorized into 2 groups, an interval between biopsy and surgery of 180 days or less (Group A) and longer than 180 days (Group B). Factors retrospectively analyzed for the BCR for the interval between the biopsy and RARP included patient's characteristics, intraoperative and postoperative results. The Kaplan–Meier method and Cox proportional hazards regression model were used to evaluate the predictors of BCR. ResultsThe median follow-up was 42 months, with 24 patients developing BCR at a mean of 13.5 months after RARP. There was no difference in the rate of BCR in Group A and Group B. Multivariate analysis showed that BMI (<23.5 kg/m2, p = 0.034), worst GS of the biopsy (≥8, p = 0.007), and without lymph node dissection (p = 0.034) were significant predictors of BCR. Analysis of the interval from the biopsy showed that there was no significant difference between Group A and Group B, when tested according to the NCCN risk stratification (low risk: p = 0.871, intermediate risk: p = 0.205, high risk: p = 0.287). ConclusionThe preoperative predictors of BCR included BMI (<23.5 kg/m2) and worst GS of the biopsy, and without lymph node dissection. A long duration from biopsy to RARP did not influence the probability of BCR, even in patients considered to be at a high risk.

CD8-positive T cells and CD204-positive M2-like macrophages predict postoperative prognosis of very high-risk prostate cancer

To stratify the heterogeneity of prostate cancer (PCa) with seminal vesicle invasion (SVI) immunologically after radical prostatectomy focusing on the tumor microenvironment. We retrospectively reviewed the clinicopathological data of 71 PCa patients with SVI, which is known as a factor of very high-risk PCa. Preoperative clinical variables and postoperative pathological variables were evaluated as predictors of biochemical recurrence (BCR) with a multivariate logistic regression. Immune cell infiltration including the CD8-positive cell (CD8+ cell) and CD204-positive M2-like macrophage (CD204+ cell) was investigated by immunohistochemistry. The cumulative incidence and risk of BCR were assessed with a Kaplan–Meier analysis and competing risks regression. A higher CD8+ cell count in the SVI area significantly indicated a favorable prognosis for cancers with SVI (p = 0.004). A lower CD204+ cell count in the SVI area also significantly indicated a favorable prognosis for cancers with SVI (p = 0.004). Furthermore, the combination of the CD8+ and CD204+ cell infiltration ratio of the SVI area to the main tumor area was a significant factor for BCR in the patients with the PCa with SVI (p = 0.001). In PCa patients with SVI, the combination of CD8+ and CD204+ cell infiltration is useful to predict the prognosis.

A novel serum biomarker quintet reveals added prognostic value when combined with standard clinical parameters in prostate cancer patients by predicting biochemical recurrence and adverse pathology.

The objective was to determine the prognostic utility of a new biomarker combination in prostate cancer (PCa) patients undergoing Radical Prostatectomy (RP). Serum samples and clinical data of 557 men who underwent RP for PCa with pathological stage (pT) <3 at Martini Clinic (Hamburg, Germany) were used for analysis. Clinical Grade Group and clinical stage was determined using biopsy samples while tumor marker concentrations were measured in serum using immunoassays. The prognostic utility of the proposed marker combination was assessed using Cox proportional hazard regression and Kaplan-Meier analysis. The performance was compared to the Cancer of the Prostate Risk Assessment (CAPRA) score in the overall cohort and in a low-risk patient subset. A multivariable model comprising fibronectin 1, galectin-3-binding protein, lumican, matrix metalloprotease 9, thrombospondin-1 and PSA together with clinical Grade Group (GG) and clinical stage (cT) was created. The proposed model was a significant predictor of biochemical recurrence (BCR) (HR 1.29 per 5 units score, 95%CI 1.20–1.38, p<0.001). The Kaplan-Meier analysis showed that the proposed model had a better prediction for low-risk disease after RP compared to CAPRA (respectively 5.0% vs. 9.1% chance of BCR). In a pre-defined low risk population subset, the risk of BCR using the proposed model was below 5.2% and thus lower when compared to CAPRA = 0–2 (9%), GG<2 (7%) and NCCN = low-risk (6%) subsets. Additionally, the proposed model could significantly (p<0.001) discriminate patients with adverse pathology (AP) events at RP from those without. In conclusion, the proposed model is superior to CAPRA for the prediction of BCR after RP in the overall cohort as well as a in a pre-defined low risk patient population subset. It is also significantly associated with AP at RP.

Prostatic Ductal Adenocarcinoma Controlled for Tumor Grade, Stage, and Margin Status Does Not Independently Influence the Likelihood of Biochemical Recurrence in Localized Prostate Cancer After Radical Prostatectomy.

Prostatic ductal adenocarcinoma (PDA) has historically been considered to be an aggressive subtype of prostate cancer. To investigate if PDA is independently associated with worse biochemical recurrence (BCR)-free survival after radical prostatectomy. A review of 1584 radical prostatectomies was performed to grade, stage, and assess margin status in each tumor nodule. Radical prostatectomies with localized PDA (ie, those lacking metastasis) in the tumor nodule with the highest grade and stage and worst margin status were matched with prostatic acinar adenocarcinoma according to grade, stage, and margin status. The effect of PDA on BCR was assessed by multivariable Cox regression and Kaplan-Meier analyses. Prostatic ductal adenocarcinoma was present in 171 cases. We excluded 24 cases because of lymph node metastasis (n = 13), PDA not in the highest-grade tumor nodule (n = 9), and positive surgical margin in a lower-grade tumor nodule (n = 2). The remaining 147 cases included 26 Grade Group (GG) 2, 44 GG3, 6 GG4, and 71 GG5 cancers. Seventy-six cases had extraprostatic extension, 33 had seminal vesicle invasion, and 65 had positive margins. Follow-up was available for 113 PDA and 109 prostatic acinar adenocarcinoma cases. Prostate-specific antigen density (odds ratio, 3.7; P = .001), cancer grade (odds ratio, 3.3-4.3; P = .02), positive surgical margin (odds ratio, 1.7; P = .02), and tumor volume (odds ratio, 1.3; P = .02) were associated with BCR in multivariable analysis. Prostatic ductal adenocarcinoma, its percentage, intraductal carcinoma, and cribriform Gleason pattern 4 were not significant independent predictors of BCR. Advanced locoregional stage, higher tumor grade, and positive surgical margin status rather than the mere presence of PDA are more predictive of worse BCR-free survival outcomes following radical prostatectomy in men with a component of PDA.

Identifying a Ferroptosis-Related Gene Signature for Predicting Biochemical Recurrence of Prostate Cancer.

Ferroptosis induced by lipid peroxidation is closely related to cancer biology. Prostate cancer (PCa) is not only a malignant tumor but also a lipid metabolic disease. Previous studies have identified ferroptosis as an important pathophysiological pathway in PCa development and treatment, but its role in the prognosis of PCa is less well known. In this study, we constructed a nine-ferroptosis-related gene risk model that demonstrated strong prognostic and therapeutic predictive power. The higher risk score calculated by the model was significantly associated with a higher ferroptosis potential index, higher Ki67 expression, higher immune infiltration, higher probability of biochemical recurrence, worse clinicopathological characteristics, and worse response to chemotherapy and antiandrogen therapy in PCa. The mechanisms identified by the gene set enrichment analysis suggested that this signature can accurately distinguish high- and low-risk populations, which is possibly closely related to variations in steroid hormone secretion, regulation of endocrine processes, positive regulation of humoral immune response, and androgen response. Results of this study were confirmed in two independent PCa cohorts, namely, The Cancer Genome Atlas cohort and the MSK-IMPACT Clinical Sequencing Cohort, which contributed to the body of scientific evidence for the prediction of biochemical recurrence in patients with PCa. In addition, as the main components of this signature, the effects of the AIFM2 and NFS1 genes on ferroptosis were evaluated and verified by in vivo and in vitro experiments, respectively. The above findings provided new insights and presented potential clinical applications of ferroptosis in PCa.

Improving the prediction of biochemical recurrence after radical prostatectomy with the addition of detailed pathology of the positive surgical margin and cribriform growth

The risk on biochemical recurrence (BCR) after radical prostatectomy (RP) is usually estimated using PSA and pathological stage and grading including the presence of positive surgical margins (PSM). Objective was to investigate whether the presence of cribriform growth in the primary tumor, Grade Group (GG) at the PSM, and length of the PSM have added value in the prognostication. We analyzed data of 835 patients initially treated with RP between 2000 and 2017. Cox regression models were developed to compare the baseline model (PSA, pT-stage, pN-stage, GG at RP, and presence of PSM) with an extended model (adding the presence of cribriform growth, length and GG at the PSM) using the likelihood ratio test. Discrimination was assessed at internal validation by the time-dependent area under the receiver operating characteristic curve (AUC) at 3- and 5-year. A total of 224 men experienced BCR. Median follow-up for men without BCR was 50.4 months (interquartile range, IQR 11.9–95.5). The extended model had a significant better fit, χ2(4) = 31.0, p < 0.001 than the baseline model. The AUC of the 3- and 5-year extended model was 0.85 (95% CI 0.81–0.88) compared to 0.83 (95% CI 0.79–0.87) for the baseline model. Importantly, the presence of cribriform growth in the primary tumor, and GG ≥ 2 at PSM were associated with a higher risk on BCR. In conclusion, the addition of pathological variables improved the prediction of the risk on BCR after RP slightly. However, the clinical implications of this model are important.

Prediction of Biochemical Recurrence After Radical Prostatectomy Based on Preoperative 68Ga-PSMA-11 PET/CT.

PurposeThis study was designed to investigate the prognostic role of preoperative 68Ga-PSMA-11 PET/CT in predicting biochemical recurrence (BCR) of localized prostate cancer (PCa) after radical prostatectomy (RP).MethodsA total of 77 biopsy-confirmed PCa patients with 68Ga-PSMA-11 PET/CT prior to RP were included. A PSMA-ligand PET/CT-based risk model with SUVmax, maximum diameter of the index tumor and T stage was developed for prediction of 2-year BCR using Cox regression analysis. Also, the efficacy of the developed risk model was compared with European Association of Urology risk stratification (D’Amico) and the Cancer of the Prostate Risk Assessment (CAPRA) score. C-index and calibration plot were used to assess discrimination and calibration with internal validation.ResultsWith a median follow-up of 25 months, 23 (29.9%) patients experienced BCR within 2 years after RP. Patients experienced BCR had a significant higher PSA at diagnosis (p<0.001), a higher ISUP grade of biopsy (p=0.044), as well as a higher ISUP grade (p=0.001), a higher possibility of T3 diseases (p=0.001) and positive margin (p=0.008) on postoperative pathology. SUVmax, maximum diameter of the index tumor and T stage on preoperative PSMA-ligand PET/CT were significantly associated with BCR (all p<0.01). PSMA-ligand PET/CT-based risk model had a superior discrimination (c-index 78.5%) and good calibration at internal validation. The efficacy of this model in predicting 2-year BCR after RP was better, compared with CAPRA (c-index 66.3%) and D’Amico (c-index 66.2%). The addition of the PSMA-ligand PET/CT-derived variables also improved the efficacy of the existing models in predicting 2-year BCR (C-index of 78.9% for modified CAPRA and 79.3% for modified D’Amico, respectively).ConclusionA PSMA-ligand PET/CT-based risk model showed good efficacy in predicting 2-year BCR after RP, which needed to be validated by further prospective studies.

Application of Primary/Secondary Circulating Tumor Cells for the Prediction of Biochemical Recurrence in Nonmetastatic Prostate Cancer Patients following Radical Prostatectomy or Radiotherapy: A Meta-Analysis

Background Circulating tumor cells (CTCs) have been regarded as an independent prognostic marker for metastatic castration-resistant prostate cancer (mCRPC). Its prognostic value, however, in nonmetastatic prostate cancer (NMPC) is still unclear. Purpose To elucidate whether CTCs can predict the biochemical recurrence (BCR) in NMPC patients following radical prostatectomy (RP) or radiotherapy (RT). Methods PubMed, Cochrane Database, and Embase and the references in relevant studies were systematically searched. Studies that investigated the correlation of CTCs and BCR in NMPC patients after RP or RT were identified and reviewed. Overall odds ratio (OR) of BCR in such patients with/without CTCs was pooled. We also calculated and pooled overall prevalence of BCR in such CTC-positive patients. Results In total, 12 studies comprising 1917 participants were eligible for the meta-analysis and showed that the presence of secondary circulating tumor cells (SCTCs) is associated with a higher BCR rate of 59% (95% CI: 22%-88%) in patients with NMPC after RP or RT (OR = 6.12; 95% CI: 2.22-16.85; P < 0.001). However, regardless of the presence of primary circulating tumor cells (PCTCs), it has not been shown to be associated with higher BCR. Conclusions Our research demonstrated that SCTC-positive patients are associated with higher BCR compared to SCTC-negative patients in NMPC. Therefore, it is recommended that NMPC patients undergo CTC surveillance intensively after RP or RT.

Deep learning for prediction of prostate cancer biochemical recurrence after radical prostatectomy

Deep learning for prediction of prostate cancer biochemical recurrence after radical prostatectomy

Pathological significance and prognostic role of LATS2 in prostate cancer.

BackgroundLarge tumor suppressor 2 (LATS2) is an important regulator of the Hippo pathway and it plays crucial roles in cell survival and behaviors. Herein, we evaluated the pathological roles of LATS2 in prostate cancer (PC), for which very little information is available.MethodsCell proliferation, migration, and invasion in response to the siRNA‐mediated knockdown (KD) LATS2 expression were evaluated in two PC cell lines (LNCaP and PC3). The expression of LATS2 in specimens from 204 PC patients was investigated immunohistochemically, and the relationships between its expression and clinicopathological features, proliferation index (PI; measured using an anti‐KI‐67 antibody), and biochemical recurrence (BCR) were investigated.ResultsKD of LATS2 increased the growth, migration, and invasion in LNCaP cells and only increased migration in PC3 cells. The expression of LATS2 was negatively associated with the grade group, T, N, M stage, and PI. In addition, the expression of LATS2 was a useful predictor of the histological effects of neoadjuvant hormonal therapy and BCR‐free survival periods. A multivariate analysis model including clinicopathological features showed that negative expression of LATS2 had a significantly higher risk of BCR (odds ratio = 2.95, P < 0.001).ConclusionsLATS2 acts as a tumor suppressor in PC. LATS2 expression is a useful predictor for BCR. LATS2‐related activities are possibly dependent on the androgen‐dependency of PC cells. Therefore, we suggest that LATS2 could be a potential therapeutic target and a useful predictor for outcome in patients with PC.

MP11-09 ADDED VALUE OF RADIOLOGICAL TUMOR STAGE IN PREDICTING EARLY ONCOLOGICAL OUTCOMES IN PROSTATE CANCER PATIENTS UNDERGOING RADICAL PROSTATECTOMY WITHIN CLINICAL STAGE: A STAGE-BY-STAGE ANALYSIS

MP11-09 ADDED VALUE OF RADIOLOGICAL TUMOR STAGE IN PREDICTING EARLY ONCOLOGICAL OUTCOMES IN PROSTATE CANCER PATIENTS UNDERGOING RADICAL PROSTATECTOMY WITHIN CLINICAL STAGE: A STAGE-BY-STAGE ANALYSIS

PD05-09 PET/CT ALLOWS FOR A RELIABLE ASSESSMENT OF THE REAL NODAL BURDEN IN CLINICALLY NODE-POSITIVE PROSTATE CANCER PATIENTS CANDIDATE FOR RADICAL PROSTATECTOMY: RESULTS FROM A MULTI-INSTITUTIONAL SERIES

You have accessJournal of UrologyProstate Cancer: Advanced (including Drug Therapy) I (PD05)1 Sep 2021PD05-09 PET/CT ALLOWS FOR A RELIABLE ASSESSMENT OF THE REAL NODAL BURDEN IN CLINICALLY NODE-POSITIVE PROSTATE CANCER PATIENTS CANDIDATE FOR RADICAL PROSTATECTOMY: RESULTS FROM A MULTI-INSTITUTIONAL SERIES Giorgio Gandaglia, R. Jeffrey Karnes, Charlien Berghen, Wouter Everaerts, Gaëtan Devos, Steven Joniau, Daniele Robesti, Armando Stabile, Simone Scuderi, Francesco Barletta, Vito Cucchiara, Elio Mazzone, Giuseppe Rosiello, Andrea Salonia, Pierre I Karakiewicz, Hendrik Van Poppel, Francesco Montorsi, and Alberto Briganti Giorgio GandagliaGiorgio Gandaglia More articles by this author , R. Jeffrey KarnesR. Jeffrey Karnes More articles by this author , Charlien BerghenCharlien Berghen More articles by this author , Wouter EveraertsWouter Everaerts More articles by this author , Gaëtan DevosGaëtan Devos More articles by this author , Steven JoniauSteven Joniau More articles by this author , Daniele RobestiDaniele Robesti More articles by this author , Armando StabileArmando Stabile More articles by this author , Simone ScuderiSimone Scuderi More articles by this author , Francesco BarlettaFrancesco Barletta More articles by this author , Vito CucchiaraVito Cucchiara More articles by this author , Elio MazzoneElio Mazzone More articles by this author , Giuseppe RosielloGiuseppe Rosiello More articles by this author , Andrea SaloniaAndrea Salonia More articles by this author , Pierre I KarakiewiczPierre I Karakiewicz More articles by this author , Hendrik Van PoppelHendrik Van Poppel More articles by this author , Francesco MontorsiFrancesco Montorsi More articles by this author , and Alberto BrigantiAlberto Briganti More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000001969.09AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: PET/CT scan might identify nodal metastases in prostate cancer (PCa) patients considered for radical prostatectomy (RP). No data addressed the performance characteristics and prognostic value of a positive PET/CT in the nodes for the identification of men more likely to recur METHODS: 346 patients treated with RP and an extended lymph node dissection (LND) with pelvic or retroperitoneal lymphadenopathies at cross-sectional abdominal imaging (n=283 81.8%) vs. choline or PSMA PET/CT (n=63 18.2%) were identified. Lowess analyses explored the non-linear relationship between the number of positive spots and pathologically positive nodes according to imaging (cross-sectional vs PET/CT). Kaplan-Meier analyses assessed time to biochemical recurrence (BCR) in the PET/CT cohort after stratifying patients according to the number of lymphadenopathies. The best cut-off for the number of lymphadenopathies at PET/CT predicting BCR was determined RESULTS: The rate of pN1 disease (74 vs 69%), the number of nodes removed (25 vs 24) and of positive nodes (2 vs 2) at LND did not differ (all P≥0.2). The site of lymphadenopathies was the pelvis only and the pelvis ± retroperitoneum in 84 vs. 88% and 16 vs 12% patients (p=0.3). Cross-sectional imaging was associated with underestimation of nodal burden when ≥2 lymphadenopathies were detected. PET/CT was characterized by excellent concordance with the number of positive nodes at RP. Median follow-up was 70 months and 183 patients experienced BCR. For PET/CT, the most-informative cut-off predicting BCR was 2 clinical lymphadenopathies. When patients selected based on PET/CT were stratified according to the number of positive spots (1 vs ≥2), the 3-year BCR-free survival rates were 48 vs. 21% (p=0.03). This was consistent to what observed when stratifying patients according to the number of positive nodes at final pathology (41 vs 23% p=0.04) CONCLUSIONS: While conventional imaging underestimates the nodal burden, PET/CT is characterized by excellent concordance with final pathology. This resulted in a more reliable prediction of BCR, where the presence of ≥2 positive spots at PET/CT might identify men more likely to experience early recurrence. Source of Funding: - © 2021 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 206Issue Supplement 3September 2021Page: e58-e59 Advertisement Copyright & Permissions© 2021 by American Urological Association Education and Research, Inc.MetricsAuthor Information Giorgio Gandaglia More articles by this author R. Jeffrey Karnes More articles by this author Charlien Berghen More articles by this author Wouter Everaerts More articles by this author Gaëtan Devos More articles by this author Steven Joniau More articles by this author Daniele Robesti More articles by this author Armando Stabile More articles by this author Simone Scuderi More articles by this author Francesco Barletta More articles by this author Vito Cucchiara More articles by this author Elio Mazzone More articles by this author Giuseppe Rosiello More articles by this author Andrea Salonia More articles by this author Pierre I Karakiewicz More articles by this author Hendrik Van Poppel More articles by this author Francesco Montorsi More articles by this author Alberto Briganti More articles by this author Expand All Advertisement Loading ...

MP60-04 SUPERVISED MACHINE LEARNING ALGORITHMS DEMONSTRATE COMPARATIVE ADVANTAGE OVER NOMOGRAMS IN PREDICTING BIOCHEMICAL RECURRENCE AFTER RADICAL PROSTATECTOMY

You have accessJournal of UrologyProstate Cancer: Markers (MP60)1 Sep 2021MP60-04 SUPERVISED MACHINE LEARNING ALGORITHMS DEMONSTRATE COMPARATIVE ADVANTAGE OVER NOMOGRAMS IN PREDICTING BIOCHEMICAL RECURRENCE AFTER RADICAL PROSTATECTOMY Yu Guang Tan, Kenneth Chen, John S. P. Yuen, Hong Hong Huang, Jay K. S. Lim, and Kae Jack Tay Yu Guang TanYu Guang Tan More articles by this author , Kenneth ChenKenneth Chen More articles by this author , John S. P. YuenJohn S. P. Yuen More articles by this author , Hong Hong HuangHong Hong Huang More articles by this author , Jay K. S. LimJay K. S. Lim More articles by this author , and Kae Jack TayKae Jack Tay More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000002095.04AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: After radical prostatectomy (RP), up to one-third of patients experience biochemical recurrence (BCR), which is associated with subsequent metastasis and cancer-specific mortality. We aimed to employ supervised machine learning (ML) algorithms to predict BCR after RP, and compare with conventional regression models and nomograms. METHODS: Utilizing a large prospective Uro-oncology registry, 18 clinicopathological parameters of 1130 consecutive patients who underwent RP (2009 – 2018) were recorded, yielding over 20,000 data points for analysis. The dataset was split in a 70:30 ratio for training and validation. Three ML models: Naïve Bayes (NB), Random forest (RF) and Support Vector Machine (SVM) were studied and compared with traditional regression models and nomograms (Kattan, CAPSURE, John Hopkins [JHH]) to predict BCR at 1,3 and 5 years. RESULTS: Over a median follow-up of 70.0 months, 176 (15.6%) developed BCR, at a median time of 16.0 months (IQR 11.0 – 26.0). Multivariate analyses demonstrated strongest association of BCR with PSA (p: 0.015), surgical margins (p<0.001), extraprostatic extension (p :0.002), seminal vesicle invasion ((p: 0.004) and Grade Group (p<0.001). The 3 ML models demonstrated good prediction of BCR at 1, 3 and 5 years, with the area under curves (AUC) of NB at 0.894, 0.876 and 0.894, RF at 0.846, 0.875 and 0.888 and SVM at 0.835, 0.850 and 0.855 respectively. All models demonstrated robust accuracy (>0.82), good calibration with minimal overfitting, longitudinal consistency and inter-model validity. The ML models were comparable to traditional regression analyses [AUC: 0.797, 0.848 and 0.861] outperformed the three nomograms: Kattan [AUC: 0.815, 0.798 and 0.799], JHH [AUC: 0.820, 0.757 and 0.750] and CAPSURE nomograms [AUC: 0.706, 0.720 and 0.749]. CONCLUSIONS: Supervised ML algorithms can deliver accurate performances and outperform nomograms in predicting BCR after RP. This may facilitate tailored care provisions by identifying high-risk patients who will benefit from early multimodal therapy. Source of Funding: NA © 2021 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 206Issue Supplement 3September 2021Page: e1042-e1042 Advertisement Copyright & Permissions© 2021 by American Urological Association Education and Research, Inc.MetricsAuthor Information Yu Guang Tan More articles by this author Kenneth Chen More articles by this author John S. P. Yuen More articles by this author Hong Hong Huang More articles by this author Jay K. S. Lim More articles by this author Kae Jack Tay More articles by this author Expand All Advertisement Loading ...

Early prediction of prostate cancer biochemical recurrence and identification of disease persistence using PSA isoforms and human kallikrein-2.

The role of isoforms of prostate specific antigen (PSA) and other kallikrein-related markers in early detection of biochemical recurrence (BCR) after radical prostatectomy (RP) is not well known and serum PSA is currently used in preoperative risk nomograms. The aim of this research was to study pre- and early postoperative levels of important PSA isoforms and human kallikrein-2 to determine their ability to predict BCR and identify disease persistence (DP). This study included 128 consecutive patients who underwent RP for clinically localized prostate cancer. PSA, fPSA, %fPSA, [-2]proPSA, PHI and hK2 were measured preoperatively, at 1 and 3 months after RP. We determined the ability of these markers to predict BCR and identify DP. The DP and BCR rate were 11.7%and 20.3%respectively and the median follow up was 64 months (range 3-76 months). Preoperatively, the independent predictors of BCR were PSA (p-value 0.029), [-2]proPSA (p-value 0.002) and PHI (p-value 0.0003). Post-RP, PSA was the single marker correlating with BCR, both at one (p-value 0.0047) and 3 months (p-value 0.0004). PSA, fPSA, [-2]proPSA and PHI significantly correlated to DP at 1 and 3 months post-RP (p-value < 0.05), although PSA had the most significant existing correlation (p-value < 0.0001). [-2]proPSA and PHI are preoperative predictors of BCR and DP that outperform the currently used serum PSA. At the early postoperative period there is no additional benefit of the other markers tested.