Explainable machine learning predicts survival of retroperitoneal liposarcoma: A study based on the SEER database and external validation in China.

OBJECTIVE: We have developed explainable machine learning models to predict the overall survival (OS) of retroperitoneal liposarcoma (RLPS) patients. This approach aims to enhance the explainability and transparency of our modeling results. METHODS: We collected clinicopathological information of RLPS patients from The Surveillance, Epidemiology, and End Results (SEER) database and allocated them into training and validation sets with a 7:3 ratio. Simultaneously, we obtained an external validation cohort from The First Affiliated Hospital of Naval Medical University (Shanghai, China). We performed LASSO regression and multivariate Cox proportional hazards analysis to identify relevant risk factors, which were then combined to develop six machine learning (ML) models: Cox proportional hazards model (Coxph), random survival forest (RSF), ranger, gradient boosting with component-wise linear models (GBM), decision trees, and boosting trees. The predictive performance of these ML models was evaluated using the concordance index (C-index), the integrated cumulative/dynamic area under the curve (AUC), and the integrated Brier score, as well as the Cox-Snell residual plot. We also used time-dependent variable importance, analysis of partial dependence survival plots, and the generation of aggregated survival SHapley Additive exPlanations (SurvSHAP) plots to provide a global explanation of the optimal model. Additionally, SurvSHAP (t) and survival local interpretable model-agnostic explanations (SurvLIME) plots were used to provide a local explanation of the optimal model. RESULTS: The final ML models are consisted of six factors: patient's age, gender, marital status, surgical history, as well as tumor's histopathological classification, histological grade, and SEER stage. Our prognostic model exhibits significant discriminative ability, particularly with the ranger model performing optimally. In the training set, validation set, and external validation set, the AUC for 1, 3, and 5 year OS are all above 0.83, and the integrated Brier scores are consistently below 0.15. The explainability analysis of the ranger model also indicates that histological grade, histopathological classification, and age are the most influential factors in predicting OS. CONCLUSIONS: The ranger ML prognostic model exhibits optimal performance and can be utilized to predict the OS of RLPS patients, offering valuable and crucial references for clinical physicians to make informed decisions in advance.

Survival nomogram and risk classification system for patients with adrenocortical carcinoma: a study based on the SEER database and external validation in China.

Background: Adrenocortical carcinoma (ACC) is an extremely rare and highly invasive malignant tumor. However, there is currently no reliable method to predict the prognosis of ACC. Our objective is to construct a nomogram and a risk classification system to predict the 1-year, 3-year, and 5-year overall survival (OS) of ACC. Methods: We retrieved clinicopathological data of patients diagnosed with ACC in The Surveillance, Epidemiology, and End Results (SEER) database and divided them into training and validation cohorts with a 7:3 ratio. Simultaneously, we collected an external validation cohort from The First Affiliated Hospital of Naval Medical University (Shanghai, China). Univariate and multivariate Cox analyses were performed to identify relevant risk factors, which were then combined to develop a correlation nomogram. The predictive performance of the nomogram was evaluated using the concordance index (C-index), receiver-operating characteristic curve (ROC), and calibration curves. Decision curve analysis (DCA) was applied to assess the clinical utility of the nomogram. In addition, Kaplan-Meier survival curves were generated to demonstrate the variation in OS between groups. Results: The final nomogram consisted of five factors: age, T, N, M, and history of chemotherapy. Our prognostic model demonstrated significant discriminative ability, with C-index and the area under the receiver operating characteristic (AUC) values exceeding 0.70. Additionally, DCA validated the clinical utility of the nomogram. In the entire cohort, the median OS for patients in the low- and high-risk groups was 70 and 10 months, respectively. Conclusions: A nomogram and a corresponding risk classification system were developed in order to predict the OS of patients diagnosed with ACC. These tools have the potential to provide valuable support for patient counseling and assist in the decision-making process related to treatment options.

Establishment of bladder cancer spheroids and cultured in microfluidic platform for predicting drug response.

Cisplatin-containing combination chemotherapy has been used as the standard treatment for bladder cancer patients at advanced stage. However, nearly 50% of patients are nonresponders. To guide the selection of more effective chemotherapeutic agents, a bladder cancer spheroids microfluidic drug sensitivity analysis system was established in this study. Bladder cancer spheroids were established and successfully cultured in a customized microfluidic device to assess their response to different chemotherapeutic agents. The in vitro drug sensitivity results were also compared to patient-derived xenograft (PDX) models and clinical responses of patients. As a result, bladder cancer spheroids faithfully recapitulate the histopathological and genetic features of their corresponding parental tumors. Furthermore, the in vitro drug sensitivity outcomes of spheroids (n = 8) demonstrated a high level of correlation with the PDX (n = 2) and clinical response in patients (n = 2). Our study highlights the potential of combining bladder cancer spheroids and microfluidic devices as an efficient and accurate platform for personalized selection of chemotherapeutic agents.

Patients With Muscle-Invasive Bladder Cancer With Lymphovascular Invasion in Transurethral Resection Specimen Benefits Most From Platinum-Based Neoadjuvant Chemotherapy.

PURPOSE: The survival benefit of neoadjuvant chemotherapy (NAC) before definitive radical cystectomy (RC) varied among patients, suggesting proper selection of patients for NAC to maximize the survival benefit. This study aimed to investigate the role of lymphovascular invasion (LVI) in transurethral resection (TUR) specimens in selecting patients with MIBC for NAC. METHODS: Two retrospective cohorts of patients with cT2-4aN0 MIBC who underwent RC from 2004 to 2015 provided by Lund University were included. Inverse probability weighting was applied to make the NAC-treated (NAC) and untreated (non-NAC) cohorts comparable. Survival benefits were estimated with Kaplan-Meier curves and Cox proportional hazards models. The primary endpoint was cancer-specific survival (CSS). LVI in TUR specimens and molecular taxonomies (BASE47, UNC, and LundTax) were examined, and bulk RNA-seq datasets were explored for LVI-relevant signatures. RESULTS: A total of 341 patients with cT2-4aN0 MIBC were included. The NAC cohort included 125 patients, whereas the non-NAC cohort included 216 patients. The 3-year CSS benefit of NAC was 7.1%. For patients with positive LVI in TUR specimens, the 3-year CSS benefit of NAC was 26.2% (48.1% vs. 74.3%), with a risk reduction of 56% (HR = 0.44, P = .03). A sensitivity analysis confirmed a significant interaction between LVI and NAC. This study failed to identify the molecular subtypes that maximized the survival benefit of NAC. Exploration of LVI-relevant signatures remains inconclusive. CONCLUSIONS: LVI in TUR specimens could help identify patients with MIBC who would derive maximal survival benefit from NAC. Further prospective validation is necessary.

Comprehensive identification of onco-exaptation events in bladder cancer cell lines revealed L1PA2-SYT1 as a prognosis-relevant event.

Transposable elements (TEs) can provide ectopic promoters to drive the expression of oncogenes in cancer, a mechanism known as onco-exaptation. Onco-exaptation events have been extensively identified in various cancers, with bladder cancer showing a high frequency of onco-exaptation events (77%). However, the effect of most of these events in bladder cancer remains unclear. This study identified 44 onco-exaptation events in 44 bladder cancer cell lines in 137 RNA-seq datasets from six publicly available cohorts, with L1PA2 contributing the most events. L1PA2-SYT1, L1PA2-MET, and L1PA2-XCL1 had the highest frequency not only in cell lines but also in TCGA-BLCA samples. L1PA2-SYT1 showed significant tumor specificity and was found to be activated by CpG island demethylation in its promoter. The upregulation of L1PA2-SYT1 enhances the in vitro invasion of bladder cancer and is an independent risk factor for patient's overall survival, suggesting L1PA2-SYT1 being an important event that promotes the development of bladder cancer.

Value of digital rectal examination in patients with suspected prostate cancer: a prospective cohort analysis study.

Background: Digital rectal examination (DRE) is a straightforward, cost-effective, practical, and time-honored physical examination method that plays a valuable role in the detection of prostate cancer (PCa). Nevertheless, with the advent of the prostate-specific antigen (PSA) era, the necessity of performing DRE has become a subject of debate. Our aim was to investigate the diagnostic efficacy and adjunctive role of DRE in a population (Prostate Imaging Reporting and Data System (PI-RADS), PI-RADS ≥3 or PSA ≥4 ng/mL) suspected of PCa. Methods: Five hundred and ninety-seven patients with suspected PCa requiring referral for biopsy were prospectively enrolled consecutively from February 2020 to May 2021. All patients received DRE and corresponding clinical diagnosis by a urologist before biopsy. According to the collected clinical and pathological information, the diagnostic performance of DRE in different PSA stratifications, and its association with tumor location and Gleason score (GS) were statistically analyzed. Results: Among patients with suspected cancer, the diagnostic accuracy of DRE was 63.45%. Compared with central zone or transition zone tumors, the recall rate of peripheral zone PCa with DRE-positive results was higher (65.50% vs. 34.55%). DRE-positive results were significantly correlated with GS ≥7 PCa (P<0.001), and the average GS of DRE-positive PCa patients was significantly higher than that of DRE-negative (7.92 vs. 7.11, P<0.001). Conclusions: DRE may help physicians further judge the necessity of biopsy in patients with elevated PSA, and preliminarily estimate the location and invasiveness of the tumor. However, it is still necessary to explore the value of DRE in a normal PSA population.

Association between testosterone and cancers risk in women: a two-sample Mendelian randomization study.

OBJECTIVE: Previous observational studies have explored the correlation between testosterone and cancer risk. However, the causal association between testosterone and various cancer types in women remains inconclusive. The objective of this Mendelian randomization study is to evaluate the causal links between total testosterone (TT) and bioavailable testosterone (BT) with cancer risk in females. METHODS: Initially, a rigorous quality control process was employed to identify suitable instrumental single nucleotide polymorphisms (SNPs) associated with the exposure under investigation that exhibited a significant association. The genetic causal relationship between female testosterone levels and the risk of developing cancers was examined through a two-sample Mendelian randomization. Various analytical methods, including inverse-variance weighted (IVW), MR-Egger, weighted median, simple mode, and weighted mode, were applied in the investigation. Key findings were primarily based on the results obtained via IVW (random effects), and sensitivity analyses were conducted to assess the reliability of the obtained results. Furthermore, maximum likelihood, penalized weighted median, and IVW (fixed effects) methods were utilized to further validate the robustness of the results. RESULTS: Based on the results of IVW analysis, our study indicated a positive causal relationship between BT and breast cancer (OR = 1.1407, 95%CI: 1.0627-1.2244, P = 0.0015) and endometrial cancer (OR = 1.4610, 95%CI: 1.2695-1.6813, P = 1.22E-06). Moreover, our findings also showed a positive causal association between TT and breast cancer (OR = 1.1764, 95%CI: 1.0846-1.2761, P = 0.0005), cervical cancer(OR = 1.0020, 95%CI: 1.0007-1.0032, P = 0.0077), and endometrial cancer(OR = 1.4124, 95%CI: 1.2083-1.6511, P = 0.0001). Additionally, our results demonstrated a negative causal relationship between BT and ovarian cancer (OR = 0.8649, 95%CI: 0.7750-0.9653, P = 0.0320). However, no causal relationship was found between BT, TT and other types of cancer (corrected P > 0.05). CONCLUSIONS: This study elucidates the role of testosterone on the development of breast cancer, endometrial cancer, ovarian cancer, and cervical cancer. It also hints at a potential but fragile link between testosterone and bladder cancer, as well as thyroid cancer. Nonetheless, it's worth noting that no statistically significant relationship between testosterone and various other types of cancer in females was identified.

Preoperative urine sediment chromosomal instability level predicts urothelial cancer prognosis.

PURPOSE: Urothelial carcinomas (UCs) are often characterized by frequent recurrences after surgery, making UC one of the costliest cancers. Chromosomal instability (CIN) has been proven to be a hallmark of UCs and is related to the prognosis of many cancer types. In this study, we evaluated CIN of urine sediments as a prognostic indicator for UCs. METHODS: Patients with UC were prospectively recruited. Preoperative urine samples were collected for whole genome sequencing and urine cytology tests. Patients underwent standard-of-care treatment and were followed up until disease relapse or study ended. Concordance and accuracy of CIN alone or in combination with cytology in predicting disease relapse were assessed. The value of CIN combined with European Organization for Research and Treatment of Cancer (EORTC) model were also analyzed. RESULTS: A total of 137 patients with UCs were included in this study. Median follow-up was 44.2 months and 41.61% patients suffered from cancer relapse. Patients with CIN-high indicated higher relapse rate, and this distinction was significant for patients underwent transurethral resection of bladder tumor (57.89% vs. 34.29%, P = 0.016). Combination of cytology and CIN result could further classified patients into subgroups with distinct relapse risks. Meanwhile, the combination of CIN and EORTC model significantly improved the prediction accuracy compared with EORTC alone (Harrel's C-index: 0.71 vs. 0.65). CONCLUSION: CIN level of preoperative urine exfoliated cells had robust prognostic value for bladder cancer patients underwent TURBT. The prognostic model by combining CIN and EORTC may help in stratifying patients to optimize follow-up regimen.

Knockdown of kinesin family member 4A inhibits cell proliferation, migration, and invasion while promoting apoptosis of urothelial bladder carcinoma cells.

BACKGROUND: Kinesin family member 4A (KIF4A) is upregulated in a variety of cancers. However, its expression and potential downstream targets in urothelial bladder carcinoma (UBC) remain unclear. METHODS: Expression data of KIF4A in UBC and noncancerous tissues were downloaded from the GEPIA database. Cell proliferation, migration, invasion, and apoptosis of T24 and 5637 UBC cells were examined using wound healing, transwell, colony formation, CCK-8, and flow cytometry assays. KIF4A and potential downstream genes were analyzed using qRT-PCR, western blot analysis, and immunohistochemistry. RESULTS: In UBC samples, KIF4A expression was significantly higher than in corresponding noncancerous samples. UBC patients with high KIF4A expression had poor cancer-specific survival and overall survival. Knockdown of KIF4A significantly inhibited proliferation and promoted apoptosis of UBC cells, accompanied by dephosphorylation of AKT and increased the protein level of proapoptotic factors. Additionally, knockdown of KIF4A reduced migration and invasion of UBC cells whereas overexpression of KIF4A exhibited opposite effects, along with altered protein level in epithelial-mesenchymal transition-related genes. Furthermore, overexpression of YAP1 promoted KIF4A expression whereas knockdown of YAP1 suppressed KIF4A expression in UBC cells. Alternatively, KIF4A knockdown reduced YAP1 nuclear protein level whereas KIF4A overexpression suppressed YAP1 phosphorylation and facilitated YAP1 nuclear translocation. CONCLUSIONS: KIF4A upregulation correlates with poor prognosis of UBC. Knockdown of KIF4A inhibits proliferation, migration, and invasion of UBC cells while inducing apoptosis possibly through dephosphorylation of AKT, changes in EMT-related genes, and interaction with YAP1.

Identification and validation of immunohistochemical marker panels to predict the prognosis of muscle invasive bladder cancer.

Background: Currently, the treatment regimen of bladder cancer depends on the stage and grade. Yet, patients with similar histopathological characteristics may have distinct prognosis. Luminal/basal subtyping had proved to be a satisfactory subtyping method. Here we intended to evaluate immunohistochemistry, a more clinically-practical method, in luminal/basal classification and further risk-stratification. Methods: Patients diagnosed with urothelial carcinoma of the bladder in Changhai Hospital were retrospectively recruited and corresponding formalin-fixed paraffin embedded blocks were acquired. Tissue microarrays (TMAs) of these patients were established followed by immunohistochemical (IHC) staining of 14 markers. Patients were classified into luminal or basal subtype according to CK5/6, CK14, CK20 and GATA3 expression. Further subtyping of luminal and basal tumors was performed according to the expression of other markers. Results: A total of 236 patients were included: 163 and 73 patients were assigned to training and validation cohorts, respectively. Patients with basal tumor were related with poorer prognosis compared to those with luminal tumor (P=0.025 and 0.008 in training and validation cohorts, respectively). We further revealed luminal muscle invasive bladder cancer (MIBC) patients could be further categorized into subgroups with different risks. Cytoplasmic YAP1 and CCNB1 were selected as classifier, patients with low expression of cytoplasmic YAP1 or CCNB1 were independent risk factor for poorer prognosis (hazard ratio =2.19, P=0.04). Conclusions: Molecular subtyping into luminal/basal subtype and risk stratification method using a 2-marker method by immunohistochemistry can be an economical, clinically practical method to predict patient prognosis and could help to develop treatment strategy and follow-up schedule in clinical practice.

Fatty Acid Synthase Is the Key Regulator of Fatty Acid Metabolism and Is Related to Immunotherapy in Bladder Cancer.

Fatty acid metabolism (FAM) genes are potentially useful for predicting prognosis and immunotherapy response in bladder cancer (BC). To examine this, we constructed a prognostic model and identified key FAM genes in BC. Using transcriptional expression profiles and clinical data of BC patients from public datasets and Changhai (CH) hospital, we built and validated a risk-score model based on 13 prognostic FAM genes. Differential gene expression identified fatty acid synthase (FASN) as central to fatty acid metabolism in BC. FASN was differentially expressed between normal and tumor tissue, and was related to survival. In the CH dataset, FASN independently predicted muscle-invasive BC. FASN differential expression was significantly related to immune-cell infiltration and patients with low FASN expression responded better to immune checkpoint inhibitor (ICI) treatment. SREBF1 was predicted as the most significant transcription factor for FASN. Competing endogenous RNA network analysis suggested that lncRNA AC107027.3 may upregulate FASN by competitively binding miR-27A-3p, thereby regulating the immunotherapy response in BC. Dasatinib and temsirolimus are potential FASN-targeting drugs. Our model efficiently predicted prognosis in BC. FASN is central to fatty acid metabolism, and a potential indicator and regulator of ICI treatment.

Construction and Validation of Novel Prediction Tools Based on Large Population-Based Database to Predict the Prognosis of Urachal Cancer After Surgery.

BACKGROUND: Urachal cancer is a rare neoplasm in the urological system. To our knowledge, no published study has explored to establish a model for predicting the prognosis of urachal cancer. The present study aims to develop and validate nomograms for predicting the prognosis of urachal cancer based on clinicopathological parameters. METHODS: Based on the data from the Surveillance, Epidemiology, and End Results database, 445 patients diagnosed with urachal cancer between 1975 and 2018 were identified as training and internal validation cohort; 84 patients diagnosed as urachal cancer from 2001 to 2020 in two medical centers were collected as external validation cohort. Nomograms were developed using a multivariate Cox proportional hazards regression analysis in the training cohort, and their performance was evaluated in terms of its discriminative ability, calibration, and clinical usefulness by statistical analysis. RESULTS: Three nomograms based on tumor-node-metastasis (TNM), Sheldon and Mayo staging system were developed for predicting cancer-specific survival (CSS) of urachal cancer; these nomograms all showed similar calibration and discrimination ability. Further internal (c-index 0.78) and external (c-index 0.81) validation suggested that Sheldon model had superior discrimination and calibration ability in predicting CSS than the other two models. Moreover, we found that the Sheldon model was able to successfully classify patients into different risk of mortality both in internal and external validation cohorts. Decision curve analysis proved that the nomogram was clinically useful and applicable. CONCLUSIONS: The nomogram model with Sheldon staging system was recommended for predicting the prognosis of urachal cancer. The proposed nomograms have promising clinical applicability to help clinicians on individualized patient counseling, decision-making, and clinical trial designing.

Impact of previous, simultaneous or intravesical recurrence bladder cancer on prognosis of upper tract urothelial carcinoma after nephroureterectomy: a large population-based study.

BACKGROUND: Currently, the impact of previous, simultaneous, or subsequent bladder cancer on the prognosis of upper urinary tract urothelial carcinoma (UTUC) is controversial. We aimed to investigate the impact of previous, simultaneous or intravesical recurrence (IVR) bladder cancer on the prognosis of UTUC based on a large population-based cohort from the Surveillance, Epidemiology, and End Results (SEER) database. METHODS: A total of 8,431 UTUC patients diagnosed from 2004 to 2018 met the inclusion criteria were identified based on the SEER database. We evaluated the impact of bladder cancer on the prognosis of UTUC by Kaplan-Meier method and propensity score matching (PSM). RESULTS: In all, 6,831 patients only had UTUC (UTUC-only), 880 patients with previous or simultaneous bladder cancer (UTUC-Bca), 720 patients with IVR (UTUC-IVR). After adjusting baseline covariates that varied significantly among groups, we found UTUC-Bca cohort, regardless of tumor grade and stage, had poorer prognosis than UTUC-only cohort. In general, we demonstrated IVR had no significant impact on the prognosis of UTUC compared to PSM matched patients without IVR. However, subgroup analysis revealed that UTUC patients with subsequent MIBC recurrence or shorter interval (<20 months) between UTUC and IVR had worse prognosis compared with UTUC-only cohort. CONCLUSIONS: UTUC patients with previous or simultaneous bladder cancer, IVR with MIBC, and shorter interval between UTUC and IVR were significant predictor for worse prognosis. Thus, more stringent postoperative surveillance and active treatment strategies should be considered for UTUC patients with those risk factors.

Noninvasive Detection of Urothelial Carcinoma by Cost-effective Low-coverage Whole-genome Sequencing from Urine-Exfoliated Cell DNA.

PURPOSE: Urothelial carcinoma is a malignant cancer with frequent chromosomal aberrations. Here, we investigated the application of a cost-effective, low-coverage whole-genome sequencing technology in detecting all chromosomal aberrations. EXPERIMENTAL DESIGN: Patients with urothelial carcinomas and nontumor controls were prospectively recruited in clinical trial NCT03998371. Urine-exfoliated cell DNA was analyzed by Illumina HiSeq XTen, followed by genotyping with a customized bioinformatics workflow named Urine Exfoliated Cells Copy Number Aberration Detector (UroCAD). RESULTS: In the discovery phase, urine samples from 126 patients with urothelial carcinomas and 64 nontumor disease samples were analyzed. Frequent chromosome copy-number changes were found in patients with tumor as compared with nontumor controls. A novel diagnosis model, UroCAD, was built by incorporating all the autosomal chromosomal changes. The model reached performance of AUC = 0.92 (95% confidence interval, 89.4%-97.3%). At the optimal cutoff, |Z| ≥ 3.21, the sensitivity, specificity, and accuracy were 82.5%, 96.9%, and 89.0%, respectively. The prediction positivity was found correlated with tumor grade (P = 0.01). In the external validation cohort of 95 participants, the UroCAD assay identified urothelial carcinomas with an overall sensitivity of 80.4%, specificity of 94.9%, and AUC of 0.91. Meanwhile, UroCAD assay outperformed cytology tests with significantly improved sensitivity (80.4% vs. 33.9%; P < 0.001) and comparable specificity (94.9% vs. 100%; P = 0.49). CONCLUSIONS: UroCAD could be a robust urothelial carcinoma diagnostic method with improved sensitivity and similar specificity as compared with cytology tests. It may be used as a noninvasive approach for diagnosis and recurrence surveillance in urothelial carcinoma prior to the use of cystoscopy, which would largely reduce the burden on patients.

Cuprous oxide nanoparticles trigger reactive oxygen species-induced apoptosis through activation of erk-dependent autophagy in bladder cancer.

Cisplatin-based chemotherapy is the first-line treatment for patients with advanced bladder cancer. However, as more than 50% of patients are ineligible for cisplatin-based chemotherapy, there is an urgent need to develop new drugs. Cuprous oxide nanoparticles (CONPs), as a new nano-therapeutic agent, have been proved to be effective in many kinds of tumors. In the present study, CONPs showed dose-dependent and time-dependent inhibitory effects on various bladder cancer cell lines (T24, J82, 5637, and UMUC3) and weak inhibitory effects on non-cancerous epithelial cells (SVHUCs). We found that CONPs induced cell cycle arrest and apoptosis in bladder cancer cells. We further demonstrated that the potential mechanisms of CONP-induced cytotoxicity were apoptosis, which was triggered by reactive oxygen species through activation of ERK signaling pathway, and autophagy. Moreover, the cytotoxic effect of CONPs on bladder cancer was confirmed both in orthotopic xenografts and subcutaneous nude mouse models, indicating that CONPs could significantly suppress the growth of bladder cancer in vivo. In further drug combination experiments, we showed that CONPs had a synergistic drug-drug interaction with cisplatin and gemcitabine in vitro, both of which are commonly used chemotherapy agents for bladder cancer. We further proved that CONPs potentiated the antitumor activity of gemcitabine in vivo without exacerbating the adverse effects, suggesting that CONPs and gemcitabine can be used for combination intravesical chemotherapy. In conclusion, our preclinical data demonstrate that CONPs are a promising nanomedicine against bladder cancer and provide good insights into the application of CONPs and gemcitabine in combination for intravesical bladder cancer treatment.