RESUMEN
BACKGROUND: Preclinical models recapitulating the metastatic phenotypes are essential for developing the next-generation therapies for metastatic prostate cancer (mPC). We aimed to establish a cohort of clinically relevant mPC models, particularly androgen receptor positive (AR+) bone metastasis models, from LuCaP patient-derived xenografts (PDX) that reflect the heterogeneity and complexity of mPC. METHODS: PDX tumors were dissociated into single cells, modified to express luciferase, and were inoculated into NSG mice via intracardiac injection. The progression of metastases was monitored by bioluminescent imaging. Histological phenotypes of metastases were characterized by immunohistochemistry and immunofluorescence staining. Castration responses were further investigated in two AR-positive models. RESULTS: Our PDX-derived metastasis (PDM) model collection comprises three AR+ adenocarcinomas (ARPC) and one AR- neuroendocrine carcinoma (NEPC). All ARPC models developed bone metastases with either an osteoblastic, osteolytic, or mixed phenotype, while the NEPC model mainly developed brain metastasis. Different mechanisms of castration resistance were observed in two AR+ PDM models with distinct genotypes, such as combined loss of TP53 and RB1 in one model and expression of AR splice variant 7 (AR-V7) expression in another model. Intriguingly, the castration-resistant tumors displayed inter- and intra-tumor as well as organ-specific heterogeneity in lineage specification. CONCLUSION: Genetically diverse PDM models provide a clinically relevant system for biomarker identification and personalized medicine in metastatic castration-resistant prostate cancer.
Asunto(s)
Neoplasias Óseas , Neoplasias de la Próstata , Receptores Androgénicos , Animales , Humanos , Masculino , Ratones , Adenocarcinoma/patología , Adenocarcinoma/secundario , Adenocarcinoma/metabolismo , Adenocarcinoma/genética , Neoplasias Óseas/secundario , Neoplasias Óseas/metabolismo , Carcinoma Neuroendocrino/patología , Carcinoma Neuroendocrino/metabolismo , Carcinoma Neuroendocrino/genética , Modelos Animales de Enfermedad , Neoplasias de la Próstata/patología , Neoplasias de la Próstata/metabolismo , Neoplasias de la Próstata/genética , Neoplasias de la Próstata Resistentes a la Castración/patología , Neoplasias de la Próstata Resistentes a la Castración/metabolismo , Neoplasias de la Próstata Resistentes a la Castración/genética , Receptores Androgénicos/genética , Receptores Androgénicos/metabolismoRESUMEN
OBJECTIVE: To determine the rigor, generalizability, and reproducibility of published classification and detection artificial intelligence (AI) models for prostate cancer (PCa) on MRI using the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) guidelines, a 42-item checklist that is considered a measure of best practice for presenting and reviewing medical imaging AI research. MATERIALS AND METHODS: This review searched English literature for studies proposing PCa AI detection and classification models on MRI. Each study was evaluated with the CLAIM checklist. The additional outcomes for which data were sought included measures of AI model performance (eg, area under the curve [AUC], sensitivity, specificity, free-response operating characteristic curves), training and validation and testing group sample size, AI approach, detection versus classification AI, public data set utilization, MRI sequences used, and definition of gold standard for ground truth. The percentage of CLAIM checklist fulfillment was used to stratify studies into quartiles. Wilcoxon's rank-sum test was used for pair-wise comparisons. RESULTS: In all, 75 studies were identified, and 53 studies qualified for analysis. The original CLAIM items that most studies did not fulfill includes item 12 (77% no): de-identification methods; item 13 (68% no): handling missing data; item 15 (47% no): rationale for choosing ground truth reference standard; item 18 (55% no): measurements of inter- and intrareader variability; item 31 (60% no): inclusion of validated interpretability maps; item 37 (92% no): inclusion of failure analysis to elucidate AI model weaknesses. An AUC score versus percentage CLAIM fulfillment quartile revealed a significant difference of the mean AUC scores between quartile 1 versus quartile 2 (0.78 versus 0.86, P = .034) and quartile 1 versus quartile 4 (0.78 versus 0.89, P = .003) scores. Based on additional information and outcome metrics gathered in this study, additional measures of best practice are defined. These new items include disclosure of public dataset usage, ground truth definition in comparison to other referenced works in the defined task, and sample size power calculation. CONCLUSION: A large proportion of AI studies do not fulfill key items in CLAIM guidelines within their methods and results sections. The percentage of CLAIM checklist fulfillment is weakly associated with improved AI model performance. Additions or supplementations to CLAIM are recommended to improve publishing standards and aid reviewers in determining study rigor.
Asunto(s)
Inteligencia Artificial , Próstata , Masculino , Humanos , Lista de Verificación , Reproducibilidad de los Resultados , Algoritmos , Imagen por Resonancia MagnéticaRESUMEN
The role of multiparametric MRI (mpMRI) in the detection of prostate cancer is well-established. Based on the limited role of dynamic contrast enhancement (DCE) in PI-RADS v2.1, the risk of potential side effects, and the increased cost and time, there has been an increase in studies advocating for the omission of DCE from MRI assessments. Per PI-RADS v2.1, DCE is indicated in the assessment of PI-RADS 3 lesions in the peripheral zone, with its most pronounced effect when T2WI and DWI are of insufficient quality. The aim of this study was to evaluate the methodology and reporting in the literature from the past 5 years regarding the use of DCE in prostate MRI, especially with respect to the indications for DCE as stated in PI-RADS v2.1, and to describe the different approaches used across the studies. We searched for studies investigating the use of bpMRI and/or mpMRI in the detection of clinically significant prostate cancer between January 2017 and April 2022 in the PubMed, Web of Science, and Google Scholar databases. Through the search process, a total of 269 studies were gathered and 41 remained after abstract and full-text screening. The following information was extracted from the eligible studies: general clinical and technical characteristics of the studies, the number of PI-RADS 3 lesions, different definitions of clinically significant prostate cancer (csPCa), biopsy thresholds, reference standard methods, and number and experience of readers. Forty-one studies were included in the study. Only 51% (21/41) of studies reported the prevalence of csPCa in their equivocal lesion (PI-RADS category 3 lesions) subgroups. Of the included studies, none (0/41) performed a stratified sub-analysis of the DCE benefit versus MRI quality and 46% (19/41) made explicit statements about removing MRI scans based on a range of factors including motion, noise, and image artifacts. Furthermore, the number of studies investigating the role of DCE using readers with varying experience was relatively low. This review demonstrates that a high proportion of the studies investigating whether bpMRI can replace mpMRI did not transparently report information inherent to their study design concerning the key indications of DCE, such as the number of clinically insignificant/significant PI-RADS 3 lesions, nor did they provide any sub-analyses to test image quality, with some removing bad quality MRI scans altogether, or reader-experience-dependency indications for DCE. For the studies that reported on most of the DCE indications, their conclusions about the utility of DCE were heavily definition-dependent (with varying definitions of csPCa and of the PI-RADS category biopsy significance threshold). Reporting the information inherent to the study design and related to the specific indications for DCE as stated in PI-RADS v2.1 is needed to determine whether DCE is helpful or not. With most of the recent literature being retrospective and not including the data related to DCE indications in particular, the ongoing dispute between bpMRI and mpMRI is likely to linger.
RESUMEN
RATIONALE AND OBJECTIVE: The combined use of prostate cancer radiotherapy and MRI planning is increasingly being used in the treatment of clinically significant prostate cancers. The radiotherapy dosage quantity is limited by toxicity in organs with de-novo genitourinary toxicity occurrence remaining unperturbed. Estimation of the urethral radiation dose via anatomical contouring may improve our understanding of genitourinary toxicity and its related symptoms. Yet, urethral delineation remains an expert-dependent and time-consuming procedure. In this study, we aim to develop a fully automated segmentation tool for the prostatic urethra. MATERIALS AND METHODS: This study incorporated 939 patients' T2-weighted MRI scans (train/validation/test/excluded: 657/141/140/1 patients), including in-house and public PROSTATE-x datasets, and their corresponding ground truth urethral contours from an expert genitourinary radiologist. The AI model was developed using MONAI framework and was based on a 3D-UNet. AI model performance was determined by Dice score (volume-based) and the Centerline Distance (CLD) between the prediction and ground truth centers (slice-based). All predictions were compared to ground truth in a systematic failure analysis to elucidate the model's strengths and weaknesses. The Wilcoxon-rank sum test was used for pair-wise comparison of group differences. RESULTS: The overall organ-adjusted Dice score for this model was 0.61 and overall CLD was 2.56 mm. When comparing prostates with symmetrical (n = 117) and asymmetrical (n = 23) benign prostate hyperplasia (BPH), the AI model performed better on symmetrical prostates compared to asymmetrical in both Dice score (0.64 vs. 0.51 respectively, p < 0.05) and mean CLD (2.3 mm vs. 3.8 mm respectively, p < 0.05). When calculating location-specific performance, the performance was highest at the apex and lowest at the base location of the prostate for Dice and CLD. Dice location dependence: symmetrical (Apex, Mid, Base: 0.69 vs. 0.67 vs. 0.54 respectively, p < 0.05) and asymmetrical (Apex, Mid, Base: 0.68 vs. 0.52 vs. 0.39 respectively, p < 0.05). CLD location dependence: symmetrical (Apex, Mid, Base: 1.43 mm vs. 2.15 mm vs. 3.28 mm, p < 0.05) and asymmetrical (Apex, Mid, Base: 1.83 mm vs. 3.1 mm vs. 6.24 mm, p < 0.05). CONCLUSION: We developed a fully automated prostatic urethra segmentation AI tool yielding its best performance in prostate glands with symmetric BPH features. This system can potentially be used to assist treatment planning in patients who can undergo whole gland radiation therapy or ablative focal therapy.
Asunto(s)
Hiperplasia Prostática , Neoplasias de la Próstata , Inteligencia Artificial , Humanos , Imagen por Resonancia Magnética/métodos , Masculino , Próstata/diagnóstico por imagen , Neoplasias de la Próstata/diagnóstico por imagen , Neoplasias de la Próstata/radioterapia , Dosificación Radioterapéutica , Uretra/diagnóstico por imagenRESUMEN
CT and MRI are both commonly used in prostate cancer (PCa) management, which includes a large spectrum from screening positive pre-diagnosis phase to metastatic disease. CT and MRI have continually evolved to meet the changing demands for PCa management. For CT, novel techniques such as dual energy CT and photon counting CT show promising results for tissue characterization and quantification. For MRI, the detection, staging, and management of prostate cancer has been significantly improved by the development of multiparametric, biparametric, and whole-body MRI techniques. Additionally, research on ultrasmall superparamagnetic particles of iron oxide contrast-enhanced MRI has revealed promising results for nodal staging of PCa. In this manuscript we aim to outline the current status and recent advancements of CT and MRI in PCa imaging.