Prediction of Micrometastasis (&lt; 1 cm) to Pelvic Lymph Nodes in Prostate Cancer: Role of Preoperative MRI

Park, Sung Yoon; Oh, Young Taek; Jung, Dae Chul; Cho, Nam Hoon; Choi, Young Deuk; Rha, Koon Ho

doi:10.2214/ajr.14.14138

Cited by 28 publications

(10 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Amongst all factors, ADC and MRI‐LNI had the highest weight coefficient in the Lasso‐LR + model; MRI‐LNI and SVI had the highest weight coefficient in RFE‐SVM + model; and ADC ranked top in both the mean decrease accuracy and mean decrease Gini evaluation in the RFs + distribution. Similarly making full use of preoperative variables, our present study was in line with published studies . In the study of Brembilla et al.…”

Section: Discussionsupporting

confidence: 90%

A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection

Hou

Bao

et al. 2019

BJU International

View full text Add to dashboard Cite

ObjectivesTo develop a machine learning (ML)-assisted model to identify candidates for extended pelvic lymph node dissection (ePLND) in prostate cancer by integrating clinical, biopsy, and precisely defined magnetic resonance imaging (MRI) findings. Patients and MethodsIn all, 248 patients treated with radical prostatectomy and ePLND or PLND were included. ML-assisted models were developed from 18 integrated features using logistic regression (LR), support vector machine (SVM), and random forests (RFs). The models were compared to the Memorial SloanKettering Cancer Center (MSKCC) nomogram using receiver operating characteristic-derived area under the curve (AUC) calibration plots and decision curve analysis (DCA). ResultsA total of 59/248 (23.8%) lymph node invasions (LNIs) were identified at surgery. The predictive accuracy of the ML-based models, with (+) or without (À) MRI-reported LNI, yielded similar AUCs (RFs + /RFs À : 0.906/0.885; SVM + /SVM À : 0.891/ 0.868; LR + /LR À : 0.886/0.882) and were higher than the MSKCC nomogram (0.816; P < 0.001). The calibration of the MSKCC nomogram tended to underestimate LNI risk across the entire range of predicted probabilities compared to the ML-assisted models. The DCA showed that the ML-assisted models significantly improved risk prediction at a risk threshold of ≤80% compared to the MSKCC nomogram. If ePLNDs missed was controlled at <3%, both RFs + and RFs À resulted in a higher positive predictive value (51.4%/49.6% vs 40.3%), similar negative predictive value (97.2%/97.8% vs 97.2%), and higher number of ePLNDs spared (56.9%/54.4% vs 43.9%) compared to the MSKCC nomogram. ConclusionsOur ML-based model, with a 5-15% cutoff, is superior to the MSKCC nomogram, sparing ≥50% of ePLNDs with a risk of missing <3% of LNIs.Keywords pelvic lymph node invasion, logistic regression, support vector machine, random forests, machine learning, #ProstateCancer, #PCSM

show abstract

Section: Discussionsupporting

confidence: 90%

A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection

Hou

Bao

et al. 2019

BJU International

View full text Add to dashboard Cite

show abstract

“…The inter-and intraobserver reproducibility demonstrated a coefficient of variation of less than 5% (18). Any lymph node with an SAD above 8 mm was considered positive according to CT criteria (7,8). The size, represented by volume, SADs and LADs, and node location, was recorded for every node.…”

Section: Image Acquisition Psma Pet Analysis and Volumetric Ct Histmentioning

confidence: 99%

“…Treatment varies according to the number and site of recurrence, but localizing such recurrences has been problematic with existing imaging methods such as CT and bone scanning because of their lack of sensitivity and specificity. For instance, only 11%-14% of patients with biochemical failure after radical prostatectomy have positive CT scan findings (5,6), with similar results with MRI (6)(7)(8). Therefore, detecting and localizing sites of recurrence is a strong unmet need for personalizing salvage therapy within individual patients.…”

mentioning

confidence: 99%

⁶⁸Ga-PSMA PET/CT and Volumetric Morphology of PET-Positive Lymph Nodes Stratified by Tumor Differentiation of Prostate Cancer

et al. 2017

View full text Add to dashboard Cite

Ga-prostate-specific membrane antigen (PSMA) PET/CT is a new method to detect early nodal metastases in patients with biochemical relapse of prostate cancer. In this retrospective investigation, the dimensions, volume, localization, and SUV of nodes identified by Ga-PSMA were correlated to their Gleason score (GS) at diagnosis. All PET/CT images were acquired 60 ± 10 min after intravenous injection of Ga-PSMA (mean dose, 176 MBq). In 147 prostate cancer patients (mean age, 68 y; range, 44-87 y) with prostate-specific antigen relapse (mean prostate-specific antigen level, 5 ng/mL; range, 0.25-294 ng/mL), 362Ga-PSMA PET-positive lymph nodes (LNs) were identified. These patients were classified on the basis of their histopathology at primary diagnosis into either low- (GS ≤ 6, well differentiated), intermediate- (GS = 7, moderately differentiated), or high-GS cohorts (GS ≥ 8, poorly differentiated prostate cancer). Using semiautomated LN segmentation software (Fraunhofer MEVIS), we measured node volume and short-axis dimensions (SADs) and long-axis dimensions based on CT and compared with the SUV Nodes demonstrating uptake of Ga-PSMA with an SUV of 2.0 or more were considered PSMA-positive, and nodes with an SAD of 8 mm or more were considered positive by morphologic criteria. Mean SUV was 13.5 (95% confidence interval [CI], 10.9-16.1), 12.4 (95% CI, 9.9-14.9), and 17.8 (95% CI, 15.4-20.3) within the low-, intermediate-, and high-GS groups, respectively. The morphologic assessment of the Ga-PSMA-positive LN demonstrated that the low-GS cohort presented with smallerGa-PSMA-positive LNs (mean SAD, 7.7 mm; = 113), followed by intermediate- (mean SAD, 9.4 mm; = 122) and high-GS cohorts (mean SAD, 9.5 mm; = 127). On the basis of the CT morphology criteria, only 34% of low-GS patients, 56% of intermediate-GS patients, and 53% of high-GS patients were considered CT positive. Overall,Ga-PSMA imaging led to a reclassification of stage in 90 patients (61%) from cN0 to cN1 over CT. Ga-PSMA PET is a promising modality in biochemical recurrent prostate cancer patients for N staging. Conventional imaging underestimates LN involvement compared with PSMA molecular staging score in each GS cohort. The sensitivity ofGa-PSMA PET/CT enables earlier detection of subcentimeter LN metastases in the biochemical recurrence setting.

show abstract

“…Since direct methods of imaging LN metastases are limited, several nomograms have been developed that calculate LN metastatic risk based on clinical parameters and T staging that correlates to some degree with N staging. However, these methods cannot localize the sites of LN metastases but merely forecast the likelihood that one or more malignant nodes may be present in the case ( 2 - 4 ). More recently, functional and molecular imaging methods have been proposed as alternatives ( 1 ).…”

Section: Introductionmentioning

confidence: 99%

Ultra-small superparamagnetic iron oxide contrast agents for lymph node staging of high-risk prostate cancer

Czarniecki¹,

Pesapane²,

Wood³

et al. 2018

Transl. Androl. Urol.

View full text Add to dashboard Cite

Ultrasmall superparamagnetic particles of iron oxide (USPIOs) imaged with magnetic resonance imaging (MRI) have been proposed as an experimental method for visualizing lymph node (LN) metastases. The method does not require ionizing radiation, yet can detect small nodes that are involved with metastases. USPIOs are naturally taken up by macrophages that deposit in the normal LN creating a low signal region in normal areas; areas within the node that do not show this loss of signal are likely involved by tumor although there can be other causes (fibrosis or inflammation). However, the lack of approved USPIOs that are clinically available hinders adoption and larger studies. The proposed indications for USPIO MRI, including specific compounds and imaging methods are discussed.

show abstract

Prediction of Micrometastasis (< 1 cm) to Pelvic Lymph Nodes in Prostate Cancer: Role of Preoperative MRI

Abstract: Preoperative MRI staging may play a role in prediction of micrometastasis (< 1 cm) to pelvic lymph nodes in prostate cancer.

Cited by 28 publications

References 40 publications

A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection

A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection

⁶⁸Ga-PSMA PET/CT and Volumetric Morphology of PET-Positive Lymph Nodes Stratified by Tumor Differentiation of Prostate Cancer

Ultra-small superparamagnetic iron oxide contrast agents for lymph node staging of high-risk prostate cancer

Contact Info

Product

Resources

About

Prediction of Micrometastasis (< 1 cm) to Pelvic Lymph Nodes in Prostate Cancer: Role of Preoperative MRI

Abstract: Preoperative MRI staging may play a role in prediction of micrometastasis (< 1 cm) to pelvic lymph nodes in prostate cancer.

Cited by 28 publications

References 40 publications

A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection

A machine learning‐assisted decision‐support model to better identify patients with prostate cancer requiring an extended pelvic lymph node dissection

68Ga-PSMA PET/CT and Volumetric Morphology of PET-Positive Lymph Nodes Stratified by Tumor Differentiation of Prostate Cancer

Ultra-small superparamagnetic iron oxide contrast agents for lymph node staging of high-risk prostate cancer

Contact Info

Product

Resources

About

⁶⁸Ga-PSMA PET/CT and Volumetric Morphology of PET-Positive Lymph Nodes Stratified by Tumor Differentiation of Prostate Cancer