Dataset Information

Machine learning-based analysis of [¹⁸F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.

ABSTRACT:

Purpose

Quantitative prostate-specific membrane antigen (PSMA) PET analysis may provide for non-invasive and objective risk stratification of primary prostate cancer (PCa) patients. We determined the ability of machine learning-based analysis of quantitative [¹⁸F]DCFPyL PET metrics to predict metastatic disease or high-risk pathological tumor features.

Methods

In a prospective cohort study, 76 patients with intermediate- to high-risk PCa scheduled for robot-assisted radical prostatectomy with extended pelvic lymph node dissection underwent pre-operative [¹⁸F]DCFPyL PET-CT. Primary tumors were delineated using 50-70% peak isocontour thresholds on images with and without partial-volume correction (PVC). Four hundred and eighty standardized radiomic features were extracted per tumor. Random forest models were trained to predict lymph node involvement (LNI), presence of any metastasis, Gleason score ??8, and presence of extracapsular extension (ECE). For comparison, models were also trained using standard PET features (SUVs, volume, total PSMA uptake). Model performance was validated using 50 times repeated 5-fold cross-validation yielding the mean receiver-operator characteristic curve AUC.

Results

The radiomics-based machine learning models predicted LNI (AUC 0.86?±?0.15, p ConclusionMachine learning-based analysis of quantitative [¹⁸F]DCFPyL PET metrics can predict LNI and high-risk pathological tumor features in primary PCa patients. These findings indicate that PSMA expression detected on PET is related to both primary tumor histopathology and metastatic tendency. Multicenter external validation is needed to determine the benefits of using radiomics versus standard PET metrics in clinical practice.

SUBMITTER: Cysouw MCF

PROVIDER: S-EPMC7835295 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning-based analysis of [<sup>18</sup>F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.

Cysouw Matthijs C F MCF Jansen Bernard H E BHE van de Brug Tim T Oprea-Lager Daniela E DE Pfaehler Elisabeth E de Vries Bart M BM van Moorselaar Reindert J A RJA Hoekstra Otto S OS Vis André N AN Boellaard Ronald R

European journal of nuclear medicine and molecular imaging 20200731 2

<h4>Purpose</h4>Quantitative prostate-specific membrane antigen (PSMA) PET analysis may provide for non-invasive and objective risk stratification of primary prostate cancer (PCa) patients. We determined the ability of machine learning-based analysis of quantitative [<sup>18</sup>F]DCFPyL PET metrics to predict metastatic disease or high-risk pathological tumor features.<h4>Methods</h4>In a prospective cohort study, 76 patients with intermediate- to high-risk PCa scheduled for robot-assisted rad ...[more]

PMID: 32737518

Similar Datasets

Project description:Quantitative evaluation of radiolabeled prostate-specific membrane antigen (PSMA) PET scans may be used to monitor treatment response in patients with prostate cancer (PCa). To interpret longitudinal differences in PSMA uptake, the intrinsic variability of tracer uptake in PCa lesions needs to be defined. The aim of this study was to investigate the repeatability of quantitative PET/CT measurements using 18F-DCFPyL ([2-(3-(1-carboxy-5-[(6-18F-fluoro-pyridine-3-carbonyl)-amino]-pentyl)-ureido)-pentanedioic acid], a second-generation 18F-PSMA-ligand) in patients with PCa. Methods: Twelve patients with metastatic PCa were prospectively included, of whom 2 were excluded from final analyses. Patients received 2 whole-body 18F-DCFPyL PET/CT scans (median dose, 317 MBq; uptake time, 120 min) within a median of 4 d (range, 1-11 d). After semiautomatic (isocontour-based) tumor delineation, the following lesion-based metrics were derived: mean, peak, and maximum tumor-to-blood ratio; SUVmean, SUVpeak, and SUVmax normalized to body weight; tumor volume; and total lesion uptake (TLU). Additionally, patient-based total tumor volume (TTV) (sum of PSMA-positive tumor volumes) and total tumor burden (TTB) (sum of all lesion TLUs) were derived. Repeatability was analyzed using repeatability coefficients (RC) and intraclass correlation coefficients. Additionally, the effect of point-spread function (PSF) image reconstruction on the repeatability of uptake metrics was evaluated. Results: In total, 36 18F-DCFPyL PET-positive lesions were analyzed (≤5 lesions per patient). The RCs for mean, peak, and maximum tumor-to-blood ratio were 31.8%, 31.7%, and 37.3%, respectively. For SUVmean, SUVpeak, and SUVmax, the RCs were 24.4%, 25.3%, and 31.0%, respectively. All intraclass correlation coefficients were at least 0.97. Tumor volume delineations were quite repeatable, with an RC of 28.1% for individual lesion volumes and 17.0% for TTV. TTB had an RC of 23.2% and 33.4% when based on SUVmean and mean tumor-to-blood ratio, respectively. Small lesions (<4.2 cm3) had worse repeatability for volume measurements. The repeatability of SUVpeak, TLU, and all patient-level metrics was not affected by PSF reconstruction. Conclusion: 18F-DCFPyL uptake measurements are quite repeatable and can be used for clinical validation in future treatment response assessment studies. Patient-based TTV may be preferred for multicenter studies because its repeatability was both high and robust to different image reconstructions.

Project description:PurposeIdentification of the dominant intraprostatic lesion(s) (DILs) can facilitate diagnosis and treatment by targeting biologically significant intra-prostatic foci. A PSMA ligand, [18F]DCFPyL (2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine-3-carbonyl)-amino]-pentyl}-ureido)-pentanedioic acid), is better than choline-based [18F]FCH (fluorocholine) in detecting and localizing DIL because of higher tumour contrast, particularly when imaging is delayed to 1 h post-injection. The goal of this study was to investigate whether the different imaging performance of [18F]FCH and [18F]DCFPyL can be explained by their kinetic behaviour in prostate cancer (PCa) and to evaluate whether DIL can be accurately detected and localized using a short duration dynamic positron emission tomography (PET).Methods19 and 23 PCa patients were evaluated with dynamic [18F]DCFPyL and [18F]FCH PET, respectively. The dynamic imaging protocol with each tracer had a total imaging time of 22 min and consisted of multiple frames with acquisition times from 10 to 180 s. Tumour and benign tissue regions identified by sextant biopsy were compared using standardized uptake value (SUV) and tracer kinetic parameters from kinetic analysis of time-activity curves.ResultsFor [18F]DCFPyL, logistic regression identified Ki and k4 as the optimal model to discriminate tumour from benign tissue (84.2% sensitivity and 94.7% specificity), while only SUV was predictive for [18F]FCH (82.6% sensitivity and 87.0% specificity). The higher k3 (binding) of [18F]FCH than [18F]DCFPyL explains why [18F]FCH SUV can differentiate tumour from benign tissue within minutes of injection. Superior [18F]DCFPyL tumour contrast was due to the higher k4/k3 (more rapid washout) in benign tissue compared to tumour tissue.ConclusionsDIL was detected with good sensitivity and specificity using 22-min dynamic [18F]DCFPyL PET and avoids the need for delayed post-injection imaging timepoints. The dissimilar in vivo kinetic behaviour of [18F]DCFPyL and [18F]FCH could explain their different SUV images. Clinical Trial Registration NCT04009174 (ClinicalTrials.gov).

Project description:BackgroundProstate Specific Membrane Antigen (PSMA) - positron emission tomography (PET) guides metastasis-directed radiotherapy (MDRT) in prostate cancer (PrCa). However, its value as a treatment response assessment tool after MDRT remains unclear. Importantly, there is limited understanding of the potential of radiotherapy (RT) to alter PSMA gene (folate hydrolase 1; FOLH1) expression.MethodologyWe reviewed a series of 11 men with oligo-metastatic PrCa (25 metastasis sites) treated with MDRT before re-staging with 18F-DCFPyL (PSMA) PET upon secondary recurrence. Acute effects of RT on PSMA protein and mRNA levels were examined with qPCR and immunoblotting in human wild-type androgen-sensitive (LNCap), castrate-resistant (22RV1) and castrate-resistant neuroendocrine (PC3 and DU145) PrCa cell lines. Xenograft tumors were analyzed with immunohistochemistry. Further, we examined PSMA expression in untreated and irradiated radio-resistant (RR) 22RV1 (22RV1-RR) and DU145 (DU145-RR) cells and xenografts selected for survival after high-dose RT.ResultsThe majority of MDRT-treated lesions showed lack of PSMA-PET/CT avidity, suggesting treatment response even after low biological effective dose (BED) MDRT. We observed similar high degree of heterogeneity of PSMA expression in both human specimens and in xenograft tumors. PSMA was highly expressed in LNCap and 22RV1 cells and tumors but not in the neuroendocrine PC3 and DU145 models. Single fraction RT caused detectable reduction in PSMA protein but not in mRNA levels in LNCap cells and did not significantly alter PSMA protein or mRNA levels in tissue culture or xenografts of the other cell lines. However, radio-resistant 22RV1-RR cells and tumors demonstrated marked decrease of PSMA transcript and protein expression over their parental counterparts.ConclusionsPSMA-PET may be a promising tool to assess RT response in oligo-metastatic PrCa. However, future systematic investigation of this concept should recognize the high degree of heterogeneity of PSMA expression within prostate tumors and the risk for loss of PSMA expression in tumor surviving curative courses of RT.

Project description:PurposeLocalized prostate cancer (PCa) in patients is characterized by a dominant focus in the gland (dominant intraprostatic lesion, DIL). Accurate DIL identification may enable more accurate diagnosis and therapy through more precise targeting of biopsy, radiotherapy and focal ablative therapies. The goal of this study is to validate the performance of [18F]DCFPyL PET and CT perfusion (CTP) for detecting and localizing DIL against digital histopathological images.MethodsMulti-modality image sets: in vivo T2-weighted (T2w)-MRI, 22-min dynamic [18F]DCFPyL PET/CT, CTP, and 2-h post-injection PET/MR were acquired in patients prior to radical prostatectomy. The explanted gland with implanted fiducial markers was imaged with T2w-MRI. All images were co-registered to the pathologist-annotated digital images of whole-mount mid-gland histology sections using fiducial markers and anatomical landmarks. Regions of interest encompassing DIL and non-DIL tissue were drawn on the digital histopathological images and superimposed on PET and CTP parametric maps. Logistic regression with backward elimination of parameters was used to select the most sensitive parameter set to distinguish DIL from non-DIL voxels. Leave-one-patient-out cross-validation was performed to determine diagnostic performance.Results[18F]DCFPyL PET and CTP parametric maps of 15 patients were analyzed. SUVLate and a model combining Ki and k4 of [18F]DCFPyL achieved the most accurate performance distinguishing DIL from non-DIL voxels. Both detection models achieved an AUC of 0.90 and an error rate of < 10%. Compared to digital histopathology, the detected DILs had a mean dice similarity coefficient of 0.8 for the Ki and k4 model and 0.7 for SUVLate.ConclusionsWe have validated using co-registered digital histopathological images that parameters from kinetic analysis of 22-min dynamic [18F]DCFPyL PET can accurately localize DILs in PCa for targeting of biopsy, radiotherapy, and focal ablative therapies. Short-duration dynamic [18F]DCFPyL PET was not inferior to SUVLate in this diagnostic task.Clinical trial registration numberNCT04009174 (ClinicalTrials.gov).

Project description:Background Prostate cancer recurrence is found in up to 40% of men with prior definitive (total prostatectomy or whole-prostate radiation) treatment. Prostate-specific membrane antigen PET agents such as 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine 3-carbonyl)-amino]-pentyl}-ureido)-pentanedioic acid (18F-DCFPyL) may improve detection of recurrence compared with multiparametric MRI; however, histopathologic validation is lacking. Purpose To determine the sensitivity, specificity, and positive predictive value (PPV) of 18F-DCFPyL PET/CT based on histologic analysis and to compare with pelvic multiparametric MRI in men with biochemically recurrent prostate cancer. Materials and Methods Men were prospectively recruited after prostatectomy and/or radiation therapy with rising prostate-specific antigen level (median, 2.27 ng/mL; range, 0.2-27.45 ng/mL) and a negative result at conventional imaging (bone scan and/or CT). Participants underwent 18F-DCFPyL PET/CT imaging and 3.0-T pelvic multiparametric MRI. Statistical analysis included Wald and modified χ2 tests. Results A total of 323 lesions were visualized in 77 men by using 18F-DCFPyL or multiparametric MRI, with imaging detection concordance of 25% (82 of 323) when including all lesions in the MRI field of view and 53% (52 of 99) when only assessing prostate bed lesions. 18F-DCFPyL depicted more pelvic lymph nodes than did MRI (128 vs 23 nodes). Histologic validation was obtained in 80 locations with sensitivity, specificity, and PPV of 69% (25 of 36; 95% confidence interval [CI]: 51%, 88%), 91% (40 of 44; 95% CI: 74%, 98%), and 86% (25 of 29; 95% CI: 73%, 97%) for 18F-DCFPyL and 69% (24 of 35; 95% CI: 50%, 86%), 74% (31 of 42; 95% CI: 42%, 89%), and 69% (24 of 35; 95% CI: 50%, 88%) for multiparametric MRI (P = .95, P = .14, and P = .07, respectively). In the prostate bed, sensitivity, specificity, and PPV were 57% (13 of 23; 95% CI: 32%, 81%), 86% (18 of 21; 95% CI: 73%, 100%), and 81% (13 of 16; 95% CI: 59%, 100%) for 18F-DCFPyL and 83% (19 of 23; 95% CI: 59%, 100%), 52% (11 of 21; 95% CI: 29%, 74%), and 66% (19 of 29; 95% CI: 44%, 86%) for multiparametric MRI (P = .19, P = .02, and P = .17, respectively). The addition of 18F-DCFPyL to multiparametric MRI improved PPV by 38% overall (P = .02) and by 30% (P = .09) in the prostate bed. Conclusion Findings with 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine 3-carbonyl)-amino]-pentyl}-ureido)-pentanedioic acid (18F-DCFPyL) were histologically validated and demonstrated high specificity and positive predictive value. In the pelvis, 18F-DCFPyL depicted more lymph nodes and improved positive predictive value and specificity when added to multiparametric MRI. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Zukotynski and Rowe in this issue.

Dataset Information

Machine learning-based analysis of [¹⁸F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.

Purpose

Methods

Results

Publications

Machine learning-based analysis of [<sup>18</sup>F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Dataset Information

Machine learning-based analysis of [18F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.

Purpose

Methods

Results

Publications

Machine learning-based analysis of [<sup>18</sup>F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Machine learning-based analysis of [¹⁸F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.