scholarly journals Machine learning for differentiating metastatic and completely responded sclerotic bone lesion in prostate cancer: a retrospective radiomics study

2019 ◽  
Vol 92 (1101) ◽  
pp. 20190286 ◽  
Author(s):  
Emine Acar ◽  
Asım Leblebici ◽  
Berat Ender Ellidokuz ◽  
Yasemin Başbınar ◽  
Gamze Çapa Kaya
Keyword(s):  
Prostate Cancer ◽  
Machine Learning ◽  
Texture Analysis ◽  
Area Under The Curve ◽  
Ct Images ◽  
Metastatic Lesions ◽  
Psma Pet ◽  
Pet Ct ◽  
Sclerotic Lesions ◽  
Psma Expression

Objective:Using CT texture analysis and machine learning methods, this study aims to distinguish the lesions imaged via 68Ga-prostate-specific membrane antigen (PSMA) positron emission tomography (PET)/CT as metastatic and completely responded in patients with known bone metastasis and who were previously treated.Methods:We retrospectively reviewed the 68Ga-PSMA PET/CT images of 75 patients after treatment, who were previously diagnosed with prostate cancer and had known bone metastasis. A texture analysis was performed on the metastatic lesions showing PSMA expression and completely responded sclerotic lesions without PSMA expression through CT images. Textural features were compared in two groups. Thus, the distinction of metastasis/completely responded lesions and the most effective parameters in this issue were determined by using various methods [decision tree, discriminant analysis, support vector machine (SVM), k-nearest neighbor (KNN), ensemble classifier] in machine learning.Results:In 28 of the 35 texture analysis findings, there was a statistically significant difference between the two groups. The Weighted KNN method had the highest accuracy and area under the curve, has been chosen as the best model. The weighted KNN algorithm was succeeded to differentiate sclerotic lesion from metastasis or completely responded lesions with 0.76 area under the curve. GLZLM_SZHGE and histogram-based kurtosis were found to be the most important parameters in differentiating metastatic and completely responded sclerotic lesions.Conclusions:Metastatic lesions and completely responded sclerosis areas in CT images, as determined by 68Ga-PSMA PET, could be distinguished with good accuracy using texture analysis and machine learning (Weighted KNN algorithm) in prostate cancer.Advances in knowledge:Our findings suggest that, with the use of newly emerging software, CT imaging can contribute to identifying the metastatic lesions in prostate cancer.

Comparison of Bone Uptake in Bone Scan and Ga-68 PSMA PET/CT Images in Patients with Prostate Cancer

2019 ◽  
Vol 15 (6) ◽  
pp. 589-594
Author(s):  
Emine Acar ◽  
Recep Bekiş ◽  
Berna Polack
Keyword(s):  
Prostate Cancer ◽  
Bone Metastases ◽  
Bone Scintigraphy ◽  
Bone Scan ◽  
Ct Images ◽  
Additional Contribution ◽  
Psma Pet ◽  
Pet Ct ◽  
Increased Activity ◽  
Psma Expression

Objective: The aim of this study was to compare images from Tc-99m MDP bone scan (BS) and Ga-68 PSMA PET/CT of patients with prostate cancer in terms of bone metastases. Methods: Overall, 34 patients exhibited a mean age of 66 ± 9.5 (50-88) years, mean PSA of 51 ± 159ng/ml (0-912), and mean Gleason score of 8 (6-9). BS and Ga-68 PSMA PET/CT were applied to 34 patients within 30 days, and the results were evaluated, retrospectively. In both tests, radiopharmaceutical uptake in bones were compared. Results: In 7 patients (20.5%), uptake was not significant on BS and Ga-68 PSMA PET / CT images, which is related to metastasis. In 14 (41%) patients, bone metastases were observed in both examinations. However, more metastatic lesions were observed in the Ga-68 PSMA PET/CT of 3 patients and in the bone scintigraphy of 2 patients. PSMA expression was not observed on Ga-68 PSMA PET / CT in 13 (38%) patients with increased activity in bone scintigraphy. Two (6%) of these patients were thought to be metastatic, 2 (6%) were suspicious for metastasis, and 9 (26%) had no metastasis. When a lesion-based evaluation was performed, a total of 480 activities were evaluated: increased activity uptake was found in 305 BS, and 427 PSMA expression activity was detected. Furthermore, 435 of these activities were evaluated as metastatic. Conclusion: Ga-68 PSMA PET/CT provides an additional contribution to the BS evaluation of activity areas because of the presence of PSMA expression and anatomical lesions. In 6% of the patients, activity on BS and metastatic appearance in CT images were observed and the presence of lesions in the absence of PSMA was determined. This suggests that bone metastases without PSMA expression may also be present.

Characterization of PSMA and 18F-fluciclovine transporter gene expression in localized prostate cancer.

2020 ◽  
Vol 38 (6_suppl) ◽  
pp. 295-295
Author(s):  
Carissa Chu ◽  
Mohammed Alshalalfa ◽  
Martin Sjöström ◽  
Shuang Zhao ◽  
Annika Herlemann ◽  
...  
Keyword(s):  
Gene Expression ◽  
Prostate Cancer ◽  
Clinical Outcomes ◽  
Molecular Subtypes ◽  
Localized Prostate Cancer ◽  
Transporter Gene ◽  
Psma Pet ◽  
Pet Ct ◽  
Genomic Risk ◽  
Psma Expression

295 Background: While 18F-fluciclovine PET/CT is approved in the US and recommended by the NCCN, prostate-specific membrane antigen (PSMA) PET/CT is more common in Europe/Australia and recommended by the EAU. Less is known about the biology of lesions detected by either modality. 18F-fluciclovine PET relies on radiotracer uptake by amino acid transporters LAT1-4 and ASCT1-2. PSMA PET is dependent on surface expression of PSMA. We compared relative expression of PSMA and fluciclovine transporter genes in radical prostatectomy (RP) samples to determine their distribution across subtypes and correlation with outcomes. Methods: Gene expression data of 19,102 RP samples were analyzed using the Affymetrix Human Exon 1.0 ST microarray. 1,135 patients had long term follow up. Associations between expression of PSMA and fluciclovine transporter genes (LAT1-4 and ASCT1-2) and pathologic variables, molecular subtypes, and clinical outcomes were conducted. Results: All fluciclovine transporter genes (LAT 1-4, ASCT1-2) were expressed at lower levels than PSMA (p <0.0001). PSMA expression was positively correlated with genomic risk score and pathologic Gleason score (p<0.0001), but LAT2-3 and ASCT2 were inversely correlated with genomic risk in primary tumors (p<0.0001) and less expressed in GS 9-10 tumors (p<0.0001). PSMA expression was associated with worse metastasis-free survival (MFS) (HR 1.45, p=0.001) and lymph node involvement (HR 2.14, p<0.0001). Expression of LAT2, LAT3, ASCT2 expression was associated with better MFS (HR 0.85, 0.63, 0.74, p<0.0001-0.04). After multivariable adjustment, PSMA expression remained independently prognostic of poorer MFS (HR 1.3, p=0.028). Luminal B subtype was notable for PSMA overexpression; Luminal A was enriched in ASCT2 and LAT2 (p<0.0001). PSMA expression did not correlate with ERG fusion prostate cancers, but LAT2, ASCT1, and ASCT2 were overexpressed in ERG fusion negative tumors (p<0.0001). Conclusions: PSMA expression is associated with more aggressive disease and poorer clinical outcomes than fluciclovine transporter genes in localized prostate cancer. Molecular subtypes of prostate cancer vary in PSMA and fluciclovine transporter gene expression.

scholarly journals Immunohistochemical Validation of PSMA Expression Measured by 68Ga-PSMA PET/CT in Primary Prostate Cancer

2017 ◽  
Vol 59 (2) ◽  
pp. 238-243 ◽  
Author(s):  
Nadine Woythal ◽  
Ruza Arsenic ◽  
Carsten Kempkensteffen ◽  
Kurt Miller ◽  
Jan-Carlo Janssen ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Primary Prostate Cancer ◽  
Psma Pet ◽  
Pet Ct ◽  
Psma Expression

Early 18F-FDHT PET/CT as a predictor of treatment response in mCRPC treated with enzalutamide.

2019 ◽  
Vol 37 (7_suppl) ◽  
pp. 232-232 ◽  
Author(s):  
Hilde Hoving ◽  
Selma Palthe ◽  
Marleen Vallinga ◽  
Rutger Dost ◽  
Jourik A. Gietema ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Lymph Node ◽  
Treatment Response ◽  
Area Under The Curve ◽  
Standard Uptake Value ◽  
Castration Resistant Prostate Cancer ◽  
Metastatic Lesions ◽  
Pet Ct ◽  
Predict Treatment Response

232 Background: Androgen deprivation is the mainstay in the treatment of metastatic prostate cancer. During treatment, the majority of patients will develop progressive disease despite castrate levels of testosterone; castration-resistant prostate cancer (CRPC). In vivo determination of androgen receptor status by 18F-FDHT PET/CT could be of use to predict treatment response timely and objectively. The objective of this study is to assess the value of early 18F-FDHT PET/CT to predict treatment response of enzalutamide in mCRPC. Methods: This pilot study was performed in 18 chemotherapy naïve men with mCRPC. 18F-FDHT PET/CT was performed at baseline and after 4 weeks of treatment with enzalutamide. Standard Uptake Value (SUV)max and SUVpeak of the 5 most intense and/or all bone, pleura and lymph node metastases were determined per patient. Area under the curve (AUC), sensitivity (Se) and specificity (Sp) of different characteristics of 18F-FDHT PET/CT were performed by ROC analysis. Response was determined at 12 weeks of treatment according to PCCTWG. Results: A total of 477 lesions (411 bone, 3 pleura and 63 lymph node) were found. At 12 weeks, response was seen in 16 patients, whereas 2 patients showed no response. The characteristics of 18F-FDHT PET/CT are shown in table 1. Baseline median SUVpeak of all metastatic lesions showed an AUC of 0.79 to predict response. AUC values using the 5 most intense lesions only or using the delta between baseline and 5 weeks were less accurate. Clinical trial information: NTR4086. Conclusions: Baseline 18F-FDHT PET/CT using SUVpeak of all metastatic lesions predicts treatment response in mCRPC treated with enzalutamide with an AUC of 0.79.[Table: see text]

Differential expression of PSMA and 18F-fluciclovine transporter genes in metastatic castrate-resistant and treatment-emergent small cell/neuroendocrine prostate cancer.

2020 ◽  
Vol 38 (6_suppl) ◽  
pp. 24-24
Author(s):  
Carissa Chu ◽  
Mohammed Alshalalfa ◽  
Martin Sjöström ◽  
Shuang Zhao ◽  
Annika Herlemann ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Expression Profiles ◽  
Small Cell ◽  
Imaging Modalities ◽  
Primary Tumors ◽  
Neuroendocrine Prostate Cancer ◽  
Psma Pet ◽  
Pet Ct ◽  
Castrate Resistant ◽  
Psma Expression

24 Background: 18F-fluciclovine (Axumin) PET/CT imaging is recommended by the NCCN in the setting of biochemical recurrence, while prostate-specific membrane antigen (PSMA) PET/CT is preferred by the EAU. The utility of these methods in the post-androgen deprivation therapy (ADT) setting however, is less defined. Our objective was to compare relative gene expression of the molecular targets of these imaging modalities— fluciclovine transporter genes (LAT1-4, ASCT1-2) and PSMA—in metastatic castrate resistant prostate cancer (mCRPC) and treatment-emergent small cell/neuroendocrine prostate cancer (t-SCNC). Methods: Genome-wide expression profiles of five mCRPC cohorts (Aggarwal, Grasso, Kumar, Beltran, Robinson, et al) were used to characterize relative expression of fluciclovine transporter (LAT1-4, ASC1-2) and PSMA (FOLH1) genes. 3 cohorts (Kumar, Beltran, Aggarwal) were enriched with t-SCNC tumors. The GSE35988 cohort included primary tumors and mCRPC. RNA expression profiling methods were consistent within cohorts. Results: 518 mCRPC specimens were included. In the GSE35988 cohort, PSMA expression was downregulated in mCRPC when compared to primary localized tumors (p=0.01). PSMA expression was further depressed in t-SCNC when compared with mCRPC (p<0.001). Of the fluciclovine transporter genes, LAT1 and LAT4 were overexpressed in mCRPC when compared to primary tumors, while ASC2 was less expressed (p<0.001). LAT1 was further overexpressed in t-SCNC when compared to mCRPC, while LAT2 was less expressed (p<0.001). PSMA expression was negatively correlated with LAT1 (p<0.001) but positively correlated with LAT2 (p=0.006). Other fluciclovine transporters were not correlated. Conclusions: Expression of PSMA and a subset of fluciclovine transporter genes are inversely correlated in mCRPC and t-SCNC. These findings suggest that fluciclovine-based imaging may play a role in castrate resistant states. Clinical comparison between PSMA- and fluciclovine-based imaging modalities in mCRPC and t-SCNC is warranted.

scholarly journals Non-Prostate Cancer Tumours: Incidence on 18f-Dcfpyl Psma Pet/Ct and Psma Expression Characteristics in 1445 Patients.

2021 ◽  
Author(s):  
Elisa Perry ◽  
Arpit Talwar ◽  
Sanjana Sharma ◽  
Daisy O'Connor ◽  
Lih-Ming Wong ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Second Primary ◽  
Public And Private ◽  
Histopathological Correlation ◽  
Positron Emission ◽  
Psma Pet ◽  
Pet Ct ◽  
Second Primary Malignancies ◽  
Clinical Dilemma ◽  
Psma Expression

Abstract PurposeProstate cancer (PCa) imaging has been revolutionized by Positron emission tomography (PET) tracers targeted to prostate specific membrane antigen (PSMA). Identification and characterization of non-PCa tumors has become an increasing clinical dilemma with growing use. The primary aim of this retrospective multicenter analysis was to determine atypical PSMA expression in PCa and expression in non-PCa tumors to aid providing clinically relevant reports and guiding multidisciplinary discussion.MethodsRetrospective multicenter study examining 1445 consecutive 18F-DCFPyL PSMA PET/CT between 2016-2020. Referrals were from public and private, secondary and tertiary referral centres serving New Zealand and Melbourne. Repeat studies were excluded. Lesions atypical for PCa were categorized into four groups: 1. Atypical PCa metastases 2. Non-PCa tumors 3. Benign and 4. Indeterminate.Results67 patients had lesions atypical for PCa metastases; 11.9% atypical prostate cancer metastases, 25.4% non-PCa tumors, 40.3% indeterminate, 10.4% benign. With the exception of Renal Cell Carcinoma (RCC), the non-PCa tumors, indeterminate and benign lesions demonstrated low PSMA expression. Most atypical PCa metastases demonstrated significant expression. Limitations included retrospective design and lack of histopathological correlation/follow up in some.Conclusions: Non-PCa tumor detection is low. Lesions demonstrating significant PSMA expression were almost exclusively PCa metastases. Differentiation of atypical PCa metastases from second primary malignancies is vital to avoid unnecessary investigation, delayed therapy and additional costs.

scholarly journals Evaluating a Machine Learning Tool for the Classification of Pathological Uptake in Whole-Body PSMA-PET-CT Scans

Tomography ◽  
2021 ◽  
Vol 7 (3) ◽  
pp. 301-312
Author(s):  
Annette Erle ◽  
Sobhan Moazemi ◽  
Susanne Lütje ◽  
Markus Essler ◽  
Thomas Schultz ◽  
...  
Keyword(s):  
Machine Learning ◽  
Area Under The Curve ◽  
Ground Truth ◽  
Tracer Uptake ◽  
Whole Body ◽  
Clinical Environment ◽  
Physiological Uptake ◽  
Psma Pet ◽  
Pet Ct ◽  
Unseen Data

The importance of machine learning (ML) in the clinical environment increases constantly. Differentiation of pathological from physiological tracer-uptake in positron emission tomography/computed tomography (PET/CT) images is considered time-consuming and attention intensive, hence crucial for diagnosis and treatment planning. This study aimed at comparing and validating supervised ML algorithms to classify pathological uptake in prostate cancer (PC) patients based on prostate-specific membrane antigen (PSMA)-PET/CT. Retrospective analysis of 68Ga-PSMA-PET/CTs of 72 PC patients resulted in a total of 77 radiomics features from 2452 manually delineated hotspots for training and labeled pathological (1629) or physiological (823) as ground truth (GT). As the held-out test dataset, 331 hotspots (path.:128, phys.: 203) were delineated in 15 other patients. Three ML classifiers were trained and ranked to assess classification performance. As a result, a high overall average performance (area under the curve (AUC) of 0.98) was achieved, especially to detect pathological uptake (0.97 mean sensitivity). However, there is still room for improvement to detect physiological uptake (0.82 mean specificity), especially for glands. The ML algorithm applied to manually delineated lesions predicts hotspot labels with high accuracy on unseen data and may be an important tool to assist in clinical diagnosis.

scholarly journals PSMA radioligand therapy for solid tumors other than prostate cancer: background, opportunities, challenges, and first clinical reports

2021 ◽  
Author(s):  
M. J. M. Uijen ◽  
Y. H. W. Derks ◽  
R. I. J. Merkx ◽  
M. G. M. Schilham ◽  
J. Roosen ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Cancer Patients ◽  
Current Knowledge ◽  
Salivary Gland Cancer ◽  
Clinical Implementation ◽  
Radioligand Therapy ◽  
Psma Pet ◽  
Pet Ct ◽  
Specific Membrane ◽  
Psma Expression

AbstractIn the past decade, a growing body of literature has reported promising results for prostate-specific membrane antigen (PSMA)-targeted radionuclide imaging and therapy in prostate cancer. First clinical studies evaluating the efficacy of [177Lu]Lu-PSMA radioligand therapy (PSMA-RLT) demonstrated favorable results in prostate cancer patients. [177Lu]Lu-PSMA is generally well tolerated due to its limited side effects. While PSMA is highly overexpressed in prostate cancer cells, varying degrees of PSMA expression have been reported in other malignancies as well, particularly in the tumor-associated neovasculature. Hence, it is anticipated that PSMA-RLT could be explored for other solid cancers. Here, we describe the current knowledge of PSMA expression in other solid cancers and define a perspective towards broader clinical implementation of PSMA-RLT. This review focuses specifically on salivary gland cancer, glioblastoma, thyroid cancer, renal cell carcinoma, hepatocellular carcinoma, lung cancer, and breast cancer. An overview of the (pre)clinical data on PSMA immunohistochemistry and PSMA PET/CT imaging is provided and summarized. Furthermore, the first clinical reports of non-prostate cancer patients treated with PSMA-RLT are described.

Sign in / Sign up

Export Citation Format

Share Document