In a study of 601 patients with prostate cancer, researchers found that only one patient with extensive metastasis was falsely diagnosed as non-metastatic on whole-body, low-dose computed tomography (CT) by reviewing radiologists.
For the detection of osteo-metastasis in patients with prostate cancer, whole-body computed tomography (CT) demonstrated sensitivity ranging between 81 to 89.4 percent and specificity ranging between 96.6 to 98.5 percent, according to new research.
For the study, recently published in Insights into Imaging, researchers examined the use of whole-body, low-dose CT for the diagnosis of osteo-metastasis in 601 patients (mean age of 68.7) with prostate cancer. According to the study, 35.1 percent of the cohort received imaging for initial staging and 64.9 percent had imaging to assess the extent of the disease after biochemical recurrence (BCR) of prostate cancer.
For the initial staging group, low-dose CT had a per-patient sensitivity rate ranging between 76.8 and 88.4 percent with specificity ranging between 98.6 and 99.3 percent. For patients with BCR, the researchers found a sensitivity range between 83.3 to 90 percent and a specificity rate between 95.6 to 98.6 percent for the reviewing radiologists. Overall, low-dose CT demonstrated an 81 to 89.4 percent sensitivity rate and a 96.6 to 98.5 percent specificity rate for osteo-metastasis in the study cohort, according to the authors.
In an analysis of metastatic lesions with underlying CT findings, the researchers found that lesion features such as size > 1 cm, cortical destruction, the presence of a soft tissue component and ill-defined borders were statistically associated with metastasis. (Images courtesy of Insights into Imaging.)
While previous research has shown higher sensitivity and specificity for 68Ga-PSMA PET/CT in detecting osteo-metastasis (see ref 9 in study), the study authors said whole-body, low-dose CT may provide a viable alternative that is less expensive and is more widely available than 68Ga-PSMA PET/CT.
“The high accuracy at the per-patient level indicated that CT could correctly diagnose the presence or absence of metastasis in most patients. However, in clinical management, both in staging and BCR conditions, it is important to know the burden of disease. In this issue, CT still had reliable performance, meaning most patients would receive the proper therapy,” wrote lead study author Mohammadreza Chavoshi, M.D., who is affiliated with the Department of Radiology at Shariati Hospital and the Tehran University of Medical Sciences in Iran, and colleagues.
The study authors noted that the per-lesion analysis for benign lesions with whole-body low-dose CT had significantly lower sensitivity (48.5 to 63.3 percent) as well as lower specificity (82.1 to 92.1 percent). The researchers also found that CT did not detect 30.8 percent of metastatic lesions in patients without extensive disease.
However, Chavoshi and colleagues noted that current CHAARTED and LATITUDE criteria for prostate cancer do not emphasize precision in determining the number and location of osteo-metastatic lesions as the treatment would be the same for patients deemed to be at high risk. They also noted that whole-body, low-dose CT had only one false diagnosis of a patient with extensive metastasis.
“In the initial staging of PCa patients based on the above-mentioned criteria, whole-body low-dose CT showed promising findings in terms of specificity and (positive predictive value), regardless of readers’ expertise. This means that the treatment approach would have been confidently adopted in the patients diagnosed with the high-burden disease,” noted Chavoshi and colleagues.
MRI-Based AI Radiomics Model Offers 'Robust' Prediction of Perineural Invasion in Prostate Cancer
July 26th 2024A model that combines MRI-based deep learning radiomics and clinical factors demonstrated an 84.8 percent ROC AUC and a 92.6 percent precision-recall AUC for predicting perineural invasion in prostate cancer cases.
Study: PSMA PET/CT More Advantageous than MRI for Locoregional Staging of Prostate Cancer
July 8th 2024For men with prostate cancer, new research demonstrates that 18F-PSMA-1007 PET/CT is 17 percent more accurate than mpMRI at identifying the final pathological tumor stage and 12 percent more accurate in detecting extracapsular extension.
Could MRI-Based AI Offer Better Risk Stratification for Prostate Cancer than PI-RADS?
July 3rd 2024The combination of deep learning assessment of prostate MRI with PI-RADS classification at a 15 percent risk threshold spared 49 percent of biopsies in comparison to 37 percent employing a PI-RADS>4 only cut-ff approach, according to new research.