RESUMEN
BACKGROUND: Multi-parametric magnetic resonance imaging (mp-MRI) is emerging as a useful tool for prostate cancer (PCa) detection but currently has unaddressed limitations. Computer aided diagnosis (CAD) systems have been developed to address these needs, but many approaches used to generate and validate the models have inherent biases. METHOD: All clinically significant PCa on histology was mapped to mp-MRI using a previously validated registration algorithm. Shape and size matched non-PCa regions were selected using a proposed sampling algorithm to eliminate biases towards shape and size. Further analysis was performed to assess biases regarding inter-zonal variability. RESULTS: A 5-feature Naïve-Bayes classifier produced an area under the receiver operating characteristic curve (AUC) of 0.80 validated using leave-one-patient-out cross-validation. As mean inter-class area mismatch increased, median AUC trended towards positively biasing classifiers to producing higher AUCs. Classifiers were invariant to differences in shape between PCa and non-PCa lesions (AUC: 0.82 vs 0.82). Performance for models trained and tested only in the peripheral zone was found to be lower than in the central gland (AUC: 0.75 vs 0.95). CONCLUSION: We developed a radiomics based machine learning system to classify PCa vs non-PCa tissue on mp-MRI validated on accurately co-registered mid-gland histology with a measured target registration error. Potential biases involved in model development were interrogated to provide considerations for future work in this area.
RESUMEN
PURPOSE: Our purpose was to evaluate intra-prostatic cancer volumes for salvage radiotherapy in men with recurrent prostate cancer confined to the prostate post-primary radiotherapy using mpMRI and 18F-DCFPyL PET/CT (PET). METHODS: Men with biochemical failure post-primary radiotherapy were enrolled in a multi-centre trial investigating mpMRI and PET. All men with isolated intra-prostatic recurrence are included in this secondary analysis. The intra-prostatic gross tumour volume (GTV) was manually delineated on mpMRI and was also delineated on PET using three methods: 1. manually, 2. using a 30% threshold of maximum intra-prostatic standard uptake value (SUVmax), and 3. using a 67% threshold of this SUVmax. Clinical target volumes (CTV) including expansions on each GTV were generated. Conformity indices were performed between the mpMRI CTV and each PET CTV. Correlation with biopsy and clinical outcomes were performed. RESULTS: Of the 36 men included, 30 (83%) had disease in two quadrants or less using the combination of mpMRI and PET. Mean target volume (union of CTV on mpMRI and CTV manually delineated on PET) was 12.2 cc (49% of prostate gland volume). 12/36 (33%) men had a biopsy. Per-patient sensitivity was 91% for mpMRI and 82% for PET. CONCLUSIONS: mpMRI and PET provide complementary information for delineation of intra-prostatic recurrent disease. Union of CTV on mpMRI and PET is often less than 50% of the prostate, suggesting this imaging could help define a target for focal salvage therapy.
RESUMEN
PURPOSE: Localized 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. METHODS: Multi-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. CONCLUSIONS: We 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 NUMBER: NCT04009174 (ClinicalTrials.gov).
RESUMEN
BACKGROUND AND PURPOSE: Prostate specific membrane antigen positron emission tomography imaging (PSMA-PET) has demonstrated potential for intra-prostatic lesion localization. We leveraged our existing database of co-registered PSMA-PET imaging with cross sectional digitized pathology to model dose coverage of histologically-defined prostate cancer when tailoring brachytherapy dose escalation based on PSMA-PET imaging. MATERIALS AND METHODS: Using a previously-developed automated approach, we created segmentation volumes delineating underlying dominant intraprostatic lesions for ten men with co-registered pathology-imaging datasets. To simulate realistic high-dose-rate brachytherapy (HDR-BT) treatments, we registered the PSMA-PET-defined segmentation volumes and underlying cancer to 3D trans-rectal ultrasound images of HDR-BT cases where 15 Gray (Gy) was delivered. We applied dose/volume optimization to focally target the dominant intraprostatic lesion identified on PSMA-PET. We then compared histopathology dose for all high-grade cancer within whole-gland treatment plans versus PSMA-PET-targeted plans. Histopathology dose was analyzed for all clinically significant cancer with a Gleason score of 7or greater. RESULTS: The standard whole-gland plans achieved a median [interquartile range] D98 of 15.2 [13.8-16.4] Gy to the histologically-defined cancer, while the targeted plans achieved a significantly higher D98 of 16.5 [15.0-19.0] Gy (p = 0.007). CONCLUSION: This study is the first to use digital histology to confirm the effectiveness of PSMA-PET HDR-BT dose escalation using automatically generated contours. Based on the findings of this study, PSMA-PET lesion dose escalation can lead to increased dose to the ground truth histologically defined cancer.
RESUMEN
BACKGROUND: PSMA-PET1 has shown good concordance with histology, but there is a need to investigate the ability of PSMA-PET to delineate DIL2 boundaries for guided biopsy and focal therapy planning. OBJECTIVE: To determine threshold and margin combinations that satisfy the following criteria: ≥95% sensitivity with max specificity and ≥95% specificity with max sensitivity. DESIGN, SETTING AND PARTICIPANTS: We registered pathologist-annotated whole-mount mid-gland prostatectomy histology sections cut in 4.4 mm intervals from 12 patients to pre-surgical PSMA-PET/MRI by mapping histology to ex-vivo imaging to in-vivo imaging. We generated PET-derived tumor volumes using boundaries defined by thresholded PET volumes from 1-100% of SUV3max in 1% intervals. At each interval, we applied margins of 0-30 voxels in one voxel increments, giving 3000 volumes/patient. OUTCOME MEASUREMENTS: Mean and standard deviation of sensitivity and specificity for cancer detection within the 2D oblique histologic planes that intersected with the 3D PET volume for each patient. RESULTS AND LIMITATIONS: A threshold of 67% SUV max with an 8.4 mm margin achieved a (mean ± std.) sensitivity of 95.0 ± 7.8% and specificity of 76.4 ± 14.7%. A threshold of 81% SUV max with a 5.1 mm margin achieved sensitivity of 65.1 ± 28.4% and specificity of 95.1 ± 5.2%. CONCLUSIONS: Preliminary evidence of thresholding and margin expansion of PSMA-PET images targeted at DILs validated with histopathology demonstrated excellent mean sensitivity and specificity in the setting of focal therapy/boosting and guided biopsy. These parameters can be used in a larger validation study supporting clinical translation.