Detection of Local Recurrence with 3-Tesla MRI After Radical Prostatectomy: A Useful Method for Radiation Treatment Planning?

BACKGROUND/AIM: Salvage radiotherapy improves biochemical control in patients with recurrence of prostate cancer after prostatectomy. Radiotherapy target volumes of the prostatic fossa are based on empirical data and differ between different guidelines. Localization of recurrence with multiparametric magnetic resonance imaging (MRI) might be a feasible approach to localize recurrent lesions. PATIENTS AND METHODS: Twenty-one patients with biochemical recurrence after radical prostatectomy were included (median prostate-specific antigen (PSA) =0.17 ng/ml). Multi-parametric MRI was performed using a 3-T MR system. RESULTS: Lesions were detected in seven patients with a median PSA of 0.86 ng/ml (minimum= 0.31 ng/ml). Patients without detectable recurrence had a median PSA of 0.12 ng/ml. All patients with detectable lesions responded to radiotherapy. Eleven out of 14 patients without detectable recurrence also responded. Plasma flow in suspicious lesions was correlated with PSA level. CONCLUSION: Detection of recurrence at the prostatic fossa with our approach was possible in a minority of patients with a low PSA level. Clinical relevance of plasma flow in suspicious lesions should be further investigated.

Is There a Role for Functional MRI for the Assessment of Extracapsular Extension in Prostate Cancer?

BACKGROUND/AIM: Exctracapsular extension (ECE) in prostate cancer has a high impact on treatment decision. MRI might predict presence of ECE non-invasively. PATIENTS AND METHODS: Triplanar T2w-sequences, DWI (diffusion weighted imaging) and DCE (dynamic contrast-enhanced imaging) of 34 patients with PCa were analyzed to prior prostatectomy. Sensitivity (SS) and specificity (SP) of T2w, apparent diffusion coefficient (ADC), plasma flow (PF) and mean transit time (MTT) normalized by PCa/normal tissue ratio for prediction of CI (capsular infiltration)/ECE were determined by area-under-the-receiver-operating-characteristics analysis. RESULTS: SS/SP for detecting ECE was 29/85. AUC (area under the curve) of ECE cases was 0.98/0.92/0.69 (cut-off-ratios 3.2/0.51/0.46), SS 93/100/86% and SP 95/80/50% for PF-/MTT-/ADC-ratios, respectively. PF- and MTT-ratios between CI and without CI/ECE differed significantly (PF, p<0.0001; MTT, p=0.0134) with SS/SP 84/89% for PF and SS/SP 52/100% for MTT-ratios. No significant differences regarding ADC-ratios were identified. CONCLUSION: ECE/CI can be assessed by quantitative DCE analysis with great diagnostic confidence and higher specificity than ADC.

Small Field-of-view single-shot EPI-DWI of the prostate: Evaluation of spatially-tailored two-dimensional radiofrequency excitation pulses.

PURPOSE: Spatially-tailored (RF) excitation pulses in echo-planar imaging (EPI), combined with a decreased FOV in the phase-encoding direction, enable a reduction of k-space acquisition lines, which shortens the echo train length (ETL) and reduces susceptibility artifacts. The purpose of this study was to evaluate the image quality of a zoomed EPI (z-EPI) sequence in diffusion-weighted imaging (DWI) of the prostate in comparison to a conventional single-shot EPI using single-channel (c-EPI1) and multi-channel (c-EPI2) RF excitation, with and without use of an endorectal coil. MATERIALS AND METHODS: 33 consecutive patients (mean age: 61 +/- 9 years; mean PSA: 8.67±6.23 ng/ml) with examinations between 10/2012 and 02/2014 were analyzed in this retrospective study. In 26 of 33 patients the initial multiparametric (mp)-MRI was performed on a whole-body 3T scanner (Magnetom Trio, Siemens, Erlangen, Germany) using an endorectal coil (c (conventional)-EPI1). Zoomed-EPI (Z-EPI) examinations of these patients and a complete mp-MRI protocol including c-EPI2 of 7 additional patients were carried out on another 3T wb MR scanner with two-channel dynamic parallel transmit capability (Magnetom Skyra with TimTX TrueShape, Siemens). For z-EPI, the one-dimensional spatially selective RF excitation pulse was replaced by a two-dimensional RF pulse. Degree of image blur and susceptibility artifacts (0=not present to 3= non-diagnostic), maximum image distortion (mm), apparent diffusion coefficient (ADC) values, as well as overall scan preference were evaluated. SNR maps were generated to compare c-EPI2 and z-EPI. RESULTS: Overall image quality of z-EPI was preferred by both readers in all examinations with a single exception. Susceptibility artifacts were rated significantly lower on z-EPI compared to both other methods (z-EPI vs c-EPI1: p<0.01; z-EPI vs c-EPI2: p<0.01) as well as image blur (z-EPI vs c-EPI1: p<0.01; z-EPI vs c-EPI2: p<0.01). Image distortion was not statistically significantly reduced with z-EPI (z-EPI vs c-EPI1: p=0.12; z-EPI vs c-EPI2: p=0.42). Interobserver agreement for ratings of susceptibility artifacts, image blur and overall scan preference was good. SNR was higher for z-EPI than for c-EPI1 (n=1). CONCLUSION: Z-EPI leads to significant improvements in image quality and artifacts as well as image blur reduction improving prostate DWI and enabling accurate fusion with conventional sequences. The improved fusion could lead to advantages in the field of MRI-guided biopsy suspicous lesions and performance of locally ablative procedures for prostate cancer.

Incremental learning with SVM for multimodal classification of prostatic adenocarcinoma.

Robust detection of prostatic cancer is a challenge due to the multitude of variants and their representation in MR images. We propose a pattern recognition system with an incremental learning ensemble algorithm using support vector machines (SVM) tackling this problem employing multimodal MR images and a texture-based information strategy. The proposed system integrates anatomic, texture, and functional features. The data set was preprocessed using B-Spline interpolation, bias field correction and intensity standardization. First- and second-order angular independent statistical approaches and rotation invariant local phase quantization (RI-LPQ) were utilized to quantify texture information. An incremental learning ensemble SVM was implemented to suit working conditions in medical applications and to improve effectiveness and robustness of the system. The probability estimation of cancer structures was calculated using SVM and the corresponding optimization was carried out with a heuristic method together with a 3-fold cross-validation methodology. We achieved an average sensitivity of 0.844 ± 0.068 and a specificity of 0.780 ± 0.038, which yielded superior or similar performance to current state of the art using a total database of only 41 slices from twelve patients with histological confirmed information, including cancerous, unhealthy non-cancerous and healthy prostate tissue. Our results show the feasibility of an ensemble SVM being able to learn additional information from new data while preserving previously acquired knowledge and preventing unlearning. The use of texture descriptors provides more salient discriminative patterns than the functional information used. Furthermore, the system improves selection of information, efficiency and robustness of the classification. The generated probability map enables radiologists to have a lower variability in diagnosis, decrease false negative rates and reduce the time to recognize and delineate structures in the prostate.

Diagnostic accuracy of 18F choline PET/CT using time-of-flight reconstruction algorithm in prostate cancer patients with biochemical recurrence.

PURPOSE: Image quality (IQ) of PET in voluminous body regions can be limited, which impairs the assessment of small metastatic lesions. Time-of-flight (TOF) reconstruction algorithm may deliver an increase of spatial resolution. The purpose of this study was to evaluate the impact of TOF on IQ, lesion detection rate, lesion volume (V) and SUVmax in F choline PET/CT of prostate cancer patients with biochemical recurrence compared to standard PET/CT reconstruction (standard). PATIENTS AND MATERIALS: During a period of 9 months, 32 patients with prostate cancer (mean [SD] age, 71 [7.8] years) and biochemical recurrence were included in this prospective institutional review board-approved study. Each patient underwent a state-of-the-art 3-dimensional F choline PET/CT. A total of 76 lesions were assessed by 2 board-certified nuclear medicine physicians and a third-year resident. Lesion volume and SUVmax of local recurrence, lymph nodes, and organ metastases were compared between TOF and standard. Image quality and lesion demarcation were rated according to a 5-point Likert-type scale. Interobserver agreement was assessed. RESULTS: Eight additional lesions were detected using TOF (SUVmax, 3.64 [0.95]; V, 0.58 cm [0.50]). Image quality was reduced (IQ standard, 1.28; TOF, 1.77; P < 0.01) in calculated TOF images, although quality of lesion demarcation was improved (lesion demarcation: standard, 1.66; TOF, 1.26; P < 0.01). SUVmax was significantly increased in TOF images (SUVmax standard, 6.9 [4.1]; TOF, 8.1 [4.1]; P < 0.01), whereas V did not show significant differences (V standard, 5.3 [10.4] cm; TOF, 5.4 [10.3] cm; P = 0.41). Interobserver agreement was good for combined ratings (1 + 2 and 3 + 4). CONCLUSIONS: Application of TOF seems to be of additional value to detect small metastatic lesions in patients with prostate cancer and biochemical recurrence, which may have further clinical implications for secondary treatment.

Interscanner comparison of dynamic contrast-enhanced MRI in prostate cancer: 1.5 versus 3 T MRI.

PURPOSE: The aim of the study was the comparison of the diagnostic potential of dynamic contrast-enhanced magnetic resonance imaging to differentiate between prostate carcinoma and normal prostate tissue as well as prostatitis at 2 different field strengths: 1.5 versus 3 T. METHODS: Sixty-six patients with biopsy and/or prostatectomy of the prostate were included in the study. Magnetic resonance imaging was performed at 1.5 T in 20 patients with biopsy-proven prostate cancer (PC) and in 8 patients with prostatitis; at 3 T, we analyzed 27 patients with prostatectomy-proven PC and 11 patients with prostatitis. All examinations were performed using a combined body and endorectal coil protocol and a 2-dimensional TurboFLASH T1-weighted gradient echo sequence to calculate plasma flow (PF) and mean transit time (MTT) values. A total of 28 of 38 areas of normal prostate tissue, 20 of 27 areas of PC, and 8 of 11 prostatitis were analyzed at 1.5 or 3T. For the normalization, we calculated PC/normal and prostatitis/normal tissue ratios of PF and MTT for each patient. RESULTS: Prostate cancer showed higher PF (P < 0.0001) and shorter MTT (P < 0.0001) at 3 T and at 1.5 T (P < 0.0001 for PF and P = 0.0016 for MTT) compared with the normal tissue. In comparison with the normal tissue, prostatitis had a statistically significant higher PF at 1.5 T (P = 0.0156) but not at 3 T (P = 0.17) and no significantly shorter MTT values both at 3 (P = 0.15) and 1.5 T (P = 0.25). Sensitivity and specificity for differentiating PC from prostatitis with PF were 46% and 88% at 1.5 T (cutoff ratio, 2.3) and 89% and 73% at 3 T (cutoff ratio, 1.2), respectively. Sensitivity and specificity for MTT were 77% and 100% at 1.5 T (cutoff ratio, 0.7) and 70% and 100% at 3 T (cutoff ratio, 0.6), respectively. We found no significant relationship between the Gleason score and PF/MTT (P = 0.17/0.11 for 1.5 T and P = 0.23/0.18 for 3 T). CONCLUSIONS: The differentiation between PC and the normal tissue is possible with both field strengths. Prostate cancer can be better distinguished from prostatitis at 3 T compared with 1.5 T. The differentiation between prostatitis and the normal tissue is limited at both field strengths.