SS 1006 - Clinical molecular imaging
Ultra-fast 18F-FDG PET/MRI compared to 18F-FDG PET/CT and CT in whole-body staging of females with recurrent pelvic malignancies
Purpose: To evaluate the diagnostic feasibility of an ultra-fast 18F-FDG PET/MRI protocol, including T2-w and contrast-enhanced T1-w imaging as well as metabolic assessment (PET) in comparison to 18F-FDG PET/CT and CT for whole-body staging of female patients with suspected recurrence of pelvic malignancies
Methods and Materials: 43 female patients with suspected tumour recurrence were included in this study. All patients underwent a PET/CT and a subsequent PET/MRI examination. Two readers were asked to evaluate ultra-fast PET/MRI, PET/CT as well as CT datasets of PET/CT separately for suspect lesions regarding lesion count, lesion localization and lesion characterization. Statistical analyses were performed both, on a per-patient and a per-lesion basis.
Results: Tumour relapse was present in 38 of 43 patients. Based on CT readings 25/38 tumour relapses were correctly identified. PET/CT enabled correct identification of 37/38 patients, PET/MRI correctly identified 36/38 patients with recurrent cancer. On a lesion-based analysis PET/MRI enabled the correct detection of more lesions, comprising a lesion-based sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of 50%, 58%, 76%, 31%, and 53% for CT, 97%, 83%, 93%, 94%, and 92% for PET/CT and 98%, 83%, 94%, 94%, and 94% for PET/MRI, respectively. Mean scan duration of ultra-fast PET/MRI, PET/CT and whole-body CT amounted to 18.5 ± 1 minutes, 18.2 ± 1 minutes and 3.5 minutes, respectively.
Conclusion: Ultra-fast PET/MRI provides equivalent diagnostic performance and examination time when compared to PET/CT and superior diagnostic performance to CT in restaging female patients suspected to have recurrent pelvic cancer.
Characterisation of complicated and recurrent urological infections using molecular imaging of the chemokine receptor CXCR4 in combination with diffusion-weighted MRI
Purpose: To obtain mechanistic insights into complicated recurrent urological infections in patients after kidney transplantation using molecular imaging of chemokine receptor CXCR4-expression by means of positron emission tomography (PET) in combination with diffusion-weighted magnetic resonance imaging (MRI).
Methods and Materials: Twelve patients with suspected acute kidney transplant infection were examined using PET with the specific CXCR4-ligand, gallium-68- (Ga-68-)pentixafor as well as diffusion-weighted MRI (10 b-values). The spatial distribution and intensity of up-regulation of CXCR4-expression on PET and diffusion restriction as determined by apparent diffusion coefficients (ADCs) on MRI were analyzed, and compared to serum and urine analyses, microbiologic evaluation and immunohistochemistry of kidney biopsy specimens.
Results: Patterns consistent with acute transplant infection were detected in 7/12 patients on PET and MRI. In focal lesions, ADC was reduced compared to unaffected parenchyma (1.110±0.125 vs. 1.577±0.028*10-3 mm2/s, p<0.001). Areas of ADC-reduction corresponded to up-regulated CXCR4-expression compared to parenchyma (SUVmean 5.37±2.37 vs. 3.63±1.92, p<0.01). In the other patients, imaging detected no signs of renal involvement, consistent with lower urinary tract infection. Leukocyturia was present in all patients with evidence for renal inflammation. Immunohistochemistry of biopsies of acute transplant infection showed CXCR4-expression in infiltrating leukocytes.
Conclusion: Combined PET/MR imaging with Ga-68-pentixafor detects acute renal infections and provides insights into the pathophysiology of the disease. Up-regulated CXCR4-expression in infiltrating leukocytes corresponded to areas of increased cell density on MRI. Combined PET/MRI may have the potential to confirm the presence, extent and localization of acute renal infection providing relevant information for the therapeutic management of patients.
Purpose: To explore the potentialities of CE-FDG PET/MR to provide insights into genetics and grade of differentiation of ductal invasive breast cancer.
Methods and Materials: 21 non-operated invasive ductal breast cancer patients, with known degree of expression of oestrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor 2 (HER2), antigen Ki-67 (Ki67), and histology grading, underwent CE-FDG PET/MR. SUV max, Ktransmean, Kepmean, Vemean, iAUC, and ADCmean were compared with the expression of ER, PR, HER2, KI67 and with grading, using Student’s t-test.
Results: Of the explored biomarkers only Kepmean, ADCmean, and SUVmax were capable of discriminating expression of ER, PR, HER2, Ki67, and grading. Higher Kepmean and SUVmax differentiated ER- and PR- from ER and/or PR+ cancers (Kepmean 9234±1320min-1versus 6492±2359min-1, p=0.011; SUVmax 14.19±7.17 versus 6.18±4.34, p=0.005). ADCmean, Kepmean, and SUVmax were higher in HER2- rather than in HER2+ tumours (ADC 1303±121 x10-6mm2/s versus 1022±280 x10-6mm2/s, p=0.009; Kepmean 8599±2122min-1 versus 6322±2241min-1, p=0.028, and SUVmax 11.79±7.65 versus 6.17±4.02, p=0.046, respectively). Only ADCmean discriminated Ki67≤14% versus Ki67>14% cancers (964±242x10-6 mm2/s versus 1252±214x10-6 mm2/s, p=0.011). Both Kepmean and SUVmax demonstrated lower values in G2 rather than in G3 cancers (Kepmean 6638±2392 min-1 versus 8944±1765 min-1, p=0.036; SUVmax 6.83±4.73 versus 12.89±8.08, p=0.042).
Conclusion: ADCmean, Kepmean, and SUVmax correlated with expression of ER, PR, HER2, and Ki67 in ductal breast cancer and also with grading. CE-FDG PET/MR has the potentiality to non-invasively provide insight into genetics and grading of invasive ductal breast cancer. This might aid better patient management optimizing personalized therapy.
Purpose: CXCR4 is a chemokine receptor that was reported to be overexpressed in invasive breast cancer and seems to play a major role in signalling pathways of metastases. The aim of this study was to assess the feasibility of CXCR4-directed imaging in patients with breast cancer using the novel CXCR4-targeted PET probe 68Ga-pentixafor.
Methods and Materials: 12 patients suffering from primary breast cancer underwent PET imaging (either PET/MR or PET/CT) using 68Ga-pentixafor. The lesions included 10 invasive ductal carcinomas (IDC) and two invasive lobar cancers (ILC). Five patients were known to have lymph node or distant metastases detected by previously performed imaging modalities. Maximum standardised uptake values (SUVmax) and tumour-to-background ratios (T/B ratio) were determined in the breast cancer lesions.
Results: 8 of 12 primary breast cancers were visually detectable with a mean SUVmax of 2.75 (range 1.7 to 4.5) and a mean T/B ratio of 2.2. The visually undetectable lesions included the two cases of ILC, and two cases of IDC (G2). Highest SUVmax T/B rations of 4.37/3.9 and 4.5/3.5, respectively, were demonstrated in an IDC (G3) with evidence of bone and lymph node metastases and in an IDC (G2) without any metastases. A statistical significant correlation between SUVmax, T/B ratios and occurrence of metastases was not observed.
Conclusion: CXCR4-directed PET imaging of breast cancer is feasible. Based on our first observations in this small patient cohort, tracer accumulation was significantly higher in IDC compared to ILC.
Correlation of apparent diffusion coefficient value on diffusion-weighted imaging and SUV values on PSMA PETCT in patients with biopsy-proven prostate cancer
Purpose: Correlation of apparent diffusion coefficient value on diffusion-weighted imaging and SUV values on PSMA PETCT in patients with biopsy-proven prostate cancer.
Methods and Materials: This is a prospective study conducted in our hospital, department of molecular imaging over last 6 months. 22 patients with biopsy-proven prostate cancer were included. All these patients underwent PSMA PETCT imaging as a staging investigation and were subjected to prostatic pelvic MRI (3T SKYRA 48 channels). PSMA PETCT images were analysed on GE work station (AW 4.6). Mean and maximum SUV values were obtained. MRI images were analysed on MMWP work station and ROIs were drawn for the ADC values at the regions of elevated SUV values. ADC values were compared with SUV values using one-way Pearson bivariate relationships were determined between SUV and ADC. T2W MR images were also correlated with ADC values and PET-MRI fusion with T2 W axial images was performed to assess the extent of the lesion.
Results: There was a significant negative correlation between SUV max with mean ADC (r =-525, p=0.01). There was a significant negative correlation of max SUV with min ADC (r= -0.602, p=0.005). There was a significant negative correlation between min SUV with mean ADC (-0.61, p=0.004) . The ADC values were significantly low and mean SUV values were high with Gleason's score above 8.
Conclusion: Quantitative ADC values in correlation with PSMA PETCT help in localisation, assessing the extent of the disease and also determining the aggressiveness of the disease.
Purpose: To evaluate diagnostic accuracy of 68Ga-PSMA- PET/MRI in suspected recurrent prostate cancer.
Methods and Materials: 25 patients with suspected recurrent PCa underwent 68Ga-PSMA PET/CT followed by integrated PET/MRI. Data from both investigations were analysed separately and compared regarding tumour detection rate, image quality and radiotracer uptake in tumour lesions and muscle tissue. Image analysis was prospectively performed by different groups of nuclear medicine physicians and radiologists with respect to the detection of lymph node metastases, bone metastases and local recurrence of the tumour. Image quality was evaluated visually based on a three-point ordinal scale.
Results: 68Ga-PSMA-PET/MRI provided diagnostic image quality in all examined patients. Overall detectability of PET-positive lesions was higher with PET/MRI compared to PET/CT (43 vs 36), identifying 14 vs 9 PET-positive local recurrences, 23 vs 20 PET-positive lymph nodes, 2 vs 3 PET-positive soft-tissue lesions, and 4 vs 4 PET-positive bone lesions, respectively. Mean PET image quality obtained with 68Ga-PSMA-PET/MRI and -PET/CT was 2.8±0.4 and 2.7±0.5, respectively (p=0.1604).
Conclusion: 68Ga-PSMA-PET/MRI is superior to 68Ga-PSMA-PET/CT in the detection of PSMA-expressing prostate bed recurrences and lymph node lesions and as reliable as 68Ga-HBED-CC-PSMA-PET/CT in the detection of bone lesions.
Detection efficacy of hybrid 68Ga-PSMA ligand PET/CT in prostate cancer patients with biochemical recurrence after primary radiation therapy defined by Phoenix criteria
Purpose: The aim of this study was to evaluate the detection rate of 68Ga-PSMA PET/CT in patients with biochemical recurrent prostate cancer defined by Phoenix criteria after external beam radiotherapy or brachytherapy as primary treatment.
Methods and Materials: 118 patients were included for this retrospective analysis. Of these, 45 were receiving androgen deprivation therapy (ADT) within at least 6 months prior to the PET/CT. The detection rates were stratified by PSA. The influence of primary Gleason score (GS) and ADT was assessed. Correlations between standardised uptake values (SUV) and GS or ADT in patients with positive findings were analysed.
Results: 90.7% (107/118) patients showed pathological findings indicative for tumour recurrence in 68Ga-PSMA PET/CT. The detection rates were 81.8% (36/44), 95.3% (41/43) and 96.8% (30/31) for PSA of 2 to <5, 5 to <10 and ≥10 ng/mL, respectively. The detection rate was significantly higher in patients with ADT (97.7%) vs without ADT (86.3%, p=0.0381), but independent from primary GS ≥8 (92.0%) vs ≤7 (90.2%, p=0.6346). SUVmax and SUVmean were significantly higher in patients with ADT (p=0.0025 and 0.0044, respectively) and a clear trend to higher values was observed for patients with GS ≥8 (p=0.0502 and 0.0514, respectively).
Conclusion: 68Ga-PSMA PET/CT demonstrates high detection rates in patients with biochemical recurrence of prostate cancer after primary radiation therapy. The detection rate positively correlated to increasing PSA as well as concomitant ADT. 68Ga-PSMA PET/CT enables discrimination of local vs systemic disease and thus might have a crucial impact on further clinical management.
Purpose: To analyse the diagnostic performance of 68Ga-DOTATATE PET-CT and contrast-enhanced MRI for detection of transosseous extension of intracranial meningiomas and to find associations with imaging parameters.
Methods and Materials: In this retrospective study, subjects were selected from a cohort of 325 consecutive patients who underwent 68Ga-DOTATATE PET-CT of the head for evaluation of confirmed or suspected meningioma. Inclusion criteria were (1) MR imaging within 1 month of either pre- or postoperative matching PET-CT imaging, and (2) as standard of reference pathology-confirmed meningioma diagnosis with inclusion or exclusion of transosseous extension. Imaging was independently analyzed by two readers with respect to osseous involvement, absolute tracer uptake measured as SUVmax, volume of meningioma and surrounding edema. Chi-square, Mann-Whitney U or exact McNemar's tests as well as receiver operating characteristics (ROC) analyses were performed to compare variables and diagnostic test performance.
Results: Eighty-two patients fulfilled the inclusion criteria. Patients with transosseous meningiomas (n=67) showed significantly larger lesions (median 10ml vs 3ml, p=0.002) and significantly higher SUVmax (median 14 vs 8, p=0.032) compared to extraosseous meningiomas. 68Ga-DOTATATE PET-CT in comparison to MRI performed at a considerably higher sensitivity (98.5% vs 55.2%) while maintaining high specificity (86.7% vs 100.0%); statistically significant differences between the tests were observed (p<0.001). In ROC analysis, PET-CT assessment performed best (AUC: 0.933), followed by MRI assessment (0.866), meningioma volume (0.765) and SUVmax (0.680) while surrounding oedema added no valuable information (0.483).
Conclusion: 68Ga-DOTATATE PET-CT enables improved detection of the transosseous extension of intracranial meningiomas compared to MRI.
Purpose: There is an emerging demand for dedicated brain PET for the study and diagnosis of neurological disorders such as dementia. We have proposed a helmet PET geometry with add-on detector for high sensitivity, high resolution and low cost brain PET. Our first prototype had an add-on detector at the chin position (helmet-chin PET). However, the gantry design gave oppressive feeling to the patient and the movement of the chin detector was inconvenient for the patient set-up. On the other hand, our previous simulation showed that add-on detectors covering the neck has equivalent effect. In this study, therefore, we developed a helmet-neck PET, which has additional detectors at the neck.
Methods and Materials: We developed the helmet PET prototype using 54 four-layer depth-of-interaction detectors with 16×16×4 array of 2.8×2.8×7.5 mm3 GSO crystals. In the prototype, 47 detectors were arrangement in a hemisphere, and seven detectors were placed as the add-on detector. We compared the helmet-chin PET and the helmet-neck PET regarding the effects of the add-on detectors for increasing the sensitivity.
Results: The sensitivity measured with a hemispherical phantom for the brain region was increased by 12% in the helmet-chin PET and by 19% in the helmet-neck PET compared with the geometry without add-on detectors. This is because the arrangement of the detector at chin required larger margin for designing gantry, in which detector position became farther than the case of neck.
Conclusion: The helmet-neck PET has promising performance for high sensitivity brain imaging and improved convenience for patient studies.