Prospective validation of 18F-Fluoroethylcholine as a tracer in PET/MRI for the evaluation of breast lesions and prediction of lymph node status

This prospective, single-center EudraCT-registered (2017-003089-29) diagnostic study was approved by the national authorities and the local ethics committee. All patients gave written, informed consent. The study was performed in accordance with the Declaration of Helsinki statement for medical research involving human subjects.

Inclusion criteria were breast lesions classified as suspicious on conventional imaging (ACR BI-RADS 4 or 5 on mammography, tomosynthesis, and/or ultrasound); age above 18 years; and availability of histopathological confirmation of the lesion.

Exclusion criteria were unstable or non-compliant patients; pregnant or breast-feeding patients; radiation therapy or chemotherapy within the last 6 months or surgical interventions less than 12 weeks before the PET/MRI examination; known contraindications to MRI and/or the intravenous administration of gadolinium; renal insufficiency. Data were collected on adverse events that occurred during or after the examination.

PET-MRI of the breast

The examinations were performed using a simultaneous whole-body combined PET/MRI device (Biograph mMR system, Siemens, Erlangen, Germany), characterized by an MRI-compatible PET detector integrated with a 3.0 Tesla MRI scanner.

Radiosynthesis of 18F-Fluoroethylcholine followed a two-step reaction procedure using a remote-controlled synthesizer (Nuclear Interface, GE Healthcare, Uppsala, Sweden). 18F-Fluoride was produced on-site using [18O]H2O and a medical cyclotron via an 18O(p,n)18F reaction (GE PET trace, GE Medical Systems, Uppsala, Sweden). After azeotropic drying, 18F-Fluoride reacted with bromoethyltriflate to yield the radiolabeled synthon 18F-Bromofluoroethane, which, after distillation, reacted with dimethylaminoethanol to give the crude product 18F-Fluoroethylcholine. Purification was achieved by cation exchange (solid-phase extraction) and the final product was obtained after elution with physiological saline. Quality control was performed according to the European Pharmacopoeia [14].

Examinations were performed with the patients in a prone position using a dedicated 16-channel breast coil (Rapid Biomedical, Rimpar, Germany). PET and contrast-enhanced MRI (CE-MRI) acquisitions were performed simultaneously. CE-MRI was performed according to EUSOMA recommendations [15]. The protocol included the following sequences: T2-TSE, STIR, DWI and T1-Dixon-TWIST. T1-weighted sequences were acquired before and after intravenous administration of a paramagnetic contrast agent (Dotarem: 0.2 ml/kg), at a flow rate of 3.5 ml/s. All sequences were acquired in the axial plane.

PET acquisition started immediately after the injection of 2.5–3.5 MBq/kg of 18F-FEC. MRI-based attenuation correction was applied using Dixon-VIBE sequences, with in-phase and opposed-phase, as well as fat-saturated and water-saturated images. A three-dimensional (3D) acquisition technique was used that offered an axial field of view (FOV) of approximately 26 cm and a transverse FOV of 59 cm with a sensitivity of 13.2 cps/kBq.

Image analysis

Lesions were initially detected and evaluated on CE-MRI of the breast by an experienced breast radiologist (> 8 years of experience). Clearly benign findings on MRI, such as simple cysts or non-enhancing lesions, were discarded. Suspicious findings, requiring histological verification, were selected for further analysis and maximum lesion diameter was measured. The targeted lesion was then correlated with the 18F-FEC PET images and quantitative evaluations were performed by a nuclear medicine physician (> 7 years of experience). 18F-FEC uptake was measured with a dedicated software (Hermes 3D Hybrid Viewer, Hermes Medical Solutions, Stockholm, Sweden) by drawing a region of interest (ROI) on the lesion. In case of a suspicious lesions in the breast, the axillary lymph nodes were also sampled. The lymph node with the highest FEC uptake was considered, regardless of the morphological lymph node characteristics on MRI. The slice in which the breast lesion or axillary finding showed the maximum uptake was selected for the evaluation, and the minimum, mean, and maximum standardized uptake value (SUVmin, SUVmean, SUVmax) was recorded for each lesion (SUVT) and lymph node (SUVLN). Readers were blinded to the previous imaging examinations, clinical history of the patient, and final histology, but they were not blinded to the simultaneous CE-MRI findings.

Histopathological analysis

Included breast lesions and suspicious lymph nodes underwent image-guided core needle or vacuum-assisted biopsy.

In patients who did not undergo neoadjuvant chemotherapy and in whom a sentinel lymph node biopsy was performed, these results were considered the standard of reference for the evaluation of the lymph node status.

For this analysis, lymph nodes with macrometastases (at least one metastasis > 2.0 mm) were considered positive, while lymph nodes with micrometastases (> 0.2 mm and/or > 200 cells but < 2.0 mm) or isolated tumor cells were considered negative [16, 17]. Malignant breast lesions underwent immunohistochemical evaluation of estrogen receptor (ER), progesterone receptor (PR), and HER2 receptor status. If HER2 status was ambiguous, fluorescent in situ hybridization was performed. The MIB-1 monoclonal antibody was used to determine the proliferation activity (expression of the Ki-67 antigen as determined by standard MIB-1 antibodies). All features were dichotomized according to international guidelines [16]. Ki-67 was considered high when > 20%. The material was analyzed by dedicated breast pathologists.

Statistical analysis

Statistical analysis was performed using dedicated software (IBM SPPS Statistics for Windows, v. 20.0.0, Armonk, NY).

The comparison of the continuous 18F-FEC SUVmaxT values between different lesion types and characteristics was performed with either the Mann–Whitney U test (two independent samples) or the Kruskal–Wallis test (three or more independent samples). In addition, a Pearson correlation coefficient matrix was used to assess the correlation between SUVmaxT, SUVmaxLN, cancer characteristics, and lymph node status in invasive carcinomas. Partial correlation was used to measure the effect of lesion size.

The diagnostic performance of 18F-FEC SUVmax was evaluated using the area under the Receiver Operating Characteristics curve (ROC).

Differences were considered significant at p < 0.05. No alpha error accumulation correction was used in this exploratory study. Therefore, interpretation of statistically significant results must consider the possibility of false-positive significances.

留言 (0)

沒有登入
gif