Faculty Bibliography

Radiomics is a rapidly evolving field of research concerned with the extraction of quantitative metrics-the so-called radiomic features-within medical images. Radiomic features capture tissue and lesion characteristics such as heterogeneity and shape and may, alone or in combination with demographic, histologic, genomic, or proteomic data, be used for clinical problem solving. The goal of this continuing education article is to provide an introduction to the field, covering the basic radiomics workflow: feature calculation and selection, dimensionality reduction, and data processing. Potential clinical applications in nuclear medicine that include PET radiomics-based prediction of treatment response and survival will be discussed. Current limitations of radiomics, such as sensitivity to acquisition parameter variations, and common pitfalls will also be covered.

We evaluated the performance of radiomics and artificial intelligence (AI) from multiparametric magnetic resonance imaging (MRI) for the assessment of breast cancer molecular subtypes. Ninety-one breast cancer patients who underwent 3T dynamic contrast-enhanced (DCE) MRI and diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) mapping were included retrospectively. Radiomic features were extracted from manually drawn regions of interest (n = 704 features per lesion) on initial DCE-MRI and ADC maps. The ten best features for subtype separation were selected using probability of error and average correlation coefficients. For pairwise comparisons with >20 patients in each group, a multi-layer perceptron feed-forward artificial neural network (MLP-ANN) was used (70% of cases for training, 30%, for validation, five times each). For all other separations, linear discriminant analysis (LDA) and leave-one-out cross-validation were applied. Histopathology served as the reference standard. MLP-ANN yielded an overall median area under the receiver-operating-characteristic curve (AUC) of 0.86 (0.77-0.92) for the separation of triple negative (TN) from other cancers. The separation of luminal A and TN cancers yielded an overall median AUC of 0.8 (0.75-0.83). Radiomics and AI from multiparametric MRI may aid in the non-invasive differentiation of TN and luminal A breast cancers from other subtypes.

The macrolide clarithromycin has been reported as active for therapy of mucosa associated lymphoid tissue (MALT) lymphoma. Pharmacokinetic properties, however, require continuous daily intake over a prolonged period of time. As the macrolide azithromycin is characterized by a long half-life as well as potential antineoplastic activity in vitro, we have performed a phase II trial of long-term once-weekly oral azithromycin for treatment of MALT lymphoma. In a 2-stage-design, 16 patients (10 f/6 m) with histologically verified and measurable MALT lymphoma were included in the first phase of the trial, which could be expanded to a maximum of 46 patients depending on remissions in the first phase. Patients were given oral azithromycin 1500 mg once-weekly 4 times a month, and restaging was performed after 3 and 6 months. Two patients had gastric and 14 extragastric MALT lymphoma; 12/16 patients were treatment-naive and received azithromycin as first line treatment. Tolerance of this regimen was excellent, and 14/16 patients received 6 months of treatment as scheduled, while 1 patient each discontinued after 4 (progressive disease) and 1 cycle (personal reasons), respectively. The most commonly observed side effects were mild nausea (n = 8) and diarrhea (n = 4). Efficacy, however, was low as only 4/16 patients (25%) responded, with 2 complete and 2 partial remissions, 9 patients (56%) had stable disease, and 3 patients 19%) were rated as progressive disease. As the predefined activity of more than 7/16 patients responding was not reached, the study was stopped after 16 patients. Although long-term once-weekly oral azithromycin showed some antilymphoma activity, the response rate was below the predefined threshold of interest. However, based on our data, one cannot rule out suboptimal dosing in our study; attempts to study azithromycin at a different mode of application might be warranted in the future.

In this case report, we present a case of false positive CA 19-9 and CA 125 levels in a patient with suspected endometriotic cysts. The patient is a 34-year-old nulliparous woman with heavy black tea consumption and elevated CA 19-9 and CA 125 levels. After discontinuation of black tea intake and careful exploration of other possible factors, CA 19-9 and CA 125 levels dropped markedly. As a conclusion, heavy black tea consumption can lead to false positive results of elevated CA 19-9 and CA 125 levels.

PURPOSE:F]FDG/positron emission tomography (PET) are established outcome predictors in FDG-avid lymphomas. We therefore investigated whether these biomarkers also have prognostic value in extranodal marginal zone B cell lymphoma of the mucosa-associated lymphoid tissue (MALT lymphoma), with a focus on patients treated with anti-CD20 antibody-based immunotherapy. PROCEDURES:F]FDG/PET parameters, and Kaplan-Meier estimates with log rank tests were performed. RESULTS:After 2 years, progression had occurred in 12/61 patients (CD20-anitbody group 6/35). TLG emerged as the only significant prognostic factor for 2-year PFS in the multivariate analyses with forward selection, both in entire cohort (hazard ratio HR, 1.001; 95 % CI, 1.001-1.002; P < 0.0001) and in the CD20-antibody group (HR, 1.001; 95 % CI, 1.001-1.002; P = 0.001). However, in the entire population, where 8/26 patients with a TLG > 90 (30.8 %) vs. 4/35 patients with a TLG ≤ 90 (11.4 %) showed progression within the 2-year observation period, TLG-based separation of risk groups failed (HR, 0.35; 95 % CI, 0.10-1.15; P = 0.069); whereas in the CD20-antibody group, where 6/16 patients with a TLG > 90 (37.5 %) vs. 0/19 patients with a TLG ≤ 90 (0.0 %) showed progression, risk group separation was successful (HR, 0.010; 95 % CI, 0.0001-8.068; P = 0.003). CONCLUSIONS:TLG may improve early risk stratification of MALT lymphoma patients treated with CD20-antibody-based immunotherapy.

PURPOSE:Aim of present study was to determine whether the currently recommended 13-cm cranio-caudal diameter cut-off on CT for assessment of splenic involvement in lymphoma offers adequate sensitivity and specificity. MATERIALS AND METHODS:Patients with histologically proven lymphoma who had undergone [18F]FDG-PET/CT before therapy were included. Cranio-caudal diameters of the spleen were measured on the CT component of PET/CT, and ROC analyses with calculation of respective areas under the curve (AUC) were used to determine cut-off values of cranio-caudal measurements with their respective sensitivities and specificities, using [18F]FDG-PET as the reference standard. RESULTS:In 93 patients, we found a sensitivity of 74.1% and a specificity of 47% for the 13-cm splenic diameter cut-off. CONCLUSIONS:Our results show reasonable, though far from perfect sensitivities and specificities for the currently recommend 13-cm splenic diameter cut-off.

PURPOSE/OBJECTIVE:The aim of this study was to compare the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1, the immune RECIST (iRECIST) criteria, and the Positron Emission Tomography Response Criteria in Solid Tumors (PERCIST) 1.0 in patients with advanced non-small cell lung cancer treated with programmed cell death protein 1 (PD-1)/programmed cell death protein 1 ligand (PD-L1) inhibitors. METHODS:This prospective study of 42 patients treated with a PD-1/PD-L1 inhibitor was approved by our institutional review board, and all patients gave written, informed consent. Tumor burden dynamics were assessed on F-FDG PET/CT before and after treatment initiation. Immunotherapeutic responses were evaluated according to RECIST 1.1, iRECIST, and PERCIST 1.0 for the dichotomous groups, responders versus nonresponders. Cohen κ and Wilcoxon signed rank tests were used to evaluate concordance among these criteria. We assessed progression-free survival and overall survival using the Kaplan-Meier estimator. RESULTS:The RECIST 1.1 and PERCIST 1.0 response classifications were discordant in 6 patients (14.2%; κ = 0.581). RECIST 1.1 and iRECIST were discordant in 2 patients, who evidenced pseudoprogression after treatment initiation. Median progression-free survival, as well as overall survival, was significantly longer for responders compared with nonresponders for all criteria (P < 0.001), with no significant difference between the 3 criteria (P > 0.05). CONCLUSIONS:RECIST 1.1 and PERCIST 1.0 show only moderate agreement, but both can predict treatment response to PD-1/PD-L1 inhibitor therapy. In case of pseudoprogression, metabolic tumor activity may help to correctly classify treatment response.

PURPOSE/OBJECTIVE:F]FDG PET/CT-derived radiomic features alone or in combination with clinical, laboratory and biological parameters are predictive of 2-year progression-free survival (PFS) in patients with mantle cell lymphoma (MCL), and whether they enable outcome prognostication. METHODS:F]FDG PET/CT scans. A multilayer perceptron neural network in combination with logistic regression analyses for feature selection was used for prediction of 2-year PFS. International prognostic indices for MCL (MIPI and MIPI-b) were calculated and combined with the radiomic data. Kaplan-Meier estimates with log-rank tests were used for PFS prognostication. RESULTS:SUVmean (OR 1.272, P = 0.013) and Entropy (heterogeneity of glucose metabolism; OR 1.131, P = 0.027) were significantly predictive of 2-year PFS: median areas under the curve were 0.72 based on the two radiomic features alone, and 0.82 with the addition of clinical/laboratory/biological data. Higher SUVmean in combination with higher Entropy (SUVmean >3.55 and entropy >3.5), reflecting high "metabolic risk", was associated with a poorer prognosis (median PFS 20.3 vs. 39.4 months, HR 2.285, P = 0.005). The best PFS prognostication was achieved using the MIPI-bm (MIPI-b and metabolic risk combined): median PFS 43.2, 38.2 and 20.3 months in the low-risk, intermediate-risk and high-risk groups respectively (P = 0.005). CONCLUSION/CONCLUSIONS:F]FDG PET/CT-derived radiomic features SUVmean and Entropy may improve prediction of 2-year PFS and PFS prognostication. The best results may be achieved using a combination of metabolic, clinical, laboratory and biological parameters.

Based on results of two pilot trials, lenalidomide (LEN) was found to be active and safe as monotherapy and showed an increased response rate of 80% in combination with rituximab (R) for patients with mucosa-associated lymphoid tissue (MALT) lymphoma. While initial results were promising, there are currently no data on long-term outcome, and larger international phase II/III trials on LEN for indolent lymphoma lack specific subgroup analyses. Thus, we have systematically analyzed 50 patients treated with LEN-based therapy (LEN-monotherapy n = 16, R-LEN n = 34) at the Medical University of Vienna 2009 to 2019 and investigated long-term outcome and relapse patterns. At a follow-up of more than 5 years (median 68 months), 54% of patients are free of relapse, and estimated median progression-free survival (PFS) was 72 months (95%CI 49-96). There was no difference in PFS according to stage of disease, i.e. localized versus disseminated disease (P = .67) and previous systemic treatment (P = .16). Interestingly, but with the caveat of the limited number of patients included in this series, primary extragastric disease had a superior PFS compared with gastric lymphoma (P = .04) and also depth of response, i.e. complete or partial response versus stable disease was associated with significantly prolonged PFS (P = .01). We documented four patients (8%) with pronounced improvement of response during follow-up including three patients initially rated as partial remission and finally achieving complete remission at 12 to 32 months. This highlights the potential of delayed responses to LEN treatment. Estimated overall survival at 5 years was excellent at 92%. These "real-world" data confirm long-term activity of LEN in MALT lymphoma.

The response evaluation criteria in lymphoma (RECIL) classification for lymphoma treatment response assessment was introduced in 2017, but it has not yet been compared to the established Lugano classification. Also, the value of the provisional "minor response" (MiR) category of RECIL is unclear. In 54 patients with FDG-avid non-Hodgkin lymphomas (41 diffuse large B-cell lymphomas (DLBCL) and 13 follicular lymphomas), [¹⁸F]FDG-PET/CT-based response according to RECIL and Lugano was determined at interim and end-of-treatment (EOT) restaging. Rates of agreement and Cohen's kappa (κ) coefficients were calculated. The relationship between RECIL and Lugano responses and 2-year complete remission (CR) status of DLBCL patients was determined. At interim restaging, MiR was observed in 14.8%, and at EOT, in 5.6% of patients. When MiR was recoded as partial remission, agreement between RECIL and Lugano was 83.3% at interim restaging (κ = 0.69), and 90.7% at EOT (κ = 0.79). 85.4%, of DLBCL patients with responding disease at interim restaging according to both RECIL and Lugano achieved 2-year CR status; whereas, at EOT, 82.9% of patients with responding disease according to Lugano, and 85.4% of patients with responding disease according to RECIL, achieved 2-year CR status. Thus, RECIL and Lugano classifications show comparable performance for treatment response assessment, and a similar association with 2-year CR status in FDG-avid lymphomas.