RPS 208 - Key insights in soft tissue neck imaging

Author Block: J-P. T. Vierula, J. Velhonoja, A. Sirén, J. Nurminen, M. J. Nyman, K. Mattila, J. Hirvonen; Turku/FI
Purpose: In patients with acute neck infections, MRI shows reactive edema patterns that predict disease severity: retropharyngeal edema (RPE) and mediastinal edema (ME). How well radiologists and clinicians with diverse backgrounds and neck MRI experience can detect these edema patterns is unknown.
Methods or Background: This prospective, blinded, multidisciplinary, multicenter interobserver study evaluated human performance in detecting RPE and ME from axial in-phase and water T2-weighted Dixon images. Readers (N=28, including radiologists, neuroradiologists, radiology residents, head and neck surgeons) from all five university hospitals in Finland were briefly trained and assessed the presence of RPE and ME (yes/no) and rated their confidence (1-5) blinded to clinical data. Emergency MRI images were obtained from 60 patients with acute neck infections. Ten patients appeared twice to assess intraobserver variability. Sensitivity, specificity, accuracy, and interobserver agreement were assessed.
Results or Findings: Overall sensitivity, specificity, and accuracy were 0.89, 0.81, and 0.85 for RPE and 0.85, 0.81, and 0.82 for ME. ME accuracy correlated with confidence (p=0.002), whereas RPE accuracy did not (p=0.580). Radiologists achieved higher RPE sensitivity (p=0.01), RPE accuracy (p=0.04), ME sensitivity (p=0.01), and ME accuracy (p=0.02) than clinicians, whereas other group comparisons were not significant. High confidence was found for RPE (4.3) and ME (4.1). Overall, interobserver kappa was 0.61 (substantial) for RPE and 0.52 (moderate) for ME, with radiologists showing higher agreement than clinicians. Median intraobserver accuracy was 90% for both RPE and ME.
Conclusion: We show high diagnostic accuracy and substantial interobserver agreement for detecting clinically relevant reactive edema patterns on MRI in patients with acute neck infections. These results encourage using these biomarkers in clinical practice.
Limitations: Limited availability of emergency MRI.
Funding for this study: This study was financially supported by the Sigrid Jusélius Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: A waiver for patient consent was not sought because it is not required by the national legislature for retrospective studies of existing data.

Author Block: S. Khongwirotphan, A. Prayongrat, S. Kitpanit, D. Kannarunimit, C. Chakkabat, V. Shotelersuk, S. Sriswasdi, C. Lertbutsayanukul, Y. Rakvongthai; Bangkok/TH
Purpose: Accurately prediction of overall survival (OS) is essential for optimizing treatment strategies in nasopharyngeal carcinoma (NPC), thereby improving patient outcomes. This study aimed to enhance OS prediction by integrating radiomic features from both intra- and peritumoral areas, offering a novel biomarker approach beyond traditional clinical features.
Methods or Background: We analyzed 251 NPC patients treated with chemoradiotherapy between 2010 and 2019, all followed for a minimum of three years. Radiomics features were extracted from the gross tumor volume (GTV) contours and a 3-mm peritumoral area of the pre-treatment CT images using PyRadiomics v3.0.1. The robustness and predictive power of radiomic features were assessed by intraclass correlation and univariate Cox regression. Selected radiomics features were combined with clinical data (age, gender, T-stage, N-stage) for a multivariate analysis. Cox regression models were optimized with recursive feature elimination and 20 repetitions of five-fold cross-validation, reserving 20% of dataset for model testing.
Results or Findings: Addition of peritumoral radiomic features significantly (P < 0.05) improved survival predictions (C-index: 0.787±0.067 validation; 0.669 test), over intratumoral only (C-index: 0.755±0.063 validation; 0.626 test set). Integrating clinical data with intra- and peritumoral radiomics yielded the best model, with a C-index of 0.832±0.052 in validation and 0.727 in test set, that outperformed (P < 0.05) the model with clinical and intratumoral features (C-index of 0.769±0.066 and 0.705, respectively). The baseline clinical model yielded a C-index of 0.703±0.100 in validation and 0.618 in test set.
Conclusion: Integrating radiomic features from both intra- and peritumoral areas significantly improved OS predictions in NPC, surpassing traditional approaches that utilize only clinical and intratumoral radiomics. This could lead to more personalized treatment strategies, potentially improving patient outcomes.
Limitations: External validation is recommended for this single-center retrospective study before clinical use.
Funding for this study: This research project is supported by National Research Council of Thailand (NRCT) and grants for development of new faculty staff, Ratchadaphiseksomphot Fund, Chulalongkorn University
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study received ethics approval from the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University, Thailand (IRB no. 0630/66).

Author Block: A. Longo, P. Hudler, P. Strojan, G. Plavc, L. Umek, K. Surlan Popovic; Ljubljana/SI
Purpose: Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can assess tumour vascularity, which depends on the process of angiogenesis and affects tumour response to treatment. Our study explored the associations between DCE-MRI parameters and the expression of plasma angiogenic factors in human papilloma virus (HPV)-negative oropharyngeal cancer, as well as their predictive value for response to concurrent chemoradiotherapy (cCRT).
Methods or Background: 25 patients with locally advanced HPV-negative oropharyngeal carcinoma were prospectively enrolled in the study. DCE-MRI and blood plasma sampling were conducted before cCRT, after receiving a radiation dose of 20 Gy, and after the completion of cCRT. Perfusion parameters ktrans, kep, Ve, initial area under the curve (iAUC) and plasma expression levels of angiogenic factors (vascular endothelial growth factor [VEGF], connective tissue growth factor [CTGF], platelet-derived growth factor [PDGF]-AB, angiogenin [ANG], endostatin [END] and thrombospondin-1 [THBS1]) were measured at each time-point. Patients were stratified into responders and non-responders based on clinical evaluation. Differences and correlations between measures were used to generate
prognostic models for response prediction.
Results or Findings: Higher perfusion parameter ktrans and higher plasma VEGF levels successfully discriminated responders from non-responders across all measured time-points, whereas higher iAUC and higher plasma PDGF-AB levels were also discriminative at selected time points. Using early intra-treatment measurements of ktrans and VEGF, a predictive model was created with cut-off values of 0.259 min-1 for ktrans and 62.5 pg/mL for plasma VEGF.
Conclusion: Early intra-treatment DCE-MRI parameter ktrans and plasma VEGF levels may be valuable early predictors of response to cCRT in HPV-negative oropharyngeal cancer.
Limitations: The small sample size and absence of healthy controls for angiogenic factors limit our findings. Perfusion parameter values in DCE-MRI can vary with different post-processing software, so comparisons between studies should be made with caution.
Funding for this study: This research was funded by the Slovenian Research and Innovation Agency (ARIS), grant number P3-0307.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the National Medical Ethics Committee of the Republic of Slovenia (No. 0120-247/2019/4, 12 June 2019) and the Committee for Medical Ethics of the Institute of Oncology Ljubljana (OI: 28.5.2019, ERIDEK-0064/2019). Written informed consent was obtained from all patients.

Author Block: S. Ruggiero¹, S. Lucchese², V. Dolcetti¹, S. Marzi¹, A. Vidiri¹; ¹Rome/IT, ²Naples/IT
Purpose: To conduct a multifactorial assessment in patients with oropharyngeal squamous cell carcinoma (OPSCC) that includes: 1) clinical characteristics; 2) detection of the amount of circulating tumor HPV-DNA in plasma (ctHPVDNA); 3) MRI-based volumetric analysis of both the primary tumor and lateral cervical lymph node metastases.
Methods or Background: A prospective study was conducted on patients with OPSCC. As controls, patients with suspected HPV-negative OPSCC were used. Both the primary tumor and lateral cervical lymph node metastases were manually delineated, slice by slice, by two expert head and neck radiologists on T2-weighted axial images at diagnosis. In patients with multiple lymph node metastases, the total volume was obtained by summing the volume of each lymph node. This measurement was correlated with the number of ctHPVDNA copies using the Mann-Whitney test.
Results or Findings: A total of 95 patients were included, of which 58 (61%) were p16+/HPV16+, 3 (3%) were p16-/HPV33+, 23 (24%) were p16-/HPV-, 1 (1%) was p16-/HPV16+, and 10 (10%) were p16+/HPV-. No association was found between the number of ctHPVDNA copies and the primary tumor volume. However, significant correlations emerged with the volume of lymph node metastases (Rho = 0.42, p = 0.004) and with the combined volume of the lymph node metastases and primary tumor (Rho = 0.51, p < 0.001).
Conclusion: ctHPVDNA is a promising biomarker that could potentially eliminate the need for solid biopsy for diagnosis. The data demonstrate an excellent correlation between p16+/HPV DNA testing, ctHPVDNA, and the volume of the primary tumor and lymph node metastases. This suggests that liquid biopsy could be useful in identifying the subgroup of patients with better oncological outcomes.
Limitations: Principal limitation of the study is the number of enrolled patients, but it was statistically appropriate.
Funding for this study: Partially supported by LILT 2020-21 Program and MAECI-Call for Joint Project Proposals Italy-Brazil # BR22GR03.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: All enrolled patients signed an informed consent to the
protocol approved by the IRCCS Regina Elena National Cancer Institute, Istituti Fisioterapici Ospitalieri, Institu- tional Review Board (RS1647/22).

Author Block: R. Abdalla-Aslan¹, G. E. Diana¹, A. Rchmiel¹, G. Merhav¹, S. Akrish¹, M. Javitt², D. Shilo¹, O. Emodi¹, N. Beck-Razi¹; ¹Haifa/IL, ²Miami/US
Purpose: Simple yet reliable methods for differentiating between benign and malignant soft tissue tumours of the oral cavity are currently lacking. Our primary aim was to assess the correlation between pre-operative intraoral ultrasound (US) variables and malignancy of oral lesions.
Methods or Background: This is a cross-sectional prospective study of consecutive patients over the age of 18 years from both genders, with a clinical diagnosis of a soft tissue lesion in the oral mucosa. Within a 2-weeks interval, patients who are scheduled for biopsy and histopathological examination underwent high-resolution intraoral US obtained using a 7-15 MHz-L15-7io-linear-ultrasound-transducer-‘hockey stick-probe’ on a Philips-Epiq-5, 7 machines (Philips Medical, Netherlands). Sonographic variables included: size in 3-dimensions, echogenicity, presence of cystic areas, presence of calcifications, margins and vascularity. The sonographic findings were compared with histopathology.
Results or Findings: Full data was available for 52 patients with 53 tumors. Included were 24 females and 28 males, with a mean age 60.11±16.7 years [range 18-90]. Following histopathological results, 22 patients with 23 tumors were diagnosed with squamous cell carcinoma (SCC), 3 patients with dysplasia (1 mild, 1 moderate and 1 severe), and the remaining 26 patients with benign lesions.
Sonographic variables of maximal diameter (a cutoff of 12 mm with 0.76 sensitivity and 0.77 specificity), margins (ill-defined, p<0.001) and vascularity (high or type III/IV, p=0.002) proved to be significantly correlated to SCC group, compared to benign lesions group.
Conclusion: The utility of intraoral US in the differential diagnosis of benign and SCC tumors in the oral cavity is demonstrated by this prospective clinical study, using sonographic features of maximal diameter, margins and vascularity.
Limitations: Small sample size and operator dependent technique.
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ehics committee of Rambam Health Care Campus reference RMB-19-0596.

Author Block: I. Burck, A. Gleich, R. Winkelmann, M. Fleischmann, E. Herrmann, J-E. Scholtz, P. Thönissen, T. Vogl, D. Pinto Dos Santos; Frankfurt/DE
Purpose: To explore radiological MR imaging features to predict response to neoadjuvant radiochemotherapy in patients with locally advanced oral-cavity tumour.
Methods or Background: We included 30 patients (15 women, mean age 60±10 years) with oral cavity cancer (stage IVa) who underwent neoadjuvant radiochemotherapy (RTX) before surgery. MRI scans were performed before RTX, 15 days after its initiation and preoperatively. Two radiologists retrospectively evaluated the images for overall tumour signal intensity (SI), SI change over time, and tumour extent using a Likert scale. Quantitative analysis was performed for the absolute SI of the tumor in ADC-, DWI-, and T2-weighted sequences normalized to the spinal cord. Tumour volume (TV) was calculated manually in a contrast-enhanced T1 sequence. Differences and ratios of ADC, DWI and T2-SI and TV were calculated. Patients with a stage pT1 or T0 were classified as responders, all others as non-responders.
Results or Findings: ADC-SIs at 2nd and 3rd MRI differed significantly between responders and non-responders (p = 0.010 and p = 0.013), as did the ratio between baseline and preoperative DWI-SIs (p = 0.041) and the difference between normalized baseline and preoperative ADC-SIs (p = 0.049). Non-responders showed an increase in TV at the 2nd MRI, while responders showed a significant decrease in TV, so the calculated percentage decrease and ratio are significant markers of response progression.
Conclusion: Diffusion weighted imaging parameters as well as tumour volumetry may predict response to neoadjuvant radiochemotherapy in oral cavity cancer and may be benefical for image guided treatment potentially be used to guide treatment or extent of surgery in these patients.
Limitations: Limitations of this study include small sample size, single-center and retrospective study design.
Funding for this study: None.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethics approval was obtained by the institutional review board (approvals number 208/12).

Author Block: F. Albisinni¹, C. Carbone¹, C. Zacchi¹, M. Ravanelli¹, D. Farina¹, B. Van Deberge²; ¹Brescia/IT, ²Leuven/BE
Purpose: This study aimed to assess the impact of deep learning (DL)-based reconstructions on T2 image quality in head and neck MRI. Additionally, potential time savings from using DL in various sequences were evaluated.
Methods or Background: Three sequences were compared: (A) TSE T2 without DL, with three signal averages and an acquisition time of 2'35''; (B) TSE T2 with DL at intermediate strength, with two signal averages and an acquisition time of 1'25''; and (C) TSE T2 with DL at maximal strength, with one signal average and acquisition time of 43''. Images from 52 patients were randomly and blindly evaluated by three radiologists with varying levels of experience using MR scanners from different vendors. For each patient, two images were analyzed at the level of the nasopharynx and oral cavity. Four categories were assessed: overall image quality, artifacts, edge sharpness, and noise, each rated on a 3-point Likert scale. Additionally, radiologists were tasked with identifying the correct sequence for each image. The two main endpoints evaluated were inter-rater reproducibility and comparison of the sequences
Results or Findings: A total of 636 images were rated by three radiologists. Inter-rated reproducibility was poor across all categories. The sequences type was correctly identified in only 44% cases. The overall quality scores for sequence A, B and C were 2.63, 2.52, and 2.52 respectively (p=0.055). Artifacts scores were 2.61, 2.48, and 2.55 (p=0.07); edge sharpness scores were 2.57, 2.39, and 2.43 (p=0.02, with significant differences between sequences A and B); and noise scores were 2.51, 2.41, and 2.41 (p=0.1)
Conclusion: The performances of the three sequences were similar overall. DL-based sequences for head and neck MRI were shown effective and offered significant time savings, enabling potential ultra-fast imaging protocols
Limitations: Small sample
Funding for this study: No funding was provided for this study
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: Not applicable