Diagnostic performance of computed tomography features in detecting oropharyngeal squamous cell carcinoma extranodal extension

International Multi-Specialty Expert Physician Preoperative Identification of Extranodal Extension n Oropharyngeal Cancer Patients using Computed Tomography: Prospective Blinded Human Inter-Observer Performance Evaluation

Importance

Extranodal extension (pENE) is a critical prognostic factor in oropharyngeal cancer (OPC) that drives therapeutic disposition. Determination of pENE from radiological imaging has been associated with high inter-observer variability. However, the impact of clinician specialty on human observer performance of imaging-detected extranodal extension (iENE) remains poorly understood.

Objective

To characterize the impact of clinician specialty on the accuracy of pre-operative iENE in human papillomavirus-positive (HPV+) OPC using computed tomography (CT) images.

Design Setting and Participants

This prospective observational human performance study analyzed pre-therapy CT images from 24 HPV+ OPC patients, with duplication of 6 scans (n=30) of which 21 were pathologically confirmed pENE. Thirty-four expert observers, including 11 radiologists, 12 surgeons, and 11 radiation oncologists, independently assessed these scans for iENE and reported human-detected radiologic criteria and observer confidence.

Main Outcomes and Measures

The primary outcomes included accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC), and Brier score for each physician, compared to ground-truth pENE. The significance of radiographic signs for prediction of pENE were determined through logistic regression analysis. Fleiss' kappa measured interobserver agreement, and Hanley-MacNeil AUC discrimination testing.

Results

Median accuracy across all specialties was 0.57 (95%CI 0.39 to 0.73), with no specialty showing discriminate performance greater than random estimation (median AUC 0.64, 95%CI 0.44 to 0.83). Significant differences between radiologists and surgeons in Brier scores (0.33 vs. 0.26, p < 0.01), radiation oncologists and surgeons in sensitivity (0.48 vs. 0.69, p > 0.1), and radiation oncologists and radiologists/surgeons in specificity (0.89 vs. 0.56, p > 0.1). Indistinct capsular contour and nodal necrosis were significant predictors of correct pENE status among all specialties. Interobserver agreement was weak for all the radiographic criteria, regardless of specialty (κ<0.6).

Conclusions and Relevance

Multiobserver testing shows physician discrimination of HPV+OPC pENE on pre-operative CT remains non-different than blind guessing, with high interrater variability and low diagnostic accuracy, regardless of clinician specialty. While minor differences in diagnostic performance among specialties are noted, they do not significantly affect the overall poor agreement and discrimination rates observed. The findings underscore the need for further research into automated detection systems or enhanced imaging techniques to improve the accuracy and reliability of iENE assessments in clinical practice.

Extranodal extension in oral squamous cell carcinoma: clinical and histopathological analysis

OBJECTIVE

Extranodal extension (ENE) is an important prognostic factor in oral squamous cell carcinoma (OSCC). This study aimed to evaluate the clinical significance of stratifying minor or major ENE in OSCC.

STUDY DESIGN

This retrospective cohort study included 75 patients who had undergone neck dissection for OSCC and were classified as pN+. ENE was measured using hematoxylin-eosin-stained specimens and stratified into major (ENEma, >2 mm) and minor (ENEmi, ≤2 mm) by distance. Their association with survival, locoregional relapse, and distant metastases were assessed.

RESULTS

Of 49 patients with pathological ENE, 23 had ENEmi, and 26 had ENEma. The 5-year overall survival (OS) was 38%, 66%, and 76% in the ENEma, ENEmi, and no ENE groups, respectively. Compared with no ENE, ENEma was associated with significantly decreased 5-year cumulative OS and disease-specific survival. ENEma was a risk factor for decreased OS (HR: 2.54, 95% CI: 1.04-6.18, P = .040) in the multivariable Cox regression analysis, and was associated with distant metastasis.

CONCLUSION

In patients with OSCC, ENEma is associated with a significantly poorer prognosis; therefore stratifying ENE is clinically relevant. ENEma may increase the risk of distant metastasis; therefore, new treatment modalities that contribute to distant metastasis control are required.

Extranodal extension in head and neck squamous cell carcinoma: need for accurate pretherapeutic staging to select optimum treatment and minimize toxicity

PURPOSE OF REVIEW

In 2017, the American Joint Committee on Cancer (AJCC) introduced the inclusion of extracapsular nodal extension (ENE) into the N staging of nonviral head and neck squamous cell carcinoma (HNSCC), while retaining the traditional N classification based on the number and sizes of metastatic nodes. The extent of ENE was further defined as microscopic ENE (ENEmi) and major ENE (ENEma) based on extent of disease beyond the nodal capsule (≤ or > 2 mm). This article reviews the evidence and progress made since these changes were introduced.

RECENT FINDINGS

The 'gold standard' for evaluation ENE is histopathologic examination, the current preferred primary treatment of patients with HNSCC is by radiation-based therapy ± chemotherapy or biotherapy. The current pretreatment staging is by imaging, which needs improved reliability of radiologic rENE assessment with reporting needs to consider both sensitivity and specificity (currently computed tomography images have high-specificity but low-sensitivity). Adjuvant chemotherapy is indicated for patients with ENEma to enhance disease control, whereas for patients with ENEmi, there is a need to assess the benefit of adjuvant chemotherapy. Evidence that the presence of pENE in HPV-positive oropharyngeal carcinoma is an independent prognostic factor and should be considered for inclusion in future AJCC editions has recently emerged.

SUMMARY

There remains a paucity of data on the reliability of imaging in the staging of rENE, more so the for the accurate assessment of ENEmi. Optimistic early results from use of artificial intelligence/deep learning demonstrate progress and may pave the way for better capabilities in tumor staging, treatment outcome prediction, resulting in improved survival outcomes.

Free-Breathing StarVIBE Sequence for the Detection of Extranodal Extension in Head and Neck Cancer: An Image Quality and Diagnostic Performance Study

(1) Background: This study aims to evaluate the image quality of abnormal cervical lymph nodes in head and neck cancer and the diagnostic performance of detecting extranodal extension (ENE) using free-breathing StarVIBE. (2) Methods: In this retrospective analysis, 80 consecutive head and neck cancer patients underwent StarVIBE before neck dissection at an academic center. Image quality was compared with conventional VIBE available for 28 of these patients. A total of 73 suspicious metastatic lymph nodes from 40 patients were found based on morphology and enhancement pattern on StarVIBE. Sensitivity (SN), specificity (SP), and odds ratios were calculated for each MR feature from StarVIBE to predict pathologic ENE. (3) Results: StarVIBE showed significantly superior image quality, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) for enlarged lymph nodes compared to VIBE. The MR findings of "invading adjacent planes" (SN, 0.54; SP, 1.00) and "matted nodes" (SN, 0.72; SP, 0.89) emerged as notable observations. The highest diagnostic performance was attained by combining these two features (SN, 0.93; SP, 0.89). (4) Conclusions: This study confirms that StarVIBE offers superior image quality for abnormal lymph nodes compared to VIBE, and it can accurately diagnose ENE by utilizing a composite MR criterion in head and neck cancer.

Prediction of extranodal extension in head and neck squamous cell carcinoma by CT images using an evolutionary learning model

BACKGROUND

Extranodal extension (ENE) in head and neck squamous cell carcinoma (HNSCC) correlates to poor prognoses and influences treatment strategies. Deep learning may yield promising performance of predicting ENE in HNSCC but lack of transparency and interpretability. This work proposes an evolutionary learning method, called EL-ENE, to establish a more interpretable ENE prediction model for aiding clinical diagnosis.

METHODS

There were 364 HNSCC patients who underwent neck lymph node (LN) dissection with pre-operative contrast-enhanced computerized tomography images. All the 778 LNs were divided into training and test sets with the ratio 8:2. EL-ENE uses an inheritable bi-objective combinatorial genetic algorithm for optimal feature selection and parameter setting of support vector machine. The diagnostic performances of the ENE prediction model and radiologists were compared using independent test datasets.

RESULTS

The EL-ENE model achieved the test accuracy of 80.00%, sensitivity of 81.13%, and specificity of 79.44% for ENE detection. The three radiologists achieved the mean diagnostic accuracy of 70.4%, sensitivity of 75.6%, and specificity of 67.9%. The features of gray-level texture and 3D morphology of LNs played essential roles in predicting ENE.

CONCLUSIONS

The EL-ENE method provided an accurate, comprehensible, and robust model to predict ENE in HNSCC with interpretable radiomic features for expanding clinical knowledge. The proposed transparent prediction models are more trustworthy and may increase their acceptance in daily clinical practice.