TL;DR
This study explores the use of Dual-Energy Computed Tomography (DECT) to predict the aggressiveness of Oral Tongue Squamous Cell Carcinoma (OTSCC) before surgery. Researchers found that specific quantitative DECT parameters, particularly those measured during the venous phase, can effectively differentiate between early and advanced tumor stages, predict histological differentiation, and identify lymph node metastasis, offering a promising non-invasive tool for treatment planning.
What Problem Does This Paper Address?
Oral Tongue Squamous Cell Carcinoma (OTSCC) is a significant subtype of oral cancer, known for its aggressiveness and potential for recurrence. Accurately assessing its biological aggressiveness before treatment is crucial for tailoring surgical approaches and avoiding overtreatment or undertreatment. Current methods rely on histological analysis after resection, which can lead to delays and suboptimal patient management. While imaging techniques like MRI have shown promise, they have limitations such as contraindications and variability in analysis. Conventional CT offers some morphological information but lacks specificity for certain aspects like lymph node metastasis. This paper investigates whether DECT, a more advanced CT technique, can provide quantitative parameters that correlate with OTSCC aggressiveness, thereby improving preoperative assessment.
Methods
The study employed a prospective, single-center design involving 93 patients with pathologically confirmed OTSCC. Before surgical resection, all patients underwent dual-phase (arterial and venous) DECT scans. Quantitative parameters were extracted from these scans, including:
- λHu (slope of the spectral Hounsfield unit curve): Reflects the material composition and energy dependence of X-ray attenuation.
- Normalized Iodine Concentration (nIC): Quantifies iodine uptake, indicative of vascularity and contrast enhancement.
- Normalized Effective Atomic Number (nZeff): Represents the average atomic number of the tissue, related to its material composition.
- Normalized Electron Density (nRho): Measures the electron density of the tissue.
These parameters were normalized to the contralateral carotid artery to account for variations in contrast delivery. The DECT parameters were then compared between different groups based on:
- Pathologic TNM stages (early vs. advanced)
- Histologic differentiation (well-differentiated vs. poorly differentiated)
- Lymph node metastasis (present vs. absent)
- Perineural invasion (PNI) (present vs. absent)
Statistical analysis involved logistic regression to identify independent predictors of aggressiveness and Receiver Operating Characteristic (ROC) curve analysis to evaluate the diagnostic performance of these parameters.
Datasets
The study utilized a dataset of 93 patients with pathologically confirmed OTSCC. This dataset was collected prospectively between August 2019 and March 2021. The paper does not introduce a new benchmark dataset but rather uses its collected patient data for analysis.
Key Insights and Ideas
- Venous Phase Dominance: Quantitative DECT parameters measured during the venous phase (VP) consistently showed higher diagnostic ability for predicting tumor aggressiveness compared to those measured during the arterial phase (AP). This suggests that the retention of contrast in the tumor interstitial space during the venous phase is more sensitive to tumor characteristics than the initial arterial filling.
- Hypoxia and Vascularity Link: Lower DECT parameter values (λHu, nIC, nZeff) in advanced-stage and poorly differentiated OTSCCs are hypothesized to be linked to increased tumor hypoxia and reduced vascularity, which are common in more aggressive tumors.
- DECT as a Non-Invasive Biomarker: The study demonstrates that DECT can provide quantitative, non-invasive biomarkers that correlate with key prognostic factors of OTSCC, potentially reducing the need for invasive histological assessment prior to treatment decisions.
- PNI Remains Elusive: Unlike other aggressiveness indicators, quantitative DECT parameters did not show a significant difference between OTSCCs with and without perineural invasion (PNI), suggesting PNI might require different imaging approaches or parameters for prediction.
Key Results
- Tumor Stage Prediction: λHu in the venous phase (VP) was identified as an independent predictor of tumor stage, with an Area Under the Curve (AUC) of 0.80.
- Histologic Differentiation: nIC in VP was an independent predictor of histologic differentiation, achieving an AUC of 0.78.
- Lymph Node Metastasis: λHu in VP was an independent predictor of lymph node status, with an AUC of 0.74.
- Parameter Trends: Advanced-stage OTSCCs generally showed lower mean λHu and nIC in AP, and lower λHu, nZeff, and nIC in VP compared to early-stage lesions. Poorly differentiated lesions also exhibited lower λHu and nIC compared to well-differentiated ones.
- Lymph Node Metastasis: Lesions with lymph node metastasis had lower mean λHu and nIC in VP compared to those without metastasis.
Limitations and Future Work
- Single-Center Study: The study was conducted at a single center, which could introduce potential biases. The authors recommend large-sample, multicenter studies for validation.
- Image-Histology Mismatch: There was an inherent limitation in precisely matching CT image slices with corresponding histological slices, which could affect the correlation degree.
- Limited Scope: While the study assessed pathologic stages, histologic differentiation, and lymph node status, it did not explore other aggressive features like extranodal extension, which could be a focus for future research.
- PNI Prediction: The lack of significant findings for PNI suggests further investigation is needed to identify DECT parameters or combinations that can predict this aggressive feature.
- Clinical Utility: The authors suggest that DECT parameters could be valuable for deciding treatment strategies and determining prognosis, implying future work should focus on integrating these findings into clinical decision-making pathways.
Summary updated