Vision Transformer-Based Multilabel Survival Prediction for Oropharynx Cancer After Radiation Therapy.

PURPOSE: A reliable and comprehensive cancer prognosis model for oropharyngeal cancer (OPC) could better assist in personalizing treatment. In this work, we developed a vision transformer-based (ViT-based) multilabel model with multimodal input to learn complementary information from available pretreatment data and predict multiple associated endpoints for radiation therapy for patients with OPC. METHODS AND MATERIALS: A publicly available data set of 512 patients with OPC was used for both model training and evaluation. Planning computed tomography images, primary gross tumor volume masks, and 16 clinical variables representing patient demographics, diagnosis, and treatment were used as inputs. To extract deep image features with global attention, we used a ViT module. Clinical variables were concatenated with the learned image features and fed into fully connected layers to incorporate cross-modality features. To learn the mapping between the features and correlated survival outcomes, including overall survival, local failure-free survival, regional failure-free survival, and distant failure-free survival, we employed 4 multitask logistic regression layers. The proposed model was optimized by combining the multitask logistic regression negative-log likelihood losses of different prediction targets. RESULTS: We employed the C-index and area under the curve metrics to assess the performance of our model for time-to-event prediction and time-specific binary prediction, respectively. Our proposed model outperformed corresponding single-modality and single-label models on all prediction labels, achieving C-indices of 0.773, 0.765, 0.776, and 0.773 for overall survival, local failure-free survival, regional failure-free survival, and distant failure-free survival, respectively. The area under the curve values ranged between 0.799 and 0.844 for different tasks at different time points. Using the medians of predicted risks as the thresholds to identify high-risk and low-risk patient groups, we performed the log-rank test, the results of which showed significantly larger separations in different event-free survivals. CONCLUSION: We developed the first model capable of predicting multiple labels for OPC simultaneously. Our model demonstrated better prognostic ability for all the prediction targets compared with corresponding single-modality models and single-label models.

Low rates of contralateral neck failure in unilaterally treated oropharyngeal squamous cell carcinoma with prospectively defined criteria of lateralization.

BACKGROUND: Unilateral radiotherapy (RT) of oropharyngeal carcinomas is accepted for patients with lateralized primary and low-volume nodal disease. Utilizing prospectively defined criteria of laterality and staging positron emission tomography (PET)/CT, we studied outcomes in patients with advanced-stage oropharyngeal cancer undergoing unilateral RT. METHODS: Thirty-seven patients with oropharyngeal tumors >1 cm from midline regardless of node status underwent unilateral RT and were followed prospectively. Patient characteristics: T1 = 11; T2 = 22; T3 = 4; N0 = 3; N1 = 9; N2a = 3; N2b = 21; and Nx = 1. Dosimetry were determined and weekly National Comprehensive Cancer Network (NCCN) distress thermometer data were collected. RESULTS: At median follow-up of 32 months, 3-year locoregional control, contralateral regional failure, distant metastasis-free survival, and disease-free survival were 96%, 0%, 7%, and 93%, respectively. CONCLUSION: Low rates of contralateral neck failure are demonstrated utilizing prospectively defined criteria for unilateral RT. The tolerances of contralateral organs are respected and patients report low to moderate levels of distress throughout treatment.

Selective intraarterial chemoradiation therapy for oropharyngeal carcinoma with high-dose cisplatin.

PURPOSE: Cisplatin has shown a high tumor response rate among head and neck carcinomas, and the tumor response is related to the cisplatin dosage. The purpose of this study was to evaluate the efficacy and toxicity of selective intraarterial chemoradiation therapy for oropharyngeal carcinomas with high-dose cisplatin. MATERIALS AND METHODS: This retrospective study consisted of 21 patients with oropharyngeal carcinoma, stages II-IVB, in whom intraarterial chemoradiation therapy was performed between 2000 and 2008. All patients were given two courses of selective intraarterial infusions of cisplatin (300 mg/m(2)), systemic chemotherapy with 5-fluorouracil, and simultaneous radiation therapy (58-61 Gy/30 fractions), with a 1-week rest period. RESULTS: The 2-year overall survival rate of the 15 patients who completed the therapeutic regimen was 71.3%. The 2-year locoregional control rate and disease-free survival rate were 95.0% and 67.7%, respectively. CONCLUSION: Selective intraarterial high-dose cisplatin chemotherapy with concomitant radiation therapy shows results similar to those of original methods in terms of survival and locoregional control with a reduction in the number of procedure times.

Oral cavity and oropharyngeal squamous cell cancer: key imaging findings for staging and treatment planning.

The imaging findings in squamous cell carcinoma (SCC) of the oral cavity and oropharynx vary widely, depending on the site of origin of the primary tumor and the extent of its involvement of other regions. Knowledge of the complex anatomy of the oral cavity and oropharynx, as well as the most common routes by which SCC spreads from various anatomic sites, allows the radiologist to accurately determine the extent of disease and help clinicians plan appropriate treatment. SCCs that originate in the oral cavity tend to behave differently than those that originate in the oropharynx, with the latter group exhibiting more aggressive growth. Furthermore, primary tumors in certain anatomic subsites within the oral cavity or oropharynx have a greater propensity to spread by direct extension along muscle, bone, or neurovascular bundles or to be disseminated along lymphatic drainage pathways to regional or distant nodes. Imaging findings of deep muscular, neurovascular, osseous, or nodal involvement are indicative of an advanced stage of disease for which management options are limited.

Prognostic significance of computed tomography in tumors of the oral cavity and oropharynx treated with neoadjuvant chemotherapy.

BACKGROUND: Despite its high response rate, the use of neoadjuvant chemotherapy remains controversial. Pretherapeutic identification of subgroups of patients who are likely to respond to chemotherapy is of the utmost importance. PURPOSE: In this study, we have attempted to determine the relationship between specific radiological parameters and the response to neoadjuvant chemotherapy. In addition, we have determined if these parameters could yield prognostic information on recurrence and/or survival. PATIENTS AND METHODS: Fifty-four patients with a squamous cell carcinoma of the oral cavity or base of tongue who had had a contrast-enhanced CT scan and neoadjuvant chemotherapy were included in this analysis. All clinical, radiological, surgical, histological, and radiotherapeutical parameters as well as the follow-up data were analyzed by a chi-square test. The method of Kaplan-Meyer was used to determine disease-free intervals and crude survival. The log-rank method was used for testing differences in local failures and survival. RESULTS: Twenty-eight patients were classified as having isodense nodes and 20 patients as having hypodense nodes. Nodal density was not related to tumor size or primary site. N stage was not correlated with the density of the nodes. Patients with hypodense nodes had a significantly lower disease-free interval and survival than patients with isodense nodes. The relation between overall response to chemotherapy and the hypodensity of the nodes didn't reach a significant level. CONCLUSION: No relation was found between overall response to chemotherapy and N-stage or tumor density. Disease-free interval and crude survival was strongly related to response to chemotherapy.