Efficacy of FDG-PET for defining gross tumor volume of head and neck cancer

A comparison of methods for fully automatic segmentation of tumors and involved nodes in PET/CT of head and neck cancers

Target volume delineation is a vital but time-consuming and challenging part of radiotherapy, where the goal is to deliver sufficient dose to the target while reducing risks of side effects. For head and neck cancer (HNC) this is complicated by the complex anatomy of the head and neck region and the proximity of target volumes to organs at risk. The purpose of this study was to compare and evaluate conventional PET thresholding methods, six classical machine learning algorithms and a 2D U-Net convolutional neural network (CNN) for automatic gross tumor volume (GTV) segmentation of HNC in PET/CT images. For the latter two approaches the impact of single versus multimodality input on segmentation quality was also assessed. 197 patients were included in the study. The cohort was split into training and test sets (157 and 40 patients, respectively). Five-fold cross-validation was used on the training set for model comparison and selection. Manual GTV delineations represented the ground truth. Tresholding, classical machine learning and CNN segmentation models were ranked separately according to the cross-validation Sørensen-Dice similarity coefficient (Dice). PET thresholding gave a maximum mean Dice of 0.62, whereas classical machine learning resulted in maximum mean Dice scores of 0.24 (CT) and 0.66 (PET; PET/CT). CNN models obtained maximum mean Dice scores of 0.66 (CT), 0.68 (PET) and 0.74 (PET/CT). The difference in cross-validation Dice between multimodality PET/CT and single modality CNN models was significant (p ≤ 0.0001). The top-ranked PET/CT-based CNN model outperformed the best-performing thresholding and classical machine learning models, giving significantly better segmentations in terms of cross-validation and test set Dice, true positive rate, positive predictive value and surface distance-based metrics (p ≤ 0.0001). Thus, deep learning based on multimodality PET/CT input resulted in superior target coverage and less inclusion of surrounding normal tissue.

Deep learning-based auto-delineation of gross tumour volumes and involved nodes in PET/CT images of head and neck cancer patients

PURPOSE

Identification and delineation of the gross tumour and malignant nodal volume (GTV) in medical images are vital in radiotherapy. We assessed the applicability of convolutional neural networks (CNNs) for fully automatic delineation of the GTV from FDG-PET/CT images of patients with head and neck cancer (HNC). CNN models were compared to manual GTV delineations made by experienced specialists. New structure-based performance metrics were introduced to enable in-depth assessment of auto-delineation of multiple malignant structures in individual patients.

METHODS

U-Net CNN models were trained and evaluated on images and manual GTV delineations from 197 HNC patients. The dataset was split into training, validation and test cohorts (n= 142, n = 15 and n = 40, respectively). The Dice score, surface distance metrics and the new structure-based metrics were used for model evaluation. Additionally, auto-delineations were manually assessed by an oncologist for 15 randomly selected patients in the test cohort.

RESULTS

The mean Dice scores of the auto-delineations were 55%, 69% and 71% for the CT-based, PET-based and PET/CT-based CNN models, respectively. The PET signal was essential for delineating all structures. Models based on PET/CT images identified 86% of the true GTV structures, whereas models built solely on CT images identified only 55% of the true structures. The oncologist reported very high-quality auto-delineations for 14 out of the 15 randomly selected patients.

CONCLUSIONS

CNNs provided high-quality auto-delineations for HNC using multimodality PET/CT. The introduced structure-wise evaluation metrics provided valuable information on CNN model strengths and weaknesses for multi-structure auto-delineation.

Gross tumour volume comparison in oropharynx carcinomas using different intelligent imaging software. A retrospective analysis

Purpose

To compare gross tumour volume (GTV) in oropharynx carcinomas using different intelligent imaging software and to evaluate which method is more reliable for tumour volume definition in comparison with 3D ProSoma software.

Material and methods

We retrospectively studied 32 patients with histopathologically confirmed oropharynx carcinomas on dual-source computed tomography (CT) (all patients underwent multislice CT examination after applying 75 ml iodinated non-ionic contrast media). One radiologist calculated the tumour volume – manually measuring tumour length (L), width (W), and height (H) – and then calculated the tumour volume using the formula 0.5236 × L × W × H. The other radiologist used the syngo.CT-Liver-Analysis software to calculate the tumour volumes. Both volume measuring methods were compared with the 3D ProSoma software, which is used by radiotherapists to calculate tumour volumes. Graphpad Prism software was used for statistical data.

Results

syngo.CT-Liver-Analysis software for gross tumour volume determination has greater reliability than the standard manual method with Syngo Plaza in comparison with the 3D ProSoma software.

Conclusions

syngo.CT-Liver-Analysis software is a reliable tool for GTV calculation, with a high correlation score, like that of radiotherapeutic 3D ProSoma software.

18F-FDG PET/CT in staging and delineation of radiotherapy volume for head and neck cancer

PURPOSE

The aim is to investigate the use of ¹⁸F-FDG (fluorine-18 fluorodeoxyglucose) PET/CT in head and neck cancer (HNC) staging and its effect on the therapeutic strategy and radiotherapy (RT) planning.

METHODS AND MATERIALS

One hundred patients with HNC were included. Primary tumor sites: 18% oral cavity, 20% oropharynx, 12% hypopharynx, 11% nasopharynx, 37% larynx, 2% paranasal sinuses. Patients were staged according to the American Joint Committee of Cancer 7th edition. Stage: 5% stage I, 7% stage II, 14% stage III, 61% stage IVA, 7% stage IVB and 6% stage IVC. A contrast-enhanced CT and a ¹⁸F-FDG PET/CT acquired under RT position were performed. Both exams were compared to analyze patients' staging reclassification. Changes in therapeutic strategy were analyzed.

RESULTS

¹⁸F-FDG PET/CT detected 6 distant metastases and treatment intention changed to palliative. Eight synchronous tumors were detected; one received palliative treatment. ¹⁸F-FDG PET/CT reclassified cTNM staging in 27patients. Tumor extension changed in 28 (14% up-staged; 14% down-staged), implying a change in GTV (Gross Tumor Volume) delineation. Nodal detection was reclassified in 47 patients: 8 patients down-staged (N2C to N2A/N2B/N1) and 2 were false positive. Nineteen patients were false negatives and 5 staged as N+(N1/N2A/N2B) turned out into N2C. These staging modifications imply adapting the nodal volume to be irradiated.

CONCLUSIONS

¹⁸F-FDG PET/CT reclassification was higher than 10% in almost all categories studied (cTNM, tumor extension and nodal disease) and detects more metastases and synchronous tumors than conventional studies, which has an impact on the therapeutic patient management and RT planning.

Molecular Imaging-Guided Radiotherapy for the Treatment of Head-and-Neck Squamous Cell Carcinoma: Does it Fulfill the Promises?

With the routine use of intensity modulated radiation therapy for the treatment of head-and-neck squamous cell carcinoma allowing highly conformed dose distribution, there is an increasing need for refining both the selection and the delineation of gross tumor volumes (GTV). In this framework, molecular imaging with positron emission tomography and magnetic resonance imaging offers the opportunity to improve diagnostic accuracy and to integrate tumor biology mainly related to the assessment of tumor cell density, tumor hypoxia, and tumor proliferation into the treatment planning equation. Such integration, however, requires a deep comprehension of the technical and methodological issues related to image acquisition, reconstruction, and segmentation. Until now, molecular imaging has had a limited value for the selection of nodal GTV, but there are increasing evidences that both FDG positron emission tomography and diffusion-weighted magnetic resonance imaging has a potential value for the delineation of the primary tumor GTV, effecting on dose distribution. With the apprehension of the heterogeneity in tumor biology through molecular imaging, growing evidences have been collected over the years to support the concept of dose escalation/dose redistribution using a planned heterogeneous dose prescription, the so-called "dose painting" approach. Validation trials are ongoing, and in the coming years, one may expect to position the dose painting approach in the armamentarium for the treatment of patients with head-and-neck squamous cell carcinoma.

Overall and disease-free survival outcomes of patients receiving intensity-modulated radiation therapy (IMRT) with PET-CT-based planning for cancers of the head and neck

Introduction

Combined modality treatment regimens have provided modest gains in locoregional control rates of cancers of the head and neck (HNC), and intensity-modulated radiation therapy (IMRT) has gained widespread use. The methodology for determining contours of the gross tumour volume (GTV) in the radiation treatment plan is often based on combined anatomic and metabolic data from positron emission tomography–computed tomography (PET-CT). This study aimed to retrospectively evaluate the overall survival and disease-free survival outcomes of patients with HNC who received definitive IMRT with or without chemotherapy, planned with PET-CT.

Materials and Methods

A total of 1,200 patients underwent treatment for HNC during the study period, from 1 January 2002 to 31 December 2010. Of those, 261 cases had evaluable data that met the inclusion criteria for the study. The incidence and timing of locoregional recurrence, distant metastatic disease, new primary malignancies and death were evaluated retrospectively. Overall and disease-free survival (survival to time of first recurrence) were determined by the life table method. Incidence of distance metastatic disease and additional cancers were also studied.

Results

Median follow-up from treatment initiation was 26·4 months (range 1·2–84·7 months). Overall survival and disease-free survival rates were 0·883 and 0·791, respectively, at 1 year; 0·793 and 0·688, respectively, at 2 years; and 0·732 and 0·619, respectively, at 3 years. The cumulative risk of recurrence was 22·6% at 3 years following definitive IMRT and the median time to recurrence was 345 days. There was an overall low incidence of distant metastatic disease (3·07%) and additional cancers (8·05%).

Conclusion

Overall and disease-free survival outcomes of a large cohort of HNC patients treated with definitive IMRT radiotherapy following treatment planning with PET-CT shows a similar high level of disease control and mortality rate as previously published outcome studies of shorter terms and/or smaller numbers of patients.