New Automated Method for Lung Functional Volumes Delineation with Lung Perfusion PET/CT Imaging

Prediction of Acute Radiation-Induced Lung Toxicity After Stereotactic Body Radiation Therapy Using Dose-Volume Parameters From Functional Mapping on Gallium 68 Perfusion Positron Emission Tomography/Computed Tomography

PURPOSE

The aim of this work was to compare anatomic and functional dose-volume parameters as predictors of acute radiation-induced lung toxicity (RILT) in patients with lung tumors treated with stereotactic body radiation therapy.

METHODS AND MATERIALS

Fifty-nine patients treated with stereotactic body radiation therapy were prospectively included. All patients underwent gallium 68 lung perfusion positron emission tomography (PET)/computed tomography (CT) imaging before treatment. Mean lung dose (MLD) and volumes receiving x Gy (VxGy, 5-30 Gy) were calculated in 5 lung volumes: the conventional anatomic volume (AV) delineated on CT images, 3 lung functional volumes (FVs) defined on lung perfusion PET imaging (FV50%, FV70%, and FV90%; ie, the minimal volume containing 50%, 70%, and 90% of the total activity within the AV), and a low FV (LFV; LFV = AV - FV90%). The primary endpoint of this analysis was grade ≥2 acute RILT at 3 months as assessed with National Cancer Institute Common Terminology Criteria for Adverse Events version 5. Dose-volume parameters in patients with and without acute RILT were compared. Receiver operating characteristic curves assessing the ability of dose-volume parameters to discriminate between patients with and without acute RILT were generated, and area under the curve (AUC) values were calculated.

RESULTS

Of the 59 patients, 10 (17%) had grade ≥2 acute RILT. The MLD and the VxGy in the AV and LFV were not statistically different between patients with and without acute RILT (P > .05). All functional parameters were significantly higher in acute RILT patients (P < .05). AUC values (95% CI) for MLD AV, LFV, FV50%, FV70%, and FV90% were 0.66 (0.46-0.85), 0.60 (0.39-0.80), 0.77 (0.63-0.91), 0.77 (0.64-0.91), and 0.75 (0.58-0.91), respectively. AUC values for V20Gy AV, LFV, FV50%, FV70%, and FV90% were 0.65 (0.44-0.87), 0.64 (0.46-0.83), 0.82 (0.69-0.95), 0.81 (0.67-0.96), and 0.75 (0.57-0.94), respectively.

CONCLUSIONS

The predictive value of PET perfusion-based functional parameters outperforms the standard CT-based dose-volume parameters for the risk of grade ≥2 acute RILT. Functional parameters could be useful for guiding radiation therapy planning and reducing the risk of acute RILT.

Optimisation of the air fraction correction for lung PET/CT: addressing resolution mismatch

BACKGROUND

Increased pulmonary [Formula: see text]F-FDG metabolism in patients with idiopathic pulmonary fibrosis, and other forms of diffuse parenchymal lung disease, can predict measurements of health and lung physiology. To improve PET quantification, voxel-wise air fractions (AF) determined from CT can be used to correct for variable air content in lung PET/CT. However, resolution mismatches between PET and CT can cause artefacts in the AF-corrected image.

METHODS

Three methodologies for determining the optimal kernel to smooth the CT are compared with noiseless simulations and non-TOF MLEM reconstructions of a patient-realistic digital phantom: (i) the point source insertion-and-subtraction method, [Formula: see text]; (ii) AF-correcting with varyingly smoothed CT to achieve the lowest RMSE with respect to the ground truth (GT) AF-corrected volume of interest (VOI), [Formula: see text]; iii) smoothing the GT image to match the reconstruction within the VOI, [Formula: see text]. The methods were evaluated both using VOI-specific kernels, and a single global kernel optimised for the six VOIs combined. Furthermore, [Formula: see text] was implemented on thorax phantom data measured on two clinical PET/CT scanners with various reconstruction protocols.

RESULTS

The simulations demonstrated that at [Formula: see text] iterations (200 i), the kernel width was dependent on iteration number and VOI position in the lung. The [Formula: see text] method estimated a lower, more uniform, kernel width in all parts of the lung investigated. However, all three methods resulted in approximately equivalent AF-corrected VOI RMSEs (<10%) at [Formula: see text]200i. The insensitivity of AF-corrected quantification to kernel width suggests that a single global kernel could be used. For all three methodologies, the computed global kernel resulted in an AF-corrected lung RMSE <10%  at [Formula: see text]200i, while larger lung RMSEs were observed for the VOI-specific kernels. The global kernel approach was then employed with the [Formula: see text] method on measured data. The optimally smoothed GT emission matched the reconstructed image well, both within the VOI and the lung background. VOI RMSE was <10%, pre-AFC, for all reconstructions investigated.

CONCLUSIONS

Simulations for non-TOF PET indicated that around 200i were needed to approach image resolution stability in the lung. In addition, at this iteration number, a single global kernel, determined from several VOIs, for AFC, performed well over the whole lung. The [Formula: see text] method has the potential to be used to determine the kernel for AFC from scans of phantoms on clinical scanners.