Pre-treatment factors associated with detecting additional brain metastases at stereotactic radiosurgery

Radiological Kinetics of Brain Metastases and Clinical Implications for Patients Treated With Stereotactic Radiosurgery

AIMS

Select patients with brain metastases receive stereotactic radiosurgery (SRS) with the objective of improving survival and intracranial disease control. Brain metastases number and volume are prognostic factors used to inform patient selection. The aim of this study was to assess the rate of change of brain metastases size and number (growth kinetics) between the diagnostic and day of SRS magnetic resonance imaging (MRI) scans.

MATERIALS AND METHODS

All patients treated with Gamma Knife SRS between October 2015 and April 2017 were included in this single-centre retrospective analysis. Brain metastases number and diameter were recorded at diagnosis and treatment. For patients with multiple brain metastases, the largest lesion was the index lesion. Distant intracranial control and overall survival were reported from the date of SRS.

RESULTS

In total, 146 patients received 156 episodes of SRS. The median interval between diagnostic and SRS MRI was 20 days (range 1-68). Interval growth in the index lesion of at least 3 mm or the development of a new brain metastasis was noted in 60.2% of patients. This was associated with age less than 60 years (P = 0.001), Eastern Cooperative Oncology Group (ECOG) performance status 2 or above (P = 0.04), non-small cell lung carcinoma (NSCLC) (P = 0.03) or melanoma histologies (P = 0.05) and uncontrolled extracranial disease (P = 0.05). These patients were also more likely to develop distant intracranial recurrence (P = 0.046). Clinically significant growth was not associated with scan interval or differences in overall survival. The Kaplan-Meier estimate of probability of survival at 12 months was 59.3% (95% confidence interval 46.7-75.2%) for all patients.

CONCLUSION

Intracranial progression between diagnosis and day of SRS is common. Risk factors are uncontrolled extracranial disease, poorer performance status, NSCLC or melanoma histologies and age less than 60 years. These patients would benefit from an MRI closer to treatment to inform patient selection and target delineation for SRS planning.

A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery

Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients' data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.