Purpose: Treatment effect or radiation necrosis after stereotactic radiosurgery (SRS) for brain metastases is a common phenomenon often indistinguishable from true progression. Radiomics is an emerging field that promises to improve on conventional imaging. In this study, we sought to apply a radiomics-based prediction model to the problem of diagnosing treatment effect after SRS. Methods and Materials: We included patients in the Johns Hopkins Health System who were treated with SRS for brain metastases who subsequently underwent resection for symptomatic growth. We also included cases of likely treatment effect in which lesions grew but subsequently regressed spontaneously. Lesions were segmented semiautomatically on preoperative T1 postcontrast and T2 fluid-attenuated inversion recovery magnetic resonance imaging, and radiomic features were extracted with software developed in-house. Top-performing features on univariate logistic regression were entered into a hybrid feature selection/classification model, IsoSVM, with parameter optimization and further feature selection performed using leave-one-out cross-validation. Final model performance was assessed by 10-fold cross-validation with 100 repeats. All cases were independently reviewed by a board-certified neuroradiologist for comparison. Results: We identified 82 treated lesions across 66 patients, with 77 lesions having pathologic confirmation. There were 51 radiomic features extracted per segmented lesion on each magnetic resonance imaging sequence. An optimized IsoSVM classifier based on top-ranked radiomic features had sensitivity and specificity of 65.38% and 86.67%, respectively, with an area under the curve of 0.81 on leave-one-out cross-validation. Only 73% of cases were classifiable by the neuroradiologist, with a sensitivity of 97% and specificity of 19%. Conclusions: Radiomics holds promise for differentiating between treatment effect and true progression in brain metastases treated with SRS. A predictive model built on radiomic features from an institutional cohort performed well on cross-validation testing. These results warrant further validation in independent datasets. Such work could prove invaluable for guiding management of individual patients and assessing outcomes of novel interventions.

No abstract

OBJECTIVE There are a variety of salvage options available for patients with brain metastases who experience local failure after stereotactic radiosurgery (SRS). These options include resection, whole-brain radiation therapy, laser thermoablation, and repeat SRS. There is little data on the safety and efficacy of repeat SRS following local failure of a prior radiosurgical procedure. This study evaluates the clinical outcomes and dosimetric characteristics of patients who experienced tumor recurrence and were subsequently treated with repeat SRS. METHODS Between 2002 and 2015, 32 patients were treated with repeat SRS for local recurrence of ≥ 1 brain metastasis following initial SRS treatment. The Kaplan-Meier method was used to estimate time-to-event outcomes including overall survival (OS), local failure, and radiation necrosis. Cox proportional hazards analysis was performed for predictor variables of interest for each outcome. Composite dose-volume histograms were constructed for each reirradiated lesion, and these were then used to develop a predictive dosimetric model for radiation necrosis. RESULTS Forty-six lesions in 32 patients were re-treated with a second course of SRS after local failure. A median dose of 20 Gy (range 14-22 Gy) was delivered to the tumor margin at the time of repeat SRS. Local control at 1 year was 79% (95% CI 67%-94%). Estimated 1-year OS was 70% (95% CI 55%-88%). Twelve patients had died at the most recent follow-up, with 8/12 patients experiencing neurological death (as described in Patchell et al.). Eleven of 46 (24%) lesions in 11 separate patients treated with repeat SRS were associated with symptomatic radiation necrosis. Freedom from radiation necrosis at 1 year was 71% (95% CI 57%-88%). Analysis of dosimetric data revealed that the volume of a lesion receiving 40 Gy (V) was the most predictive factor for the development of radiation necrosis (p = 0.003). The following V thresholds were associated with 10%, 20%, and 50% probabilities of radiation necrosis, respectively: 0.28 cm (95% CI 3%-28%), 0.76 cm (95% CI 9%-39%), 1.60 cm (95% CI 26%-74%). CONCLUSIONS Repeat SRS appears to be an effective salvage option for patients with brain metastases experiencing local failure following initial SRS treatment. This series demonstrates durable local control and, although rates of radiation necrosis are significant, repeat SRS may be indicated for select cases of local disease recurrence. Because the V is predictive of radiation necrosis, limiting this value during treatment planning may allow for a reduction in radiation necrosis rates.

No abstract

BackgroundTo determine the clinical benefits of systemic targeted agents across multiple histologies after stereotactic radiosurgery (SRS) for brain metastases.MethodsBetween 2000 and 2013, 737 patients underwent upfront SRS for brain metastases. Patients were stratified by whether or not they received targeted agents with SRS. 167 (23%) received targeted agents compared to 570 (77%) that received other available treatment options. Time to event data were summarized using Kaplan-Meier plots, and the log rank test was used to determine statistical differences between groups.ResultsPatients who received SRS with targeted agents vs those that did not had improved overall survival (65% vs. 30% at 12 months, p < 0.0001), improved freedom from local failure (94% vs 90% at 12 months, p = 0.06), improved distant failure-free survival (32% vs. 18% at 12 months, p = 0.0001) and improved freedom from whole brain radiation (88% vs. 77% at 12 months, p = 0.03). Improvement in freedom from local failure was driven by improvements seen in breast cancer (100% vs 92% at 12 months, p < 0.01), and renal cell cancer (100% vs 88%, p = 0.04). Multivariate analysis revealed that use of targeted agents improved all cause mortality (HR = 0.6, p < 0.0001).ConclusionsTargeted agent use with SRS appears to improve survival and intracranial outcomes.