Search

Simple Summary: Diagnosing true progression (TP) versus radiation necrosis (RN) in brain metastases treated with stereotactic radiosurgery is a significant clinical challenge that can lead to delays of care or unnecessary neurosurgical procedures. We implemented a novel machine-learning framework that uses both multiparametic radiomics (mpRad) and tumor connectomics analysis to probe the textural properties and structural networks within radiographically progressive lesions, respectively. Our predictive model was able to distinguish histopathologically proven cases of TP from RN with excellent discrimination and may ultimately serve as a useful tool to inform clinical decision making. We introduce tumor connectomics, a novel MRI-based complex graph theory framework that describes the intricate network of relationships within the tumor and surrounding tissue, and combine this with multiparametric radiomics (mpRad) in a machine-learning approach to distinguish radiation necrosis (RN) from true progression (TP). Pathologically confirmed cases of RN vs. TP in brain metastases treated with SRS were included from a single institution. The region of interest was manually segmented as the single largest diameter of the T1 post-contrast (T1C) lesion plus the corresponding area of T2 FLAIR hyperintensity. There were 40 mpRad features and 6 connectomics features extracted, as well as 5 clinical and treatment factors. We developed an Integrated Radiomics Informatics System (IRIS) based on an Isomap support vector machine (IsoSVM) model to distinguish TP from RN using leave-one-out cross-validation. Class imbalance was resolved with differential misclassification weighting during model training using the IRIS. In total, 135 lesions in 110 patients were analyzed, including 43 cases (31.9%) of pathologically proven RN and 92 cases (68.1%) of TP. The top-performing connectomics features were three centrality measures of degree, betweenness, and eigenvector centralities. Combining these with the 10 top-performing mpRad features, an optimized IsoSVM model was able to produce a sensitivity of 0.87, specificity of 0.84, AUC-ROC of 0.89 (95% CI: 0.82–0.94), and AUC-PR of 0.94 (95% CI: 0.87–0.97). [ABSTRACT FROM AUTHOR]

Background: Improvements in systemic therapy continue to increase survival for patients with brain metastases. Updated dosimetric models are required to optimize long-term safety of stereotactic radiosurgery (SRS) for this indication. Methods: Patients at a single institution receiving SRS from December 2011 to December 2014 were retrospectively reviewed. Patients with radiographic progression of at least one lesion, and with at least 6 months of follow-up from the start of SRS were included. Grade 3 necrosis was defined as requiring surgical intervention. This data were combined with two additional published datasets to construct logistic models describing necrosis risk as a function of dose and volume. Results: From our institution, 294 brain metastases across 57 patients in 139 treatment plans met inclusion criteria. Primary histologies included non-small cell lung cancer (n = 19), melanoma (n = 13), breast carcinoma (n = 9), renal cell carcinoma (n = 7), and other (n = 9). Median follow-up from SRS of first cranial metastasis was 21.7 months (range: 6.3-56.6) and median overall survival was 25.6 months (range: 6.5-56.6). There were eight cases of Grade 1-2 and two cases of Grade 3 necrosis. As a useful clinical reference point, 20 cc of total brain receiving a single-fraction equivalent dose ≥14 Gy corresponded to 12.1% risk for Grade 1-3 (P < 0.003) and 3.4% risk for Grade 3 necrosis (P < 0.001). Conclusions: These results compare favorably with the QUANTEC brain tolerance estimates for radiosurgery, providing optimism for lower toxicity in the modern era. Additional studies are needed to determine dose tolerance parameters across a broad spectrum of patients. [ABSTRACT FROM AUTHOR]

• 57% of KRAS-mutated NSCLC brain mets patients experienced intracranial progression on sotorasib. • Median ICPFS from sotorasib initiation was 3 months for patients with active disease at baseline. • Median ICPFS from sotorasib initiation was 15 months for patients with stable disease at baseline. • Multidisciplinary discussion might still be warranted before starting sotorasib as monotherapy. Sotorasib has emerged as a treatment option for patients with KRAS-mutated non-small cell lung cancer (NSCLC); however, its effect in patients with brain metastases is not well described. We assessed the intracranial response of sotorasib in a retrospective case series of patients with brain metastases (BMs) at a single institution. Patients with KRAS-mutated NSCLC with BMs who received sotorasib at Mass General Brigham Hospitals were included. Patients were stratified into three groups: patients with active BM without local therapy within one month of sotorasib initiation (group 1), patients with active BM with local therapy (surgery or radiation) within one month of sotorasib initiation (group 2), and patients with stable BM (group 3). Intracranial progression-free survival (ICPFS) and overall survival (OS) were explored using Kaplan Meier curves that were compared through log-rank test. Thirty patients were included (five in group 1; seven in group 2; 18 in group 3). Mean age at sotorasib initiation was 60 years. Most (67 %) patients had between one and four BMs at sotorasib initiation. Median ICPFS was three months (95 % CI: 0– 7.7) from start of sotorasib for group 1, two months (0–5.7) for group 2, and 15 months (6.0–24.0) for group 3 (p-value = 0.02). Median OS was four months (1.9–6.1) for group 1, six months (0–13.7) for group 2, and 12 months (3.5–20.5) for group 3 (p-value = 0.13). 57 % of patients experienced intracranial progression, including 44 % of patients who had stable BM at sotorasib initiation. While sotorasib may have some intracranial activity, a multidisciplinary approach to BM therapy is still warranted, as are future studies with larger patient samples, controls, and extended follow-up. [ABSTRACT FROM AUTHOR]