A Dose-Response Model of Local Tumor Control Probability After Stereotactic Radiosurgery for Brain Metastases Resection Cavities

Factors associated with the local control of brain metastases: a systematic search and machine learning application

BACKGROUND

Enhancing Local Control (LC) of brain metastases is pivotal for improving overall survival, which makes the prediction of local treatment failure a crucial aspect of treatment planning. Understanding the factors that influence LC of brain metastases is imperative for optimizing treatment strategies and subsequently extending overall survival. Machine learning algorithms may help to identify factors that predict outcomes.

METHODS

This paper systematically reviews these factors associated with LC to select candidate predictor features for a practical application of predictive modeling. A systematic literature search was conducted to identify studies in which the LC of brain metastases is assessed for adult patients. EMBASE, PubMed, Web-of-Science, and the Cochrane Database were searched up to December 24, 2020. All studies investigating the LC of brain metastases as one of the endpoints were included, regardless of primary tumor type or treatment type. We first grouped studies based on primary tumor types resulting in lung, breast, and melanoma groups. Studies that did not focus on a specific primary cancer type were grouped based on treatment types resulting in surgery, SRT, and whole-brain radiotherapy groups. For each group, significant factors associated with LC were identified and discussed. As a second project, we assessed the practical importance of selected features in predicting LC after Stereotactic Radiotherapy (SRT) with a Random Forest machine learning model. Accuracy and Area Under the Curve (AUC) of the Random Forest model, trained with the list of factors that were found to be associated with LC for the SRT treatment group, were reported.

RESULTS

The systematic literature search identified 6270 unique records. After screening titles and abstracts, 410 full texts were considered, and ultimately 159 studies were included for review. Most of the studies focused on the LC of the brain metastases for a specific primary tumor type or after a specific treatment type. Higher SRT radiation dose was found to be associated with better LC in lung cancer, breast cancer, and melanoma groups. Also, a higher dose was associated with better LC in the SRT group, while higher tumor volume was associated with worse LC in this group. The Random Forest model predicted the LC of brain metastases with an accuracy of 80% and an AUC of 0.84.

CONCLUSION

This paper thoroughly examines factors associated with LC in brain metastases and highlights the translational value of our findings for selecting variables to predict LC in a sample of patients who underwent SRT. The prediction model holds great promise for clinicians, offering a valuable tool to predict personalized treatment outcomes and foresee the impact of changes in treatment characteristics such as radiation dose.

Immune microenvironment remodeling after radiation of a progressing brain metastasis

Radiation is commonly used in the treatment of many cancers. However, its effects on anti-tumor immune responses are incompletely understood. Here, we present a detailed immunological analysis of two tumors from a patient with multiple non-small cell lung cancer metastases to the brain. One tumor was resected without treatment; the second was irradiated to a total dose of 30 Gy and resected following further progression. Comprehensive single-cell analysis reveals a substantially reduced immune cell fraction in the irradiated tumor, including the depletion of tissue-resident macrophages and infiltration of pro-inflammatory monocytes. Despite the presence of similar somatic mutations in both tumors, radiation is associated with the depletion of exhausted, tumor-resident T cell clones and their replacement by circulating clones unlikely to contribute to tumor-specific immunity. These results provide insight into the local effects of radiation on anti-tumor immunity and raise important considerations for the combination of radiation and immunotherapy.

Predicting the effect of indirect cell kill in the treatment of multiple brain metastases via single-isocenter/multitarget volumetric modulated arc therapy stereotactic radiosurgery

Purpose

Due to spatial uncertainty, patient setup errors are of major concern for radiosurgery of multiple brain metastases (m‐bm) when using single‐isocenter/multitarget (SIMT) volumetric modulated arc therapy (VMAT) techniques. However, recent clinical outcome studies show high rates of tumor local control for SIMT‐VMAT. In addition to direct cell kill (DCK), another possible explanation includes the effects of indirect cell kill (ICK) via devascularization for a single dose of 15 Gy or more and by inducing a radiation immune intratumor response. This study quantifies the role of indirect cell death in dosimetric errors as a function of spatial patient setup uncertainty for stereotactic treatments of multiple lesions.

Material and Methods

Nine complex patients with 61 total tumors (2‐16 tumors/patient) were planned using SIMT‐VMAT with geometry similar to HyperArc with a 10MV‐FFF beam (2400 MU/min). Isocenter was placed at the geometric center of all tumors. Average gross tumor volume (GTV) and planning target volume (PTV) were 1.1 cc (0.02–11.5) and 1.9 cc (0.11–18.8) with an average distance to isocenter of 5.4 cm (2.2–8.9). The prescription was 20 Gy to each PTV. Plans were recalculated with induced clinically observable patient setup errors [±2 mm, ±2^o] in all six directions. Boolean structures were generated to calculate the effect of DCK via 20 Gy isodose volume (IDV) and ICK via 15 Gy IDV minus the 20 Gy IDV. Contributions of each IDV to the PTV coverage were analyzed along with normal brain toxicity due to the patient setup uncertainty. Induced uncertainty and minimum dose covering the entire PTV were analyzed to determine the maximum tolerable patient setup errors to utilize the ICK effect for radiosurgery of m‐bm via SIMT‐VMAT.

Results

Patient setup errors of 1.3 mm /1.3° in all six directions must be maintained to achieve PTV coverage of the 15 Gy IDV for ICK. Setup errors of ±2 mm/2° showed clinically unacceptable loss of PTV coverage of 29.4 ± 14.6% even accounting the ICK effect. However, no clinically significant effect on normal brain dosimetry was observed.

Conclusions

Radiosurgery of m‐bm using SIMT‐VMAT treatments have shown positive clinical outcomes even with small residual patient setup errors. These clinical outcomes, while largely due to DCK, may also potentially be due to the ICK. Potential mechanisms, such as devascularization and/or radiation‐induced intratumor immune enhancement, should be explored to provide a better understanding of the radiobiological response of stereotactic radiosurgery of m‐bm using a SIMT‐VMAT plan.

Stereotactic Radiosurgery for Postoperative Metastatic Surgical Cavities: A Critical Review and International Stereotactic Radiosurgery Society (ISRS) Practice Guidelines

PURPOSE

The purpose of this critical review is to summarize the literature specific to single-fraction stereotactic radiosurgery (SRS) and multiple-fraction stereotactic radiation therapy (SRT) for postoperative brain metastases resection cavities and to present practice recommendations on behalf of the ISRS.

METHODS AND MATERIALS

The Medline and Embase databases were used to apply the Preferred Reporting Items for Systematic Reviews and Meta-Analyses approach to search for manuscripts reporting SRS/SRT outcomes for postoperative brain metastases tumor bed resection cavities with a search end date of July 20, 2018. Prospective studies, consensus guidelines, and retrospective series that included exclusively postoperative brain metastases and had at minimum 100 patients were considered eligible.

RESULTS

The Embase search revealed 157 manuscripts, of which 77 were selected for full-text screening. PubMed yielded 55 manuscripts, of which 23 were selected for full text screening. We deemed 8 retrospective series, 1 phase 2 prospective study, 3 randomized controlled trials, and 1 consensus contouring paper appropriate for inclusion. The data suggest that SRS/SRT to surgical cavities with prescription doses of 30 to 50 Gy equivalent effective dose (EQD) 2₁₀, 50 to 70 Gy EQD2₅, and 70 to 90 EQD2₂ are associated with rates of local control ranging from 60.5% to 91% (median, 80.5%). Randomized data suggest improved local control with single-fraction SRS compared with observation and improved cognitive outcomes compared with whole-brain radiation therapy (WBRT). The toxicity of SRS/SRT in the postoperative setting was limited and is reviewed herein.

CONCLUSIONS

Although randomized data raise concern for poorer local control after resection cavity SRS than WBRT, these findings may be driven by factors such as conservative prescription doses used in the SRS arm. Retrospective studies suggest high rates of local control after single-fraction SRS and hypofractionated SRT for postoperative brain metastases. With a superior neurocognitive profile and no survival disadvantage to withholding WBRT, the ISRS recommends SRS as first-line treatment for eligible postoperative patients. Emerging data suggest that fractionated SRT may provide superior local control compared with single-fraction SRS, in particular, for large tumor cavity volumes/diameters and potentially for patients with a preoperative diameter greater than 2.5 cm.