A new prognostic score for predicting survival in patients treated with robotic stereotactic radiotherapy for brain metastases

Predicting Survival with Brain Metastases in the Stereotactic Radiosurgery Era: are Existing Prognostic Scores Still Relevant? Or Can we do Better?

Predicting survival is essential to tailoring treatment for patients diagnosed with brain metastases. We have evaluated the performance of widely used, validated prognostic scoring systems (Graded Prognostic Assessment and diagnosis-specific Graded Prognostic Assessment) in over 1000 'real-world' patients treated with stereotactic radiosurgery to the brain, selected according to National Health Service commissioning criteria. Survival outcomes from our dataset were consistent with those predicted by the prognostic systems, but with certain cancer subtypes showing a significantly better survival than predicted. Although performance status remains the simplest tool for prediction, total brain tumour volume emerges as an independent prognostic factor, and a new, improved, prognostic scoring system incorporating this has been developed.

Prediction of the treatment response and local failure of patients with brain metastasis treated with stereotactic radiosurgery using machine learning: A systematic review and meta-analysis

BACKGROUND

Stereotactic radiosurgery (SRS) effectively treats brain metastases. It can provide local control, symptom relief, and improved survival rates, but it poses challenges in selecting optimal candidates, determining dose and fractionation, monitoring for toxicity, and integrating with other modalities. Practical tools to predict patient outcomes are also needed. Machine learning (ML) is currently used to predict treatment outcomes. We aim to investigate the accuracy of ML in predicting treatment response and local failure of brain metastasis treated with SRS.

METHODS

PubMed, Scopus, Web of Science (WoS), and Embase were searched until April 16th, which was repeated on October 17th, 2023 to find possible relevant papers. The study preparation adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline. The statistical analysis was performed by the MIDAS package of STATA v.17.

RESULTS

A total of 17 articles were reviewed, of which seven and eleven were related to the clinical use of ML in predicting local failure and treatment response. The ML algorithms showed sensitivity and specificity of 0.89 (95% CI: 0.84-0.93) and 0.87 (95% CI: 0.81-0.92) for predicting treatment response. The positive likelihood ratio was 7.1 (95% CI: 4.5-11.1), the negative likelihood ratio was 0.13 (95% CI: 0.08-0.19), and the diagnostic odds ratio was 56 (95% CI: 25-125). Moreover, the pooled estimates for sensitivity and specificity of ML algorithms for predicting local failure were 0.93 (95% CI: 0.76-0.98) and 0.80 (95% CI: 0.53-0.94). The positive likelihood ratio was 4.7 (95% CI: 1.6-14.0), the negative likelihood ratio was 0.09 (95% CI: 0.02-0.39), and the diagnostic odds ratio was 53 (95% CI: 5-606).

CONCLUSION

ML holds promise in predicting treatment response and local failure in brain metastasis patients receiving SRS. However, further studies and improvements in the treatment process can refine the models and effectively integrate them into clinical practice.

Analysis of prognostic factors and the role of epilepsy in neurosurgical patients with brain metastases

Background

Brain metastases (BMs) represent the most frequent brain tumors in adults. The identification of key prognostic factors is essential for choosing the therapeutic strategy tailored to each patient. Epilepsy can precede several months of other clinical presentations of BMs. This work aimed to study the impact of epilepsy and other prognostic factors on BMs patients' survival.

Methods

This retrospective study included 51 patients diagnosed with BMs and who underwent neurosurgery between 2010 and 2021. The impact of BM features and patient's clinical characteristics on the overall survival (OS) was analyzed through uni- and multivariate analysis.

Results

The average OS was 25.98 months and differed according to the histology of the primary tumor. The primary tumor localization and the presence of extracranial metastases had a statistically significant impact on the OS, and patients with single BM showed a superior OS to those with multifocal lesions. The localization of BMs in the temporal lobe correlated with the highest OS. The OS was significantly higher in patients who presented seizures in their clinical onset and in those who had better post-surgical Karnofsky performance status, no post-surgical complications, and who underwent post-surgical treatment.

Conclusion

Our study has highlighted prognostically favorable patient and tumor factors. Among those, a clinical onset with epileptic seizures can help identify brain metastasis hitherto silent. This could lead to immediate diagnostic-therapeutic interventions with more aggressive therapies after appropriate multidisciplinary evaluation.

Navigating Post-Operative Outcomes: A Comprehensive Reframing of an Original Graded Prognostic Assessment in Patients with Brain Metastases

BACKGROUND

Graded Prognostic Assessment (GPA) has been proposed for various brain metastases (BMs) tailored to the primary histology and molecular profiles. However, it does not consider whether patients have been operated on or not and does not include surgical outcomes as prognostic factors. The residual tumor burden (RTB) is a strong predictor of overall survival. We validated the GPA score and introduced "volumetric GPA" in the largest cohort of operated patients and further explored the role of RTB as an additional prognostic factor.

METHODS

A total of 630 patients with BMs between 2007 and 2020 were included. The four GPA components were analyzed. The validity of the original score was assessed using Cox regression, and a modified index incorporating RTB was developed by comparing the accuracy, sensitivity, specificity, F1-score, and AUC parameters.

RESULTS

GPA categories showed an association with survival: age (p < 0.001, hazard ratio (HR) 2.9, 95% confidence interval (CI) 2.5-3.3), Karnofsky performance status (KPS) (p < 0.001, HR 1.3, 95% CI 1.2-1.5), number of BMs (p = 0.019, HR 1.4, 95% CI 1.1-1.8), and the presence of extracranial manifestation (p < 0.001, HR 3, 95% CI 1.6-2.5). The median survival for GPA 0-1 was 4 months; for GPA 1.5-2, it was 12 months; for GPA 2.5-3, it was 21 months; and for GPA 3.5-4, it was 38 months (p < 0.001). RTB was identified as an independent prognostic factor. A cut-off of 2 cm3 was used for further analysis, which showed a median survival of 6 months (95% CI 4-8) vs. 13 months (95% CI 11-14, p < 0.001) for patients with RTB > 2 cm3 and <2 cm3, respectively. RTB was added as an additional component for a modified volumetric GPA score. The survival rates with the modified GPA score were: GPA 0-1: 4 months, GPA 1.5-2: 7 months, GPA 2.5-3: 18 months, and GPA 3.5-4: 34 months. Both scores showed good stratification, with the new score showed a trend towards better discrimination in patients with more favorable prognoses.

CONCLUSION

The prognostic value of the original GPA was confirmed in our cohort of patients who underwent surgery for BM. The RTB was identified as a parameter of high prognostic significance and was incorporated into an updated "volumetric GPA". This score provides a novel tool for prognosis and clinical decision making in patients undergoing surgery. This method may be useful for stratification and patient selection for further treatment and in future clinical trials.

Concomitant radiosurgical and targeted oncological treatment improves the outcome of patients with brain metastases from gastrointestinal cancer

BACKGROUND

So far, only limited studies exist that evaluate patients with brain metastases (BM) from GI cancer and associated primary cancers who were treated by Gamma Knife Radiosurgery (GKRS) and concomitant immunotherapy (IT) or targeted therapy (TT).

METHODS

Survival after GKRS was compared to the general and specific Graded Prognostic Assessment (GPA) and Score Index for Radiosurgery (SIR). Further, the influence of age, sex, Karnofsky Performance Status Scale (KPS), extracranial metastases (ECM) status at BM diagnosis, number of BM, the Recursive Partitioning Analysis (RPA) classes, GKRS1 treatment mode and concomitant treatment with IT or TT on the survival after GKRS was analyzed. Moreover, complication rates after concomitant GKRS and mainly TT treatment are reported.

RESULTS

Multivariate Cox regression analysis revealed IT or TT at or after the first Gamma Knife Radiosurgery (GKRS1) treatment as the only significant predictor for overall survival after GKRS1, even after adjusting for sex, KPS group, age group, number of BM at GKRS1, RPA class, ECM status at BM diagnosis and GKRS treatment mode. Concomitant treatment with IT or TT did not increase the rate of adverse radiation effects. There was no significant difference in local BM progression after GKRS between patients who received IT or TT and patients without IT or TT.

CONCLUSION

Good local tumor control rates and low rates of side effects demonstrate the safety and efficacy of GKRS in patients with BM from GI cancers. The concomitant radiosurgical and targeted oncological treatment significantly improves the survival after GKRS without increasing the rate of adverse radiation effects. To provide local tumor control, radiosurgery remains of utmost importance in modern GI BM management.

Validation of the ALK-Brain Prognostic Index for patients with ALK-rearranged lung cancer and brain metastases

BACKGROUND

Brain metastases (BMs) are a key challenge in the management of anaplastic lymphoma kinase-rearranged non-small-cell lung cancer (ALK+ NSCLC), but prognostic scores are complicated or rely on data before the era of tyrosine kinase inhibitors (TKIs). This study aimed to validate the novel ALK-Brain Prognostic Index (ALK-BPI), which was originally proposed based on 44 TKI-treated ALK+ NSCLC patients from Karolinska University Hospital, using an external clinical cohort.

PATIENTS AND METHODS

TKI-treated ALK+ NSCLC patients with BM from Heidelberg (n = 82, cohort 1) were retrospectively analyzed alone and together with the original Karolinska cohort (n = 126, cohort 2). Cox regression models were used to determine the association of clinical variables and scores with overall survival (OS) after BM diagnosis (BM-related OS).

RESULTS

Both cohorts showed a similar median age (58 years), roughly balanced sex distributions (52%-56% females), and Eastern Cooperative Oncology Group performance status (PS) 0-2 for most patients (87%-92%) at the time of BM development, which were present already at initial diagnosis in 36%-38% of the patients. Most patients had received next-generation ALK inhibitors (54%-63%), while 55%-56% of patients did not receive any radiotherapy. The ALK-BPI identified poor-risk patients (i.e. featuring ≥ 2/3 risk factors: PS > 2, male sex, development of BM after initial diagnosis) with a significantly shorter BM-related OS than other patients in both cohorts: 32/82 in cohort 1 with 21.3 versus 62.2 months in median [hazard ratio (HR) = 2.5, P < 0.001]; 59/126 in cohort 2 with 23.1 versus 67.2 months in median (HR = 2.6, P < 0.001). The five-parameter Lung-molGPA score did not achieve statistical significance and/or clear prognostic separation in all four groups, while the Disease-Specific Graded Prognostic Assessment score did not show consistent results.

CONCLUSIONS

The ALK-BPI is a reliable tool for easy prognostic dichotomization of TKI-treated ALK+ NSCLC patients with BM in daily clinical practice, without the complexity of previous models.

Optimizing treatment of brain metastases in an era of novel systemic treatments: a single center consecutive series

BACKGROUND

The multidisciplinary management of patients with brain metastases consists of surgical resection, radiation treatment and systemic treatment. Tailoring and timing these treatment modalities is challenging. This study presents real-world data from consecutively treated patients and assesses the impact of all treatment strategies and their relation with survival. The aim is to provide new insights to improve multidisciplinary decisions towards individualized treatment strategies in patients with brain metastases.

METHODS

A retrospective consecutive cohort study was performed. Patients with brain metastases were included between June 2018 and May 2020. Brain metastases of small cell lung carcinoma were excluded. Overall survival was analyzed in multivariable models.

RESULTS

676 patients were included in the study, 596 (88%) received radiotherapy, 41 (6%) awaited the effect of newly started or switched systemic treatment and 39 (6%) received best supportive care. Overall survival in the stereotactic radiotherapy group was 14 months (IQR 5-32) and 32 months (IQR 11-43) in patients who started or switched systemic treatment and initially did not receive radiotherapy. In patients with brain metastases without options for local or systemic treatment best supportive care was provided, these patients had an overall survival of 0 months (IQR 0-1). Options for systemic treatment, Karnofsky Performance Score ≥ 70 and breast cancer were prognostic for a longer overall survival, while progressive extracranial metastases and whole-brain-radiotherapy were prognostic for shorter overall survival.

CONCLUSIONS

Assessing prognosis in light of systemic treatment options is crucial after the diagnosis of brain metastasis for the consideration of radiotherapy versus best supportive care.

Stratified assessment of an FDA-cleared deep learning algorithm for automated detection and contouring of metastatic brain tumors in stereotactic radiosurgery

PURPOSE

Artificial intelligence-based tools can be leveraged to improve detection and segmentation of brain metastases for stereotactic radiosurgery (SRS). VBrain by Vysioneer Inc. is a deep learning algorithm with recent FDA clearance to assist in brain tumor contouring. We aimed to assess the performance of this tool by various demographic and clinical characteristics among patients with brain metastases treated with SRS.

MATERIALS AND METHODS

We randomly selected 100 patients with brain metastases who underwent initial SRS on the CyberKnife from 2017 to 2020 at a single institution. Cases with resection cavities were excluded from the analysis. Computed tomography (CT) and axial T1-weighted post-contrast magnetic resonance (MR) image data were extracted for each patient and uploaded to VBrain. A brain metastasis was considered "detected" when the VBrain- "predicted" contours overlapped with the corresponding physician contours ("ground-truth" contours). We evaluated performance of VBrain against ground-truth contours using the following metrics: lesion-wise Dice similarity coefficient (DSC), lesion-wise average Hausdorff distance (AVD), false positive count (FP), and lesion-wise sensitivity (%). Kruskal-Wallis tests were performed to assess the relationships between patient characteristics including sex, race, primary histology, age, and size and number of brain metastases, and performance metrics such as DSC, AVD, FP, and sensitivity.

RESULTS

We analyzed 100 patients with 435 intact brain metastases treated with SRS. Our cohort consisted of patients with a median number of 2 brain metastases (range: 1 to 52), median age of 69 (range: 19 to 91), and 50% male and 50% female patients. The primary site breakdown was 56% lung, 10% melanoma, 9% breast, 8% gynecological, 5% renal, 4% gastrointestinal, 2% sarcoma, and 6% other, while the race breakdown was 60% White, 18% Asian, 3% Black/African American, 2% Native Hawaiian or other Pacific Islander, and 17% other/unknown/not reported. The median tumor size was 0.112 c.c. (range: 0.010-26.475 c.c.). We found mean lesion-wise DSC to be 0.723, mean lesion-wise AVD to be 7.34% of lesion size (0.704 mm), mean FP count to be 0.72 tumors per case, and lesion-wise sensitivity to be 89.30% for all lesions. Moreover, mean sensitivity was found to be 99.07%, 97.59%, and 96.23% for lesions with diameter equal to and greater than 10 mm, 7.5 mm, and 5 mm, respectively. No other significant differences in performance metrics were observed across demographic or clinical characteristic groups.

CONCLUSION

In this study, a commercial deep learning algorithm showed promising results in segmenting brain metastases, with 96.23% sensitivity for metastases with diameters of 5 mm or higher. As the software is an assistive AI, future work of VBrain integration into the clinical workflow can provide further clinical and research insights.

Predicting survival after radiosurgery in patients with lung cancer brain metastases using deep learning of radiomics and EGFR status

The early prediction of overall survival (OS) in patients with lung cancer brain metastases (BMs) after Gamma Knife radiosurgery (GKRS) can facilitate patient management and outcome improvement. However, the disease progression is influenced by multiple factors, such as patient characteristics and treatment strategies, and hence satisfactory performance of OS prediction remains challenging. Accordingly, we proposed a deep learning approach based on comprehensive predictors, including clinical, imaging, and genetic information, to accomplish reliable and personalized OS prediction in patients with BMs after receiving GKRS. Overall 1793 radiomic features extracted from pre-GKRS magnetic resonance images (MRI), clinical information, and epidermal growth factor receptor (EGFR) mutation status were retrospectively collected from 237 BM patients who underwent GKRS. DeepSurv, a multi-layer perceptron model, with 4 different aggregation methods of radiomics was applied to predict personalized survival curves and survival status at 3, 6, 12, and 24 months. The model combining clinical features, EGFR status, and radiomics from the largest BM showed the best prediction performance with concordance index of 0.75 and achieved areas under the curve of 0.82, 0.80, 0.84, and 0.92 for predicting survival status at 3, 6, 12, and 24 months, respectively. The DeepSurv model showed a significant improvement (p < 0.001) in concordance index compared to the validated lung cancer BM prognostic molecular markers. Furthermore, the model provided a novel estimate of the risk-of-death period for patients. The personalized survival curves generated by the DeepSurv model effectively predicted the risk-of-death period which could facilitate personalized management of patients with lung cancer BMs.

Association between red blood cell distribution width and mortality in patients with metastatic brain tumors: A retrospective single-center cohort study

Metastatic brain tumor has been associated with high mortality and poor prognosis. However, information on indicators predicting surgical prognosis in patients with brain metastases is limited. This study aimed to investigate the association between preoperative red blood cell distribution width (RDW) and mortality in patients who underwent surgery for metastatic brain tumors. This study analyzed 282 patients who underwent metastatic brain tumor surgery between August 1999 and March 2020. Patients were divided into two groups based on preoperative RDW cut-off values (<13.2 and ≥13.2). The surgical outcomes were compared between the two groups. Additionally, we performed Cox regression analysis to assess the association between preoperative RDW and 1-year and overall mortality. There were significant differences in 180-day mortality (6.2% vs. 28.7%, P<0.001), 1-year mortality (23.8% vs. 46.7%, P<0.001), and overall mortality (75.0% vs. 87.7%, P=0.012) between the two groups. In the Cox regression analysis, RDW ≥ 13.2 was significantly associated with higher 1-year mortality (adjusted hazard ratio [HR], 2.14; 95% confidence interval [CI], 1.38–3.30; P<0.001) and overall mortality (HR, 1.44; 95% CI, 1.09–1.90; P=0.010). Preoperative RDW is strongly associated with high mortality in metastatic brain tumor surgery.