Search Results

You are looking at 1 - 10 of 11 items for

  • Author or Editor: Satvir Saggi x
  • Refine by Access: all x
Clear All Modify Search
Free access

Machine learning for predicting hemorrhage in pediatric patients with brain arteriovenous malformation

Satvir Saggi, Ethan A. Winkler, Simon G. Ammanuel, Ramin A. Morshed, Joseph H. Garcia, Jacob S. Young, Alexa Semonche, Heather J. Fullerton, Helen Kim, Daniel L. Cooke, Steven W. Hetts, Adib Abla, Michael T. Lawton, and Nalin Gupta

OBJECTIVE

Ruptured brain arteriovenous malformations (bAVMs) in a child are associated with substantial morbidity and mortality. Prior studies investigating predictors of hemorrhagic presentation of a bAVM during childhood are limited. Machine learning (ML), which has high predictive accuracy when applied to large data sets, can be a useful adjunct for predicting hemorrhagic presentation. The goal of this study was to use ML in conjunction with a traditional regression approach to identify predictors of hemorrhagic presentation in pediatric patients based on a retrospective cohort study design.

METHODS

Using data obtained from 186 pediatric patients over a 19-year study period, the authors implemented three ML algorithms (random forest models, gradient boosted decision trees, and AdaBoost) to identify features that were most important for predicting hemorrhagic presentation. Additionally, logistic regression analysis was used to ascertain significant predictors of hemorrhagic presentation as a comparison.

RESULTS

All three ML models were consistent in identifying bAVM size and patient age at presentation as the two most important factors for predicting hemorrhagic presentation. Age at presentation was not identified as a significant predictor of hemorrhagic presentation in multivariable logistic regression. Gradient boosted decision trees/AdaBoost and random forest models identified bAVM location and a concurrent arterial aneurysm as the third most important factors, respectively. Finally, logistic regression identified a left-sided bAVM, small bAVM size, and the presence of a concurrent arterial aneurysm as significant risk factors for hemorrhagic presentation.

CONCLUSIONS

By using an ML approach, the authors found predictors of hemorrhagic presentation that were not identified using a conventional regression approach.

Free access

Factors associated with seizures at initial presentation in pediatric patients with cerebral arteriovenous malformations

Joseph H. Garcia, Ethan A. Winkler, Ramin A. Morshed, Alex Lu, Simon G. Ammanuel, Satvir Saggi, Elaina J. Wang, Steve Braunstein, Christine K. Fox, Heather J. Fullerton, Helen Kim, Daniel L. Cooke, Steven W. Hetts, Michael T. Lawton, Adib A. Abla, and Nalin Gupta

OBJECTIVE

Children with cerebral arteriovenous malformations (AVMs) can present with seizures, potentially increasing morbidity and impacting clinical management. However, the factors that lead to seizures as a presenting sign are not well defined. While AVM-related seizures have been described in case series, most studies have focused on adults and have included patients who developed seizures after an AVM rupture. To address this, the authors sought to analyze demographic and morphological characteristics of AVMs in a large cohort of children.

METHODS

The demographic, clinical, and AVM morphological characteristics of 189 pediatric patients from a single-center database were studied. Univariate and multivariate logistic regression models were used to test the effect of these characteristics on seizures as an initial presenting symptom in patients with unruptured brain AVMs.

RESULTS

Overall, 28 of 189 patients initially presented with seizures (14.8%). By univariate comparison, frontal lobe location (p = 0.02), larger AVM size (p = 0.003), older patient age (p = 0.04), and the Supplemented Spetzler-Martin (Supp-SM) grade (0.0006) were associated with seizure presentation. Multivariate analysis confirmed an independent effect of frontal lobe AVM location and higher Supp-SM grade. All patients presenting with seizures had AVMs in the cortex or subcortical white matter.

CONCLUSIONS

While children and adults share some risk factors for seizure presentation, their risk factor profiles do not entirely overlap. Pediatric patients with cortical AVMs in the frontal lobe were more likely to present with seizures. Additionally, the Supp-SM grade was highly associated with seizure presentation. Future clinical research should focus on the effect of therapeutic interventions targeting AVMs on seizure control in these patients.

Free access

Validation of the Ruptured Arteriovenous Malformation Grading Scale in a pediatric cohort

Joseph H. Garcia, Caleb Rutledge, Ethan A. Winkler, Luis Carrete, Ramin A. Morshed, Alex Y. Lu, Satvir Saggi, Christine K. Fox, Heather J. Fullerton, Helen Kim, Daniel L. Cooke, Steven W. Hetts, Michael T. Lawton, Nalin Gupta, and Adib A. Abla

OBJECTIVE

Pediatric brain arteriovenous malformations (AVMs) are the leading cause of spontaneous intracranial hemorrhage (SICH) in children. Although the incidence of SICH is low in pediatric populations, such events cause substantial morbidity. The recently created Ruptured Arteriovenous Malformation Grading Scale (RAGS) is proposed as a reliable and novel grading system to specifically serve as a predictor of clinical outcomes in patients following AVM rupture, similar to the Hunt and Hess (HH) grade for ruptured aneurysms. While these data are promising, pediatric patients were notably absent from the original study validating the RAGS. Therefore, correlation of the RAGS score with clinical outcomes following AVM rupture in individuals younger than 18 years of age using the RAGS score is needed. The objective of this study was to validate the RAGS in a cohort of pediatric patients with AVMs who presented with hemorrhage, thereby demonstrating the score’s generalizability, and expanding its external validity.

METHODS

A cohort of children with ruptured AVMs were retrospectively reviewed. Using disability, measured by the modified Rankin Scale (mRS), as the response variable, the area under the receiver operating characteristic curve (AUROC) was calculated for patients based on their RAGS scores for three time periods. The AUROC values were then compared with those generated by two commonly used clinical grading systems, the HH classification and Glasgow Coma Scale.

RESULTS

A total of 81 children who presented with ruptured AVMs were included in the study, with a mean follow-up duration of 4 years. The RAGS score outperformed other clinical grading scales in predicting mRS scores, with AUROC values of 0.81, 0.82, and 0.81 at three distinct follow-up periods.

CONCLUSIONS

The RAGS score correlated well with the clinical outcome after AVM rupture in pediatric patients. Additional validation studies across multiple treatment centers are needed to further demonstrate the generalizability of the scoring system.

Free access

Association of CDKN2A alterations with increased postoperative seizure risk after resection of brain metastases

Daniel D. Cummins, Joseph H. Garcia, Minh P. Nguyen, Satvir Saggi, Jason E. Chung, Ezequiel Goldschmidt, Mitchel S. Berger, Philip V. Theodosopoulos, Edward F. Chang, Mariza Daras, Shawn L. Hervey-Jumper, Manish K. Aghi, and Ramin A. Morshed

OBJECTIVE

Seizures are common and significantly disabling for patients with brain metastases (BMs). Although resection can provide seizure control, a subset of patients with BMs may continue to suffer seizures postoperatively. Genomic BM characteristics may influence which patients are at risk for postoperative seizures. This work explores correlations between genomic alterations and risk of postoperative seizures following BM resection.

METHODS

All patients underwent BM resection at a single institution, with available clinical and sequencing data on more than 500 oncogenes. Clinical seizures were documented pre- and postoperatively. A random forest machine learning classification was used to determine candidate genomic alterations associated with postoperative seizures, and clinical and top genomic variables were correlated with postoperative seizures by using Cox proportional hazards models.

RESULTS

There were 112 patients with BMs who underwent 114 surgeries and had at least 1 month of postoperative follow-up. Seizures occurred preoperatively in 26 (22.8%) patients and postoperatively in 25 (21.9%). The Engel classification achieved at 6 months for those with preoperative seizures was class I in 13 (50%); class II in 6 (23.1%); class III in 5 (19.2%), and class IV in 2 (7.7%). In those with postoperative seizures, only 8 (32.0%) had seizures preoperatively, and preoperative seizures were not a significant predictor of postoperative seizures (HR 1.84; 95% CI 0.79–4.37; p = 0.156). On random forest classification and multivariate Cox analysis controlling for factors including recurrence, extent of resection, and number of BMs, CDKN2A alterations were associated with postoperative seizures (HR 3.22; 95% CI 1.27–8.16; p = 0.014). Melanoma BMs were associated with higher risk of postoperative seizures compared with all other primary malignancies (HR 5.23; 95% CI 1.37–19.98; p = 0.016). Of 39 BMs with CDKN2A alteration, 35.9% (14/39) had postoperative seizures, compared to 14.7% (11/75) without CDKN2A alteration. The overall rate of postoperative seizures in melanoma BMs was 42.9% (15/35), compared with 12.7% (10/79) for all other primary malignancies.

CONCLUSIONS

CDKN2A alterations and melanoma primary malignancy are associated with increased postoperative seizure risk following resection of BMs. These results may help guide postoperative seizure prophylaxis in patients undergoing resection of BMs.

Free access

Genomic alterations associated with rapid progression of brain metastases

Amalie S. V. Uggerly, Daniel D. Cummins, Minh P. Nguyen, Satvir Saggi, Ezequiel Goldschmidt, Edward F. Chang, Michael W. McDermott, Mitchel S. Berger, Philip V. Theodosopoulos, Shawn L. Hervey-Jumper, Mariza Daras, Manish K. Aghi, and Ramin A. Morshed

OBJECTIVE

The aim of this study was to investigate associations between genomic alterations in resected brain metastases and rapid local and distant CNS recurrence identified at the time of postoperative adjuvant radiosurgery.

METHODS

This was a retrospective study on patients who underwent resection of intracranial brain metastases. Next-generation sequencing of more than 500 coding genes was performed on brain metastasis specimens. Postoperative and preradiosurgery MR images were compared to identify rapid recurrence. Genomic data were associated with rapid local and distant CNS recurrence of brain metastases using nominal regression analyses.

RESULTS

The cohort contained 92 patients with 92 brain metastases. Thirteen (14.1%) patients had a rapid local recurrence, and 64 (69.6%) patients had rapid distant CNS progression by the time of postoperative adjuvant radiosurgery, which occurred in a median time of 25 days (range 3–85 days) from surgery. RB1 and CTNNB1 mutations were seen in 8.7% and 9.8% of the cohort, respectively, and were associated with a significantly higher risk of rapid local recurrence (RB1: OR 13.6, 95% CI 2.0–92.39, p = 0.008; and CTNNB1: OR 11.97, 95% CI 2.25–63.78, p = 0.004) on multivariate analysis. No genes were found to be associated with rapid distant CNS progression. However, the presence of extracranial disease was significantly associated with a higher risk of rapid distant recurrence on multivariate analysis (OR 4.06, 95% CI 1.08–15.34, p = 0.039).

CONCLUSIONS

Genomic alterations in RB1 or CTNNB1 were associated with a significantly higher risk of rapid recurrence at the resection site. Although no genomic alterations were associated with rapid distant recurrence, having active extracranial disease was a risk factor for new lesions by the time of adjuvant radiotherapy after resection.

Free access

Factors associated with preoperative and postoperative seizures in patients undergoing resection of brain metastases

Joseph H. Garcia, Ramin A. Morshed, Jason Chung, Miguel A. Millares Chavez, Vivek Sudhakar, Satvir Saggi, Lauro N. Avalos, Aaron Gallagher, Jacob S. Young, Mariza Daras, Michael W. McDermott, Paul A. Garcia, Edward F. Chang, and Manish K. Aghi

OBJECTIVE

Epileptic seizures are a common and potentially devastating complication of metastatic brain tumors. Although tumor-related seizures have been described in previous case series, most studies have focused on primary brain tumors and have not differentiated between different types of cerebral metastases. The authors analyzed a large surgical cohort of patients with brain metastases to examine risk factors associated with preoperative and postoperative seizures and to better understand the seizure risk factors of metastatic brain tumors.

METHODS

Patients who underwent resection of a brain metastasis at the University of California, San Francisco (UCSF), were retrospectively reviewed. Patients included in the study were ≥ 18 years of age, required resection of a brain metastasis, and were treated at UCSF. Primary cancers included melanoma, non–small cell lung adenocarcinoma, breast adenocarcinoma, colorectal adenocarcinoma, esophageal adenocarcinoma, gastric adenocarcinoma, renal cell carcinoma, urothelial carcinoma, ovarian carcinoma, cervical squamous cell carcinoma, and endometrial adenocarcinoma. Patients were evaluated for primary cancer type and seizure occurrence, as well as need for use of antiepileptic drugs preoperatively, at time of discharge, and at 6 months postoperatively. Additionally, Engel classification scores were assigned to those patients who initially presented with seizures preoperatively. Univariate and multivariate regression analyses were used to assess the association of tumor type with preoperative seizures.

RESULTS

Data were retrospectively analyzed for 348 consecutive patients who underwent surgical treatment of brain metastases between 1998 and 2019. The cohort had a mean age of 60 years at the time of surgery and was 59% female. The mean and median follow-up durations after the date of surgery for the cohort were 22 months and 10.8 months, respectively. In univariate analysis, frontal lobe location (p = 0.05), melanoma (p = 0.02), KRAS mutation in lung carcinoma (p = 0.04), intratumoral hemorrhage (p = 0.04), and prior radiotherapy (p = 0.04) were associated with seizure presentation. Postoperative checkpoint inhibitor use (p = 0.002), prior radiotherapy (p = 0.05), older age (p = 0.002), distant CNS progression (p = 0.004), and parietal lobe tumor location (p = 0.002) were associated with seizures at 6 months postoperatively. The final multivariate model confirmed the independent effects of tumor location in the frontal lobe and presence of intratumoral hemorrhage as predictors of preoperative seizures, and checkpoint inhibitor use and parietal lobe location were identified as significant predictors of seizures at 6 months postoperatively.

CONCLUSIONS

Within this surgical cohort of patients with brain metastases, seizures were seen in almost a quarter of patients preoperatively. Frontal lobe metastases and hemorrhagic tumors were associated with higher risk of preoperative seizures, whereas checkpoint inhibitor use and parietal lobe tumors appeared to be associated with seizures at 6 months postoperatively. Future research should focus on the effect of metastatic lesion–targeting therapeutic interventions on seizure control in these patients.

Free access

Supervised machine learning algorithms demonstrate proliferation index correlates with long-term recurrence after complete resection of WHO grade I meningioma

Minh P. Nguyen, Ramin A. Morshed, Cecilia L. Dalle Ore, Daniel D. Cummins, Satvir Saggi, William C. Chen, Abrar Choudhury, Akshay Ravi, David R. Raleigh, Stephen T. Magill, Michael W. McDermott, and Philip V. Theodosopoulos

OBJECTIVE

Meningiomas are the most common primary intracranial tumor, and resection is a mainstay of treatment. It is unclear what duration of imaging follow-up is reasonable for WHO grade I meningiomas undergoing complete resection. This study examined recurrence rates, timing of recurrence, and risk factors for recurrence in patients undergoing a complete resection (as defined by both postoperative MRI and intraoperative impression) of WHO grade I meningiomas.

METHODS

The authors conducted a retrospective, single-center study examining recurrence risk for adult patients with a single intracranial meningioma that underwent complete resection. Uni- and multivariate nominal logistic regression and Cox proportional hazards analyses were performed to identify variables associated with recurrence and time to recurrence. Two supervised machine learning algorithms were then implemented to confirm factors within the cohort that were associated with recurrence.

RESULTS

The cohort consisted of 823 patients who met inclusion criteria, and 56 patients (6.8%) had recurrence on imaging follow-up. The median age of the cohort was 56 years, and 77.4% of patients were female. The median duration of head imaging follow-up for the entire cohort was 2.7 years, but for the subgroup of patients who had a recurrence, the median follow-up was 10.1 years. Estimated 1-, 5-, 10-, and 15-year recurrence-free survival rates were 99.8% (95% confidence interval [CI] 98.8%–99.9%), 91.0% (95% CI 87.7%–93.6%), 83.6% (95% CI 78.6%–87.6%), and 77.3% (95% CI 69.7%–83.4%), respectively, for the entire cohort. On multivariate analysis, MIB-1 index (odds ratio [OR] per 1% increase: 1.34, 95% CI 1.13–1.58, p = 0.0003) and follow-up duration (OR per year: 1.12, 95% CI 1.03–1.21, p = 0.012) were both associated with recurrence. Gradient-boosted decision tree and random forest analyses both identified MIB-1 index as the main factor associated with recurrence, aside from length of imaging follow-up. For tumors with an MIB-1 index < 8, recurrences were documented up to 8 years after surgery. For tumors with an MIB-1 index ≥ 8, recurrences were documented up to 12 years following surgery.

CONCLUSIONS

Long-term imaging follow-up is important even after a complete resection of a meningioma. Higher MIB-1 labeling index is associated with greater risk of recurrence. Imaging screening for at least 8 years in patients with an MIB-1 index < 8 and at least 12 years for those with an MIB-1 index ≥ 8 may be needed to detect long-term recurrences.

Restricted access

Genomic alterations associated with postoperative nodular leptomeningeal disease after resection of brain metastases

Ramin A. Morshed, Daniel D. Cummins, Minh P. Nguyen, Satvir Saggi, Harish N. Vasudevan, Steve E. Braunstein, Ezequiel Goldschmidt, Edward F. Chang, Michael W. McDermott, Mitchel S. Berger, Philip V. Theodosopoulos, Mariza Daras, Shawn L. Hervey-Jumper, and Manish K. Aghi

OBJECTIVE

The relationship between brain metastasis resection and risk of nodular leptomeningeal disease (nLMD) is unclear. This study examined genomic alterations found in brain metastases with the aim of identifying alterations associated with postoperative nLMD in the context of clinical and treatment factors.

METHODS

A retrospective, single-center study was conducted on patients who underwent resection of brain metastases between 2014 and 2022 and had clinical and genomic data available. Postoperative nLMD was the primary endpoint of interest. Targeted next-generation sequencing of > 500 oncogenes was performed in brain metastases. Cox proportional hazards analyses were performed to identify clinical features and genomic alterations associated with nLMD.

RESULTS

The cohort comprised 101 patients with tumors originating from multiple cancer types. There were 15 patients with nLMD (14.9% of the cohort) with a median time from surgery to nLMD diagnosis of 8.2 months. Two supervised machine learning algorithms consistently identified CDKN2A/B codeletion and ERBB2 amplification as the top predictors associated with postoperative nLMD across all cancer types. In a multivariate Cox proportional hazards analysis including clinical factors and genomic alterations observed in the cohort, tumor volume (× 10 cm3; HR 1.2, 95% CI 1.01–1.5; p = 0.04), CDKN2A/B codeletion (HR 5.3, 95% CI 1.7–16.9; p = 0.004), and ERBB2 amplification (HR 3.9, 95% CI 1.1–14.4; p = 0.04) were associated with a decreased time to postoperative nLMD.

CONCLUSIONS

In addition to increased resected tumor volume, ERBB2 amplification and CDKN2A/B deletion were independently associated with an increased risk of postoperative nLMD across multiple cancer types. Additional work is needed to determine if targeted therapy decreases this risk in the postoperative setting.

Free access

Letter to the Editor. Seizures in pediatric patients with cerebral AVMs

Lei Zhao, Liwei Peng, Peng Wang, and Weixin Li

Free access

Letter to the Editor. Seizures in pediatric patients with cerebral AVMs

Lei Zhao, Liwei Peng, Peng Wang, and Weixin Li