Ori Barzilai, Shlomit Ben Moshe, Razi Sitt, Gal Sela, Ben Shofty, and Zvi Ram
Cognition is a key component in health-related quality of life (HRQoL) and is currently incorporated as a major parameter of outcome assessment in patients treated for brain tumors. The effect of surgery on cognition and HRQoL remains debatable. The authors investigated the impact of resection of low-grade gliomas (LGGs) on cognition and the correlation with various histopathological markers.
A retrospective analysis of patients with LGG who underwent craniotomy for tumor resection at a single institution between 2010 and 2014 was conducted. Of 192 who underwent resective surgery for LGG during this period, 49 had complete pre- and postoperative neurocognitive evaluations and were included in the analysis. These patients completed a full battery of neurocognitive tests (memory, language, attention and working memory, visuomotor organization, and executive functions) pre- and postoperatively. Tumor and surgical characteristics were analyzed, including volumetric measurements and histopathological markers (IDH, p53, GFAP).
Postoperatively, significant improvement was found in memory and executive functions. A subgroup analysis of patients with dominant-side tumors, most of whom underwent intraoperative awake mapping, revealed significant improvement in the same domains. Patients whose tumors were on the nondominant side displayed significant improvement only in memory functions. Positive staining for p53 testing was associated with improved language function and greater extent of resection in dominant-side tumors. GFAP positivity was associated with improved memory in patients whose tumors were on the nondominant side. No correlation was found between cognitive outcome and preoperative tumor volume, residual volume, extent of resection, or IDH1 status.
Resection of LGG significantly improves memory and executive function and thus is likely to improve functional outcome in addition to providing oncological benefit. GFAP and pP53 positivity could possibly be associated with improved cognitive outcome. These data support early, aggressive, surgical treatment of LGG.
Nadav Schellekes, Arianna Barbotti, Yael Abramov, Razi Sitt, Francesco Di Meco, Zvi Ram, and Rachel Grossman
Primary central nervous system lymphoma (PCNSL) is a rare CNS tumor with a poor prognosis. It is usually diagnosed by needle biopsy and treated mainly with high-dose chemotherapy. Resection is currently not considered a standard treatment option. A possible prolonged survival after resection of PCNSL lesions in selected patients has been suggested, but selection criteria for surgery, especially for solitary lesions, have never been established.
The authors retrospectively searched their patient database for records of adult patients (≥ 18 years) who were diagnosed and treated for a solitary PCNSL between 2005 and 2019. Patients were divided into groups according to whether they underwent resection or needle biopsy. Statistical analyses were performed in an attempt to identify variables affecting outcome and possible survival advantage and to characterize subgroups of patients who would benefit from resection of their tumor compared with undergoing biopsy only.
A total of 113 patients with a solitary lesion of PCNSL were identified; 36 patients underwent resection, and 77 had a diagnostic stereotactic biopsy only. The statically significant preoperative risk factors included age ≥ 70 years (adjusted HR 9.61, 95% CI 2.42–38.11; p = 0.001), deep-seated lesions (adjusted HR 3.33, 95% CI 1.13–9.84; p = 0.030), and occipital location (adjusted HR 4.26, 95% CI 1.08–16.78; p = 0.039). Having a postoperative Karnofsky Performance Scale (KPS) score < 80 (adjusted HR 3.21, 95% CI 1.05–9.77; p = 0.040) and surgical site infection (adjusted HR 4.27, 95% CI 1.18–15.47; p = 0.027) were significant postoperative risk factors after the adjustment and selection by means of other possible risk factors. In a subgroup analysis, patients younger than 70 years who underwent resection had a nonsignificant trend toward longer survival than those who underwent needle biopsy (median survival 35.0 months vs 15.2 months, p = 0.149). However, patients with a superficial tumor who underwent resection had significantly longer survival times than those who underwent needle biopsy (median survival 34.3 months vs 8.9 months, p = 0.014). Patients younger than 70 years who had a superficial tumor and underwent resection had significantly prolonged survival, with a median survival of 35.0 months compared with 8.9 months in patients from the same group who underwent needle biopsy (p = 0.007).
Specific subgroups of patients with a solitary PCNSL lesion might gain a survival benefit from resection compared with undergoing only a diagnostic biopsy.
Tal Gonen, Rachel Grossman, Razi Sitt, Erez Nossek, Raneen Yanaki, Emanuela Cagnano, Akiva Korn, Daniel Hayat, and Zvi Ram
Intraoperative seizures during awake craniotomy may interfere with patients' ability to cooperate throughout the procedure, and it may affect their outcome. The authors have assessed the occurrence of intraoperative seizures during awake craniotomy in regard to tumor location and the isocitrate dehydrogenase 1 (IDH1) status of the tumor.
Data were collected in 137 consecutive patients who underwent awake craniotomy for removal of a brain tumor. The authors performed a retrospective analysis of the incidence of seizures based on the tumor location and its IDH1 mutation status, and then compared the groups for clinical variables and surgical outcome parameters.
Tumor location was strongly associated with the occurrence of intraoperative seizures. Eleven patients (73%) with tumor located in the supplementary motor area (SMA) experienced intraoperative seizures, compared with 17 (13.9%) with tumors in the other three non-SMA brain regions (p < 0.0001). Interestingly, there was no significant association between history of seizures and tumor location (p = 0.44). Most of the patients (63.6%) with tumor in the SMA region harbored an IDH1 mutation compared with those who had tumors in non-SMA regions. Thirty-one of 52 patients (60%) with a preoperative history of seizures had an IDH1 mutation (p = 0.02), and 15 of 22 patients (68.2%) who experienced intraoperative seizures had an IDH1 mutation (p = 0.03). In a multivariate analysis, tumor location was found as a significant predictor of intraoperative seizures (p = 0.002), and a trend toward IDH1 mutation as such a predictor was found as well (p = 0.06). Intraoperative seizures were not associated with worse outcome.
Patients with tumors located in the SMA are more prone to develop intraoperative seizures during awake craniotomy compared with patients who have a tumor in non-SMA frontal areas and other brain regions. The IDH1 mutation was more common in SMA region tumors compared with other brain regions, and may be an additional risk factor for the occurrence of intraoperative seizures.