Quiñones‐Hinojosa, Alfredo

Background: The complex nature and heterogeneity involving pituitary surgery results have increased interest in machine learning (ML) applications for prediction of outcomes over the last decade. This study aims to systematically review the characteristics of ML models involving pituitary surgery outcome prediction and assess their reporting quality. Methods: We searched the PubMed, Scopus, and Web of Knowledge databases for publications on the use of ML to predict pituitary surgery outcomes. We used the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) to assess report quality. Our search strategy was based on the terms “artificial intelligence”, “machine learning”, and “pituitary”. Results: 20 studies were included in this review. The principal models reported in each article were post-surgical endocrine outcomes (n = 10), tumor management (n = 3), and intra- and postoperative complications (n = 7). Overall, the included studies adhered to a median of 65% (IQR = 60–72%) of TRIPOD criteria, ranging from 43% to 83%. The median reported AUC was 0.84 (IQR = 0.80–0.91). The most popular algorithms were support vector machine (n = 5) and random forest (n = 5). Only two studies reported external validation and adherence to any reporting guideline. Calibration methods were not reported in 15 studies. No model achieved the phase of actual clinical applicability. Conclusion: Applications of ML in the prediction of pituitary outcomes are still nascent, as evidenced by the lack of any model validated for clinical practice. Although studies have demonstrated promising results, greater transparency in model development and reporting is needed to enable their use in clinical practice. Further adherence to reporting guidelines can help increase AI’s real-world utility and improve clinical practice.

OBJECTIVE The rates of women and underrepresented racial and ethnic minority (UREM) students successfully matching into neurosurgical residency are extremely low and do not reflect the makeup of the general population. As of 2019, only 17.5% of neurosurgical residents in the United States were women, 4.95% were Black or African American, and 7.2% were Hispanic or Latinx. Earlier recruitment of UREM students will help to diversify the neurosurgical workforce. Therefore, the authors developed a virtual educational event for undergraduate students entitled "Future Leaders in Neurosurgery Symposium for Underrepresented Students’’ (FLNSUS). The primary objectives of the FLNSUS were to expose attendees to 1) neurosurgeons from diverse gender, racial, and ethnic backgrounds; 2) neurosurgical research; 3) opportunities for neurosurgical mentorship; and 4) information about life as a neurosurgeon. The authors hypothesized that the FLNSUS would increase student self-confidence, provide exposure to the specialty, and reduce perceived barriers to a neurosurgical career. METHODS To measure the change in participant perceptions of neurosurgery, pre- and postsymposium surveys were administered to attendees. Of the 269 participants who completed the presymposium survey, 250 participated in the virtual event and 124 completed the postsymposium survey. Paired pre- and postsurvey responses were used for analysis, yielding a response rate of 46%. To assess the impact of participant perceptions of neurosurgery as a field, pre- and postsurvey responses to questions were compared. The change in response was analyzed, and a nonparametric sign test was performed to check for significant differences. RESULTS According to the sign test, applicants showed increased familiarity with the field (p < 0.001), increased confidence in their abilities to become neurosurgeons (p = 0.014), and increased exposure to neurosurgeons from diverse gender, racial, and ethnic backgrounds (p < 0.001 for all categories). CONCLUSIONS These results reflect a significant improvement in student perceptions of neurosurgery and suggest that symposiums like the FLNSUS may promote further diversification of the field. The authors anticipate that events promoting diversity in neurosurgery will lead to a more equitable workforce that will ultimately translate to enhanced research productivity, cultural humility, and patient-centered care in neurosurgery.

Study Design Human bone marrow stem cells (hBMSCs) and human adipose-derived stem cells (hADSCs) have demonstrated the capability to regenerate bone once they have differentiated into osteoblasts. Objective This systematic review aimed to evaluate the in vitro osteogenic differentiation potential of these cells when seeded in a poly (lactic- co-glycolic) acid (PLGA) scaffold. Methods A literature search of 4 databases following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines was conducted in January 2021 for studies evaluating the osteogenic differentiation potential of hBMSCs and hADSCs seeded in a PLGA scaffold. Only in vitro models were included. Studies in languages other than English were excluded. Results A total of 257 studies were identified after the removal of duplicates. Seven articles fulfilled our inclusion and exclusion criteria. Four of these reviews used hADSCs and three used hBMSCs in the scaffold. Upregulation in osteogenic gene expression was seen in all the cells seeded in a 3-dimensional scaffold compared with 2-dimensional films. High angiogenic gene expression was found in hADSCs. Addition of inorganic material to the scaffold material affected cell performance. Conclusions Viability, proliferation, and differentiation of cells strongly depend on the environment where they grow. There are several factors that can enhance the differentiation capacity of stem cells. A PLGA scaffold proved to be a biocompatible material capable of boosting the osteogenic differentiation potential and mineralization capacity in hBMSCs and hADSCs.

Background Transvenous embolization is emerging as a promising treatment for cerebrospinal fluid-venous fistulas (CVF) associated with spontaneous intracranial hypotension (SIH). Objective To perform an independent validation of the efficacy and safety of the procedure and describe the procedural techniques used at our institution. Methods A retrospective review was performed including consecutive patients with SIH who had undergone CVF embolization with 3-month clinical and imaging follow-up. Clinical evaluation included the Patient Global Impression of Change (PGIC) Scale and six-item Headache Impact Test (HIT-6). Bern SIH score was used for imaging evaluation on brain MRI. Post-treatment changes in scores were assessed by Wilcoxon signed rank test. Procedural technical details, including use of upper-extremity access and dual-microcatheter pressure cooker technique, were recorded. Results 18 patients (13 female, median age 60 years) were included. 17 (94%) procedures were performed with upper-extremity access and 12 (67%) using dual-microcatheter pressure cooker technique. After embolization, 16 (89%) patients reported much or very much improved at follow-up PGIC; median (IQR) HIT-6 score improved from 68 (62–72) to 36 (36–38) and Bern SIH score improved from 8 (6–8) to 3 (1.5–3.5), p values <0.001. Side effects were transient embolization site back pain in 15 (83%) and rebound intracranial hypertension requiring medical management in 9 (50%) patients. HIT-6 and Bern SIH score changes were similar between conventional and pressure cooker techniques ( p values >0.05). Conclusion Transvenous embolization is independently validated as a highly effective and safe treatment for CVF and is feasible using upper-extremity venous access. Dual-microcatheter and balloon/coil pressure cooker techniques may be used to optimize distribution of embolic material and potentially, treatment efficacy.

Leptomeningeal carcinomatosis (LMC) is a fatal but uncommon complication occurring in 5–15% of patients with stage IV cancer. Current treatment options are ineffective at managing leptomeningeal spread, with a median overall survival (mOS) of 2–6 months. We aimed to conduct a systematic review of the literature to identify past and future therapies for LMC from solid tumors. Forty-three clinical trials (CTs) published between 1982–2022 were identified. Of these, 35 (81.4%) were non-randomized CTs and 8 (18.6%) were randomized CTs. The majority consisted of phase I (16.3%) and phase II CTs (65.1%). Trials enrolled patients with LMC from various primary histology (n = 23, 57.5%), with one CT evaluating LCM from melanoma (2.4%). A total of 21 trials evaluated a single modality treatment. Among CTs, 23.7% closed due to low accrual. Intraventricular (ITV)/intrathecal (IT) drug delivery was the most common route of administration (n = 22, 51.2%) vs. systemic drug delivery (n = 13, 30.3%). Two clinical trials evaluated the use of craniospinal irradiation for LMC with favorable results. LMC continues to carry a dismal prognosis, and over the years, increments in survival have remained stagnant. A paradigm shift towards targeted systemic therapy with continued standardization of efficacy endpoints will help to shed light on promising treatments.

OBJECTIVE The vertebral column is the most common site for skeletal metastasis, often leading to debilitating pain and weakness. Metastatic cancer has unique genetic drivers that potentiate tumorigenicity. There is an unmet need for novel targeted therapy in patients with spinal metastatic disease. METHODS The authors assessed the effect of verteporfin-induced yes-associated protein (YAP) inhibition on spine metastatic cell tumorigenicity and radiation sensitivity in vitro. Animal studies used a subcutaneous xenograft mouse model to assess the use of systemic intraperitoneal verteporfin (IP-VP) and intratumoral verteporfin microparticles (IT-VP) to inhibit the tumorigenicity of lung and breast spinal metastatic tumors from primary patient-derived tissue. RESULTS Verteporfin led to a dose-dependent decrease in migration, clonogenicity, and cell viability via inhibition of YAP and downstream effectors cyclin D1, CTGF, TOP2A, ANDRD1, MCL-1, FOSL2, KIF14, and KIF23. This was confirmed with knockdown of YAP. Verteporfin has an additive response when combined with radiation, and knockdown of YAP rendered cells more sensitive to radiation. The addition of verteporfin to YAP knockdown cells did not significantly alter migration, clonogenicity, or cell viability. IP-VP and IT-VP led to diminished tumor growth (p < 0.0001), especially when combined with radiation (p < 0.0001). Tissue analysis revealed diminished expression of YAP (p < 0.0001), MCL-1 (p < 0.0001), and Ki-67 (p < 0.0001) in tissue from verteporfin-treated tumors compared with vehicle-treated tumors. CONCLUSIONS This is the first study to demonstrate that verteporfin-mediated inhibition of YAP leads to diminished tumorigenicity in lung and breast spinal metastatic cancer cells. Targeting of YAP with verteporfin offers promising results that could be translated to human clinical trials.