Citations

TPS3159 Background: ELU001 is a novel first-in-class nanoparticle drug conjugate that consists of ~13 folic acid targeting moieties and a payload of ~22 molecules of the topoisomerase-1 inhibitor, exatecan. Folic acid and exatecan are covalently bound by non-cleavable and cathepsin-B cleavable linkers, respectively, to short polyethylene glycol chains which surround the C’Dot’s silica core. CDCs are very small (~6 nm), allowing for more efficient penetration into solid tumors compared to ADCs and a rapid elimination by the kidneys, which is expected to lead to less toxicity than ADCs that have a longer half-life in circulation. ELU001’s high avidity is designed to promote binding to FRα on the surface of cancer cells that express any level of antigen (low to high). ELU001 then internalizes into the tumor cell and traffics to the lysosome where enzymatic cleavage releases the exatecan payload. ELU001 has demonstrated efficacy in preclinical models with low, moderate and high levels of FRα. The first-in-human trial, ELU-FRα-1, is currently recruiting patients with ovarian, endometrial, colorectal, gastric, gastroesophageal junction, triple negative breast, or non-small cell lung cancers, or cholangiocarcinoma cancers, and, in the opinion of the Investigator, have no meaningful life-prolonging therapy option available. Methods: This is a Phase 1 / 2 multicenter, open label clinical trial with two parts: Part 1 Dose Escalation and Part 2 Tumor Group Expansion Cohort(s). In Part 1, patients with cancer types with a high likelihood of having FRα overexpressing tumors, (ovarian, endometrial, colorectal, gastric, gastroesophageal junction, triple negative breast, or non-small cell lung cancers, or cholangiocarcinoma) will be enrolled. FRα expression is determined retrospectively. ELU001 is administered either weekly (QW) (once a week for 3 weeks, 1 week rest), every other week (Q2W, with no rest between cycles), or every three weeks (Q3W). Initial Part 2 expansion cohorts are planned in ovarian cancer with high FRα expression, ovarian cancer with moderate or low FRα expression, and endometrial cancer with any positive FRα expression. Here, FRα will be prospectively determined. Other cancer types may be assessed in the future. The primary objective for Part 1 is to identify the MTD/RP2D. The primary objective for Part 2 is to determine the ORR. Secondary objectives include DOR, PFS, TFST, PFS2, OS, frequency, severity and tolerability of adverse events, PK, ADA, and FRα expression assessments. Part 1 Dose Escalation will recruit about 2-5 patients at each dose for each regimen (QW; Q2W; Q3W). Part 2 Stage 1 will recruit about 15 patients per tumor group expansion cohort and is expected to begin recruitment in the second half of 2023. The study is enrolling in Part 1 Q2W Cohort 301 and Q3W Cohort A in the US. Clinical trial information: NCT05001282 .

OBJECTIVE Spondylolisthesis is a common operative disease in the United States, but robust predictive models for patient outcomes remain limited. The development of models that accurately predict postoperative outcomes would be useful to help identify patients at risk of complicated postoperative courses and determine appropriate healthcare and resource utilization for patients. As such, the purpose of this study was to develop k-nearest neighbors (KNN) classification algorithms to identify patients at increased risk for extended hospital length of stay (LOS) following neurosurgical intervention for spondylolisthesis. METHODS The Quality Outcomes Database (QOD) spondylolisthesis data set was queried for patients receiving either decompression alone or decompression plus fusion for degenerative spondylolisthesis. Preoperative and perioperative variables were queried, and Mann-Whitney U-tests were performed to identify which variables would be included in the machine learning models. Two KNN models were implemented (k = 25) with a standard training set of 60%, validation set of 20%, and testing set of 20%, one with arthrodesis status (model 1) and the other without (model 2). Feature scaling was implemented during the preprocessing stage to standardize the independent features. RESULTS Of 608 enrolled patients, 544 met prespecified inclusion criteria. The mean age of all patients was 61.9 ± 12.1 years (± SD), and 309 (56.8%) patients were female. The model 1 KNN had an overall accuracy of 98.1%, sensitivity of 100%, specificity of 84.6%, positive predictive value (PPV) of 97.9%, and negative predictive value (NPV) of 100%. Additionally, a receiver operating characteristic (ROC) curve was plotted for model 1, showing an overall area under the curve (AUC) of 0.998. Model 2 had an overall accuracy of 99.1%, sensitivity of 100%, specificity of 92.3%, PPV of 99.0%, and NPV of 100%, with the same ROC AUC of 0.998. CONCLUSIONS Overall, these findings demonstrate that nonlinear KNN machine learning models have incredibly high predictive value for LOS. Important predictor variables include diabetes, osteoporosis, socioeconomic quartile, duration of surgery, estimated blood loss during surgery, patient educational status, American Society of Anesthesiologists grade, BMI, insurance status, smoking status, sex, and age. These models may be considered for external validation by spine surgeons to aid in patient selection and management, resource utilization, and preoperative surgical planning.

OBJECTIVE The purpose of this study was to evaluate the performance of different supervised machine learning algorithms to predict achievement of minimum clinically important difference (MCID) in neck pain after surgery in patients with cervical spondylotic myelopathy (CSM). METHODS This was a retrospective analysis of the prospective Quality Outcomes Database CSM cohort. The data set was divided into an 80% training and a 20% test set. Various supervised learning algorithms (including logistic regression, support vector machine, decision tree, random forest, extra trees, gaussian naïve Bayes, k–nearest neighbors, multilayer perceptron, and extreme gradient boosted trees) were evaluated on their performance to predict achievement of MCID in neck pain at 3 and 24 months after surgery, given a set of predicting baseline features. Model performance was assessed with accuracy, F1 score, area under the receiver operating characteristic curve, precision, recall/sensitivity, and specificity. RESULTS In total, 535 patients (46.9%) achieved MCID for neck pain at 3 months and 569 patients (49.9%) achieved it at 24 months. In each follow-up cohort, 501 patients (93.6%) were satisfied at 3 months after surgery and 569 patients (100%) were satisfied at 24 months after surgery. Of the supervised machine learning algorithms tested, logistic regression demonstrated the best accuracy (3 months: 0.76 ± 0.031, 24 months: 0.773 ± 0.044), followed by F1 score (3 months: 0.759 ± 0.019, 24 months: 0.777 ± 0.039) and area under the receiver operating characteristic curve (3 months: 0.762 ± 0.027, 24 months: 0.773 ± 0.043) at predicting achievement of MCID for neck pain at both follow-up time points, with fair performance. The best precision was also demonstrated by logistic regression at 3 (0.724 ± 0.058) and 24 (0.780 ± 0.097) months. The best recall/sensitivity was demonstrated by multilayer perceptron at 3 months (0.841 ± 0.094) and by extra trees at 24 months (0.817 ± 0.115). Highest specificity was shown by support vector machine at 3 months (0.952 ± 0.013) and by logistic regression at 24 months (0.747 ± 0.18). CONCLUSIONS Appropriate selection of models for studies should be based on the strengths of each model and the aims of the studies. For maximally predicting true achievement of MCID in neck pain, of all the predictions in this balanced data set the appropriate metric for the authors’ study was precision. For both short- and long-term follow-ups, logistic regression demonstrated the highest precision of all models tested. Logistic regression performed consistently the best of all models tested and remains a powerful model for clinical classification tasks.

Objective(s): PIK3CA-related overgrowth spectrum (PROS) is a group of disorders caused by somatic variants in the PIK3CA gene. We aimed to update recommendations on the severity classification, testing, and medical management of patients with PROS. Methods: Using validated consensus methodology, we convened a 13-member panel in 2020 and reviewed current evidence on how to diagnose and treat PROS. The panel was asked to rate the level of disease severity, and the appropriateness of whether to test for a mutation and medical therapy in 217 patient scenarios before a virtual meeting. Panelists discussed areas of disagreement and completed ratings following the meeting. Results: The panel developed clinical presentations and endorsed the disease severity framework defined by functional impairment, a reduction in quality of life, and risk of death. Panelists agreed it is appropriate to test for a PIK3CA gene variant in every moderately/severely affected patient. Panelists agreed it may be appropriate to consider an mammalian (mechanistic) target of rapamycin inhibitor in some severely affected patients and some moderately affected patients with progressive disease. Although clinical trials have only recently begun and the evidence remains limited, panelists also agreed it may be appropriate to consider treatment with phosphoinositide 3-kinase/serine/threonine protein kinase inhibitors in severely affected patients with a confirmed PIK3CA variant or without a confirmed variant but with progressive disease. Conclusion: These recommendations represent the consensus of experts informed by published literature and experience. Future research should validate this guidance using clinical data. Once validated, we hope these recommendations will improve outcomes for patients with PROS.

Abstract Machine learning is a powerful tool that is increasingly being used in many research areas, including neuroscience. The recent development of new algorithms and network architectures, especially in the field of deep learning, has made machine learning models more reliable and accurate and useful for the biomedical research sector. By minimizing the effort necessary to extract valuable features from datasets, they can be used to find trends in data automatically and make predictions about future data, thereby improving the reproducibility and efficiency of research. One application is the automatic evaluation of micrograph images, which is of great value in neuroscience research. While the development of novel models has enabled numerous new research applications, the barrier to use these new algorithms has also decreased by the integration of deep learning models into known applications such as microscopy image viewers. For researchers unfamiliar with machine learning algorithms, the steep learning curve can hinder the successful implementation of these methods into their workflows. This review explores the use of machine learning in neuroscience, including its potential applications and limitations, and provides some guidance on how to select a fitting framework to use in real-life research projects.

Study Design. External validation using prospectively collected data. Objectives. To determine the model performance of “Dialogue Support” (DS) in predicting outcomes after lumbar spine surgery. Summary of Background Data. To help clinicians discuss risk versus benefit with patients considering lumbar fusion surgery, DS has been made available online. As DS was created using a Swedish sample, there is a need to study how well DS performs in alternative populations. Patients and Methods. Preoperative data from patients enrolled in the Quality Outcomes Database were entered into DS. The probability for each patient to report satisfaction, achieve success (leg pain improvement ≥3), or have no leg pain 12 months after surgery was extracted and compared with their actual 12-month postoperative data. The ability of DS to identify patients in the Quality Outcomes Database who report satisfaction, achieve success, or have no leg pain 12 months after surgery was determined using Receiver operating characteristic curve analysis, goodness-of-fit tests, and calibration plots. Results. There was a significant improvement in all outcomes in 23,928 cases included in the analysis from baseline to 12 months postoperative. Most (84%) reported satisfaction, 67% achieved success, and 44% were pain-free 12 months postoperative. Receiver operating characteristic analysis showed that DS had a low ability to predict satisfaction [area under the curve (AUC) = 0.606], success (AUC = 0.546), and being pain-free (AUC = 0.578) at 12 months postoperative; poor fit for satisfaction (<0.001) and being pain-free (P = 0.004), but acceptable fit for success (P = 0.052). Calibration plots showed underestimation for satisfaction and success, but acceptable estimates for being pain-free. Conclusion. DS is not directly transferable to predict satisfaction and success after lumbar surgery in a US population. This may be due to differences in patient characteristics, weights of the variables included, or the exclusion of unknown variables associated with outcomes. Future studies to better understand and improve the transferability of these models are needed.

OBJECTIVE Clear diagnostic delineation is necessary for the development of a strong evidence base in lumbar spinal surgery. Experience with existing national databases suggests that International Classification of Diseases, Tenth Edition (ICD-10) coding is insufficient to support that need. The purpose of this study was to assess agreement between surgeon-specified diagnostic indication and hospital-reported ICD-10 codes for lumbar spine surgery. METHODS Data collection for the American Spine Registry (ASR) includes an option to denote the surgeon’s specific diagnostic indication for each procedure. For cases treated between January 2020 and March 2022, surgeon-delineated diagnosis was compared with the ICD-10 diagnosis generated by standard ASR electronic medical record data extraction. For decompression-only cases, the primary analysis focused on the etiology of neural compression as determined by the surgeon versus that determined on the basis of the related ICD-10 codes extracted from the ASR database. For lumbar fusion cases, the primary analysis compared structural pathology, which may have required fusion, as determined by the surgeon versus that determined on the basis of the extracted ICD-10 codes. This allowed for identification of agreement between surgeon delineation and extracted ICD-10 codes. RESULTS In 5926 decompression-only cases, agreement between the surgeon and ASR ICD-10 codes was 89% for spinal stenosis and 78% for lumbar disc herniation and/or radiculopathy. Both the surgeon and database indicated no structural pathology (i.e., none) suggesting the need for fusion in 88% of cases. In 5663 lumbar fusion cases, agreement was 76% for spondylolisthesis but poor for other diagnostic indications. CONCLUSIONS Agreement between surgeon-specified diagnostic indication and hospital-reported ICD-10 codes was best for patients who underwent decompression only. In the fusion cases, agreement with ICD-10 codes was best in the spondylolisthesis group (76%). In cases other than spondylolisthesis, agreement was poor due to multiple diagnoses or lack of an ICD-10 code that reflected the pathology. This study suggested that standard ICD-10 codes may be inadequate to clearly define the indications for decompression or fusion in patients with lumbar degenerative disease.

OBJECTIVE The Quality Outcomes Database (QOD) was established in 2012 by the NeuroPoint Alliance, a nonprofit organization supported by the American Association of Neurological Surgeons. Currently, the QOD has launched six different modules to cover a broad spectrum of neurosurgical practice—namely lumbar spine surgery, cervical spine surgery, brain tumor, stereotactic radiosurgery (SRS), functional neurosurgery for Parkinson’s disease, and cerebrovascular surgery. This investigation aims to summarize research efforts and evidence yielded through QOD research endeavors. METHODS The authors identified all publications from January 1, 2012, to February 18, 2023, that were produced by using data collected prospectively in a QOD module without a prespecified research purpose in the context of quality surveillance and improvement. Citations were compiled and presented along with comprehensive documentation of the main study objective and take-home message. RESULTS A total of 94 studies have been produced through QOD efforts during the past decade. QOD-derived literature has been predominantly dedicated to spinal surgical outcomes, with 59 and 22 studies focusing on lumbar and cervical spine surgery, respectively, and 6 studies focusing on both. More specifically, the QOD Study Group—a research collaborative between 16 high-enrolling sites—has yielded 24 studies on lumbar grade 1 spondylolisthesis and 13 studies on cervical spondylotic myelopathy, using two focused data sets with high data accuracy and long-term follow-up. The more recent neuro-oncological QOD efforts, i.e., the Tumor QOD and the SRS Quality Registry, have contributed 5 studies, providing insights into the real-world neuro-oncological practice and the role of patient-reported outcomes. CONCLUSIONS Prospective quality registries are an important resource for observational research, yielding clinical evidence to guide decision-making across neurosurgical subspecialties. Future directions of the QOD efforts include the development of research efforts within the neuro-oncological registries and the American Spine Registry—which has now replaced the inactive spinal modules of the QOD—and the focused research on high-grade lumbar spondylolisthesis and cervical radiculopathy.