GPT-4 Underperforms Experts in Detecting IV Fluid Contamination.

BACKGROUND: Specimens contaminated with intravenous (IV) fluids are common in clinical laboratories. Current methods for detecting contamination rely on insensitive and workflow-disrupting delta checks or manual technologist review. Herein, we assessed the utility of large language models for detecting contamination by IV crystalloids and compared its performance to multiple, but variably trained healthcare personnel (HCP). METHODS: Contamination of basic metabolic panels was simulated using 0.9% normal saline (NS), with (n = 30) and without (n = 30) 5% dextrose (D5NS), at mixture ratios of 0.10 and 0.25. A multimodal language model (GPT-4) and a diverse panel of 8 HCP were asked to adjudicate between real and contaminated results. Classification performance, mixture quantification, and confidence was compared by Wilcoxon rank sum. RESULTS: The 95% CIs for accuracy were 0.57-0.71 vs 0.73-0.80 for GPT-4 and HCP, respectively, on the NS set and 0.57-0.57 vs 0.73-0.80 on the D5NS set. HCP overestimated severity of contamination in the 0.10 mixture group (95% CI of estimate error, 0.05-0.20) for both fluids, while GPT-4 markedly overestimated the D5NS mixture at both ratios (0.16-0.33 for NS, 0.11-0.35 for D5NS). There was no correlation between reported confidence and likelihood of a correct classification. CONCLUSIONS: GPT-4 is less accurate than trained HCP for detecting IV fluid contamination of basic metabolic panel results. However, trained individuals were imperfect at identifying contaminated specimens implying the need for novel, automated tools for its detection.

High-throughput analysis of tissue microarrays using automated desorption electrospray ionization mass spectrometry.

Tissue microarrays (TMAs) are commonly used for the rapid analysis of large numbers of tissue samples, often in morphological assessments but increasingly in spectroscopic analysis, where specific molecular markers are targeted via immunostaining. Here we report the use of an automated high-throughput system based on desorption electrospray ionization (DESI) mass spectrometry (MS) for the rapid generation and online analysis of high-density (6144 samples/array) TMAs, at rates better than 1 sample/second. Direct open-air analysis of tissue samples (hundreds of nanograms) not subjected to prior preparation, plus the ability to provide molecular characterization by tandem mass spectrometry (MS/MS), make this experiment versatile and applicable to both targeted and untargeted analysis in a label-free manner. These capabilities are demonstrated in a proof-of-concept study of frozen brain tissue biopsies where we showcase (i) a targeted MS/MS application aimed at identification of isocitrate dehydrogenase mutation in glioma samples and (ii) an untargeted MS tissue type classification using lipid profiles and correlation with tumor cell percentage estimates from histopathology. The small sample sizes and large sample numbers accessible with this methodology make for a powerful analytical system that facilitates the identification of molecular markers for later use in intraoperative applications to guide precision surgeries and ultimately improve patient outcomes.

Metabolic profiles of human brain parenchyma and glioma for rapid tissue diagnosis by targeted desorption electrospray ionization mass spectrometry.

Desorption electrospray ionization mass spectrometry (DESI-MS) is well suited for intraoperative tissue analysis since it requires little sample preparation and offers rapid and sensitive molecular diagnostics. Currently, intraoperative assessment of the tumor cell percentage of glioma biopsies can be made by measuring a single metabolite, N-acetylaspartate (NAA). The inclusion of additional biomarkers will likely improve the accuracy when distinguishing brain parenchyma from glioma by DESI-MS. To explore this possibility, mass spectra were recorded for extracts from 32 unmodified human brain samples with known pathology. Statistical analysis of data obtained from full-scan and multiple reaction monitoring (MRM) profiles identified discriminatory metabolites, namely gamma-aminobutyric acid (GABA), creatine, glutamic acid, carnitine, and hexane-1,2,3,4,5,6-hexol (abbreviated as hexol), as well as the established biomarker NAA. Brain parenchyma was readily differentiated from glioma based on these metabolites as measured both in full-scan mass spectra and by the intensities of their characteristic MRM transitions. New DESI-MS methods (5 min acquisition using full scans and MS/MS), developed to measure ion abundance ratios among these metabolites, were tested using smears of 29 brain samples. Ion abundance ratios based on signals for GABA, creatine, carnitine, and hexol all had sensitivities > 90%, specificities > 80%, and accuracies > 85%. Prospectively, the implementation of diagnostic ion abundance ratios should strengthen the discriminatory power of individual biomarkers and enhance method robustness against signal fluctuations, resulting in an improved DESI-MS method of glioma diagnosis.

Intraoperative Mass Spectrometry Platform for IDH Mutation Status Prediction, Glioma Diagnosis, and Estimation of Tumor Cell Infiltration.

BACKGROUND: Surgical tumor resection is the primary treatment option for diffuse glioma, the most common malignant brain cancer. The intraoperative diagnosis of gliomas from tumor core samples can be improved by use of molecular diagnostics. Further, residual tumor at surgical margins is a primary cause of tumor recurrence and malignant progression. This study evaluates a desorption electrospray ionization mass spectrometry (DESI-MS) system for intraoperative isocitrate dehydrogenase (IDH) mutation assessment, estimation of tumor cell infiltration as tumor cell percentage (TCP), and disease status. This information could be used to enhance the extent of safe resection and so potentially improve patient outcomes. METHODS: A mobile DESI-MS instrument was modified and used in neurosurgical operating rooms (ORs) on a cohort of 49 human subjects undergoing craniotomy with tumor resection for suspected diffuse glioma. Small tissue biopsies (ntotal = 203) from the tumor core and surgical margins were analyzed by DESI-MS in the OR and classified using univariate and multivariate statistical methods. RESULTS: Assessment of IDH mutation status using DESI-MS/MS to measure 2-hydroxyglutarate (2-HG) ion intensities from tumor cores yielded a sensitivity, specificity, and overall diagnostic accuracy of 89, 100, and 94%, respectively (ncore = 71). Assessment of TCP (categorized as low or high) in tumor margin and core biopsies using N-acetyl-aspartic acid (NAA) intensity provided a sensitivity, specificity, and accuracy of 91, 76, and 83%, respectively (ntotal = 203). TCP assessment using lipid profile deconvolution provided sensitivity, specificity, and accuracy of 76, 85, and 81%, respectively (ntotal = 203). Combining the experimental data and using PCA-LDA predictions of disease status, the sensitivity, specificity, and accuracy in predicting disease status are 63%, 83%, and 74%, respectively (ntotal = 203). CONCLUSIONS: The DESI-MS system allowed for identification of IDH mutation status, glioma diagnosis, and estimation of tumor cell infiltration intraoperatively in a large human glioma cohort. This methodology should be further refined for clinical diagnostic applications.

Intraoperative detection of isocitrate dehydrogenase mutations in human gliomas using a miniature mass spectrometer.

Knowledge of the isocitrate dehydrogenase (IDH) mutation status of glioma patients could provide insights for decision-making during brain surgery. However, pathology is not able to provide such information intraoperatively. Here we describe the first application of a miniature mass spectrometer (MS) to the determination of IDH mutation status in gliomas intraoperatively. The instrumentation was modified to be compatible with use in the operating room. Tandem MS was performed on the oncometabolite, 2-hydroxyglutarate, and a reference metabolite, glutamate, which is not involved in the IDH mutation. Ratios of fragment ion intensities were measured to calculate an IDH mutation score, which was used to differentiate IDH mutant and wild-type tissues. The results of analyzing 25 biopsies from 13 patients indicate that reliable determination of IDH mutation status was achieved (p = 0.0001, using the Kruskal-Wallis non-parametric test). With its small footprint and low power consumption and noise level, this application of miniature mass spectrometers represents a simple and cost-effective platform for an important intraoperative measurement. Graphical abstract.