Citations

Abstract Modern histologic imaging platforms coupled with machine learning methods have provided new opportunities to map the spatial distribution of immune cells in the tumor microenvironment. However, there exists no standardized method for describing or analyzing spatial immune cell data, and most reported spatial analyses are rudimentary. In this review, we provide an overview of two approaches for reporting and analyzing spatial data (raster versus vector‐based). We then provide a compendium of spatial immune cell metrics that have been reported in the literature, summarizing prognostic associations in the context of a variety of cancers. We conclude by discussing two well‐described clinical biomarkers, the breast cancer stromal tumor infiltrating lymphocytes score and the colon cancer Immunoscore, and describe investigative opportunities to improve clinical utility of these spatial biomarkers. © 2023 The Pathological Society of Great Britain and Ireland.

AbstractThe clinical significance of the tumor‐immune interaction in breast cancer is now established, and tumor‐infiltrating lymphocytes (TILs) have emerged as predictive and prognostic biomarkers for patients with triple‐negative (estrogen receptor, progesterone receptor, and HER2‐negative) breast cancer and HER2‐positive breast cancer. How computational assessments of TILs might complement manual TIL assessment in trial and daily practices is currently debated. Recent efforts to use machine learning (ML) to automatically evaluate TILs have shown promising results. We review state‐of‐the‐art approaches and identify pitfalls and challenges of automated TIL evaluation by studying the root cause of ML discordances in comparison to manual TIL quantification. We categorize our findings into four main topics: (1) technical slide issues, (2) ML and image analysis aspects, (3) data challenges, and (4) validation issues. The main reason for discordant assessments is the inclusion of false‐positive areas or cells identified by performance on certain tissue patterns or design choices in the computational implementation. To aid the adoption of ML for TIL assessment, we provide an in‐depth discussion of ML and image analysis, including validation issues that need to be considered before reliable computational reporting of TILs can be incorporated into the trial and routine clinical management of patients with triple‐negative breast cancer. © 2023 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Background The impact of changes in Doppler‐derived kidney venous flow in heart failure (HF) is not well studied. We aimed to investigate the association of Doppler‐derived kidney venous stasis index (KVSI) and intrakidney venous‐flow (IKVF) patterns with adverse cardiorenal outcomes in patients with HF. Methods and Results In this observational cohort study, consecutive inpatients with HF referred to a nephrologist because of a history of diuretic resistance and abnormal kidney function (n=216) underwent spectral kidney assessments after admission (Doppler 1) and 25 to 35 days later (Doppler 2) to identify IKVF patterns (continuous/pulsatile/biphasic/monophasic) and KVSI levels. Cox proportional hazard regression models were used to evaluate the associations between KVSI/IKVF patterns at Doppler 1 as well as changes from Doppler 1 to Doppler 2 and risk of cardiorenal events up to 18 months after admission. Worsening HF or death occurred in 126 patients. Both baseline KVSI (hazard ratio [HR], 1.49 [95% CI, 1.37–1.61] per 0.1‐unit increase) and baseline IKVF pattern (HR, 2.47 [95% CI, 2.01–3.04] per 1 pattern severity increase) were significantly associated with worsening HF/death. Increases in both KVSI and IKVF pattern severity from Doppler 1 to 2 were also associated with an increased risk of worsening HF/death (HR, 3.00 [95% CI, 2.08–4.32] per 0.1‐unit increase change; and HR, 6.73 [95% CI, 3.27–13.86] per 1 pattern increase in severity change, respectively). Similar results were observed for kidney outcomes. Conclusions Baseline kidney venous flow predicted adverse cardiorenal events, and inclusion of serial kidney venous flow in cardiorenal risk stratification could facilitate clinical decision‐making for patients with HF. Registration URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03039959.

Importance Electroencephalograms (EEGs) are a fundamental evaluation in neurology but require special expertise unavailable in many regions of the world. Artificial intelligence (AI) has a potential for addressing these unmet needs. Previous AI models address only limited aspects of EEG interpretation such as distinguishing abnormal from normal or identifying epileptiform activity. A comprehensive, fully automated interpretation of routine EEG based on AI suitable for clinical practice is needed. Objective To develop and validate an AI model (Standardized Computer-based Organized Reporting of EEG–Artificial Intelligence [SCORE-AI]) with the ability to distinguish abnormal from normal EEG recordings and to classify abnormal EEG recordings into categories relevant for clinical decision-making: epileptiform-focal, epileptiform-generalized, nonepileptiform-focal, and nonepileptiform-diffuse. Design, Setting, and Participants In this multicenter diagnostic accuracy study, a convolutional neural network model, SCORE-AI, was developed and validated using EEGs recorded between 2014 and 2020. Data were analyzed from January 17, 2022, until November 14, 2022. A total of 30 493 recordings of patients referred for EEG were included into the development data set annotated by 17 experts. Patients aged more than 3 months and not critically ill were eligible. The SCORE-AI was validated using 3 independent test data sets: a multicenter data set of 100 representative EEGs evaluated by 11 experts, a single-center data set of 9785 EEGs evaluated by 14 experts, and for benchmarking with previously published AI models, a data set of 60 EEGs with external reference standard. No patients who met eligibility criteria were excluded. Main Outcomes and Measures Diagnostic accuracy, sensitivity, and specificity compared with the experts and the external reference standard of patients’ habitual clinical episodes obtained during video-EEG recording. Results The characteristics of the EEG data sets include development data set (N = 30 493; 14 980 men; median age, 25.3 years [95% CI, 1.3-76.2 years]), multicenter test data set (N = 100; 61 men, median age, 25.8 years [95% CI, 4.1-85.5 years]), single-center test data set (N = 9785; 5168 men; median age, 35.4 years [95% CI, 0.6-87.4 years]), and test data set with external reference standard (N = 60; 27 men; median age, 36 years [95% CI, 3-75 years]). The SCORE-AI achieved high accuracy, with an area under the receiver operating characteristic curve between 0.89 and 0.96 for the different categories of EEG abnormalities, and performance similar to human experts. Benchmarking against 3 previously published AI models was limited to comparing detection of epileptiform abnormalities. The accuracy of SCORE-AI (88.3%; 95% CI, 79.2%-94.9%) was significantly higher than the 3 previously published models ( P < .001) and similar to human experts. Conclusions and Relevance In this study, SCORE-AI achieved human expert level performance in fully automated interpretation of routine EEGs. Application of SCORE-AI may improve diagnosis and patient care in underserved areas and improve efficiency and consistency in specialized epilepsy centers.

9569 Background: Immune-checkpoint inhibitor (ICI) immunotherapy increases survival in patients with melanoma. Yet only half of the patients respond, and 10-40% of patients experience immune-related adverse events (irAEs). Previous studies have found a correlation between the development of an irAE and treatment response. Modifying the gut microbiome could positively affect response to ICIs and reduce toxicities. We sought to determine if the microbiome at the onset of treatment, during an irAE, and at 12 weeks, can predict the response or toxicity during ICI treatment for metastatic melanoma. Methods: Melanoma patients (pts) > 18 yo treated with ICI enrolled in a prospective observational cohort study at The Ohio State University Comprehensive Cancer Center Skin Cancer Clinic. Excluded were patients on corticosteroids at the start of ICI cycle 1, except for adrenal physiologic replacement. Stools were collected at baseline, within 2 days of an adverse event assessed by physician chart review using CTCAE (Common Terminology Criteria for Adverse Events) v5.0, and at 12 weeks. Response (ORR) to ICIs was assessed by Response Evaluation Criteria in Solid Tumors (RECIST v1.1) q12 weeks and progression-free survival recorded. Metagenomic whole-genome shotgun sequencing was performed on an Illumina NovaSeq 6000 and classified using HUMAnN3. The effect of microbe relative abundances on irAEs was modeled by logistic regression with the R package glmm. Results: Of the 88 patients enrolled, 48 had metastatic disease with 32 assessable for response. ORR at 12 weeks was 28% (1 CR, 8 PR, 17 SD, and 6 PD) and 24 patients showed PFS at 12 months. Grade > 2 irAEs were observed in 11/48 pts. Abundance of Bacteroides dorei (False Discovery Rate Corrected p-value = 9.5E-07) and Blautia species CAG:257 (padj = 0.001) were enriched in responders, Prevotella Stercorea (padj = 1.1E-07) and a phage with a predicted target of Salmonella (padj = 8.3E-09) in non-responders. Responders with an irAE had enrichment of Bacteroides plebeius (padj = 4.2E-13) and Bacteroides coprophilus (padj = 0.0002) whereas responders without irAE had enrichment of Eubacterium siraeum (padj = 7.2E-05). Conclusions: Longitudinal and event-driven biospecimen collection in patients treated with ICIs showed several bacteria and viruses but no fungi associated with response, with and without the development of an irAE. The abundance of the two high-taxonomic rank microbe groups is significantly associated with irAEs. The association with the Bacteroides Genus is consistent with previous studies and is associated with response to ICIs. A microbiome abundant in Bacteroides and lacking Prevotella is associated with response to ICI. An abundance of Bacteroides could be a biomarker for irAEs, and an abundance of Eubacterium, a biomarker for response without irAEs. Future analyses are necessary to assign causal roles for microbe biomarkers with specific irAEs. Clinical trial information: NCT05102773 .

BACKGROUND: Patients with heart failure (HF) have a high burden of symptoms and physical limitations, regardless of ejection fraction (EF). Whether the benefits of SGLT2 (sodium-glucose cotransporter-2) inhibitors on these outcomes vary across the full range of EF remains unclear. METHODS: Patient-level data were pooled from the DEFINE-HF trial (Dapagliflozin Effects on Biomarkers, Symptoms, and Functional Status in Patients With Heart Failure With Reduced Ejection Fraction) of 263 participants with reduced EF (≤40%), and PRESERVED-HF trial (Effects of Dapagliflozin on Biomarkers, Symptoms and Functional Status in Patients With Preserved Ejection Fraction Heart Failure) of 324 participants with preserved EF (≥45%). Both were randomized, double-blind 12-week trials of dapagliflozin versus placebo, recruiting participants with New York Heart Association class II or higher and elevated natriuretic peptides. The effect of dapagliflozin on the change in the Kansas City Cardiomyopathy Questionnaire (KCCQ) Clinical Summary Score (CSS) at 12 weeks was tested with ANCOVA adjusted for sex, baseline KCCQ, EF, atrial fibrillation, estimated glomerular filtration rate, and type 2 diabetes. Interaction of dapagliflozin effects on KCCQ-CSS by EF was assessed using EF both categorically and continuously with restricted cubic spline. Responder analyses, examining proportions of patients with deterioration, and clinically meaningful improvements in KCCQ-CSS were conducted using logistic regression. RESULTS: Of 587 patients randomized (293 dapagliflozin, 294 placebo), EF was ≤40, >40-≤60, and >60% in 262 (45%), 199 (34%), and 126 (21%), respectively. Dapagliflozin improved KCCQ-CSS at 12 weeks (placebo-adjusted difference 5.0 points [95% CI, 2.6–7.5]; P <0.001). This was consistent in participants with EF≤40 (4.6 points [95% CI, 1.0–8.1]; P =0.01), >40 to ≤60 (4.9 points [95% CI, 0.8–9.0]; P =0.02) and >60% (6.8 points [95% CI, 1.5–12.1]; P =0.01; P interaction =0.79). Benefits of dapagliflozin on KCCQ-CSS were also consistent when analyzing EF continuously ( P interaction =0.94). In responder analyses, fewer dapagliflozin-treated patients had deterioration and more had small, moderate, and large KCCQ-CSS improvements versus placebo; these results were also consistent regardless of EF (all P interaction values nonsignificant). CONCLUSIONS: In patients with HF, dapagliflozin significantly improves symptoms and physical limitations after 12 weeks of treatment, with consistent and clinically meaningful benefits across the full range of EF. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifiers: NCT02653482 and NCT03030235.

AbstractBackgroundThe incidence and impact of de novo fungal airway colonization and infection in lung transplant recipients (LTRs) with known chronic lung allograft dysfunction (CLAD) has not been established. We aimed to determine the 1‐year cumulative incidence and risk factors of de novo fungal colonization or infection in LTRs with CLAD and assess the impact of colonization or infection on post‐CLAD survival.MethodsProspectively collected Toronto Lung Transplant Program database and chart review were used for double‐LTRs who were diagnosed with CLAD from January 1, 2016 to January 1, 2020 and who were free of airway fungi within 1 year prior to CLAD onset. International Society for Heart and Lung Transplantation definitions were used to define clinical syndromes. Cox‐Proportional Hazards Models were used for risk‐factor analysis. Survival analysis could not be completed secondary to low number of fungal events; therefore, descriptive statistics were employed for survival outcomes.ResultsWe found 186 LTRs diagnosed with CLAD meeting our inclusion criteria. The 1‐year cumulative incidence for any fungal event was 11.8% (7.0% for infection and 4.8% for colonization). Aspergillus fumigatus was a causative pathogen in eight of 13 (61.5%) patients with infection and six of nine (66.7%) patients with colonization. No patients with fungal colonization post‐CLAD developed fungal infection. Peri‐CLAD diagnosis (3 months prior or 1 month after) methylprednisolone bolus (hazards ratio: 8.84, p = .001) increased the risk of fungal events. Most patients diagnosed with fungal infections (53.8%) died within 1‐year of CLAD onset.ConclusionDe novo IFIs and fungal colonization following CLAD onset were not common. Fungal colonization did not lead to fungal infection. Methylprednisolone bolus was a significant risk factors for post‐CLAD fungal events. image

ObjectivesTo explore the impact of female sex on the experiences of trainees and surgeons in otolaryngology from LMIC and HIC.MethodsThis study includes perspectives of five consultants and two resident physicians from the United States, United Kingdom, Uganda, Kenya and South Africa.ResultsSix themes emerged from these interviews. Overall, LMIC and HIC women shared similar experiences of microaggressions during patients, working along ancillary staff, related to pregnancy, imposter syndrome, difficulties during job search, and unique barriers as consultant.ConclusionsThe findings of this study highlight that gender disparities are present at all levels in Otolaryngology but can present differently depending on context. Laryngoscope, 133:547–551, 2023