Classical Phenotype Research Articles

GATA6 identifies an immune-enriched phenotype linked to favorable outcomes in patients with pancreatic cancer undergoing upfront surgery

This study underscores GATA6's role in distinguishing classical and basal-like pancreatic ductal adenocarcinoma (PDAC) phenotypes. Retrospective studies associate GATA6 immunohistochemistry (IHC) expression with survival outcomes, warranting prospective validation. In a prospective treatment-naive cohort of patients with resected PDAC, GATA6 IHC proves a prognostic discriminator, associating high GATA6 expression with extended survival and the classical PDAC phenotype. However, GATA6's prognostic significance is numerically lower after gemcitabine-based neoadjuvant chemoradiotherapy compared to its significance in patients treated with upfront surgery. Furthermore, GATA6 is implicated in immunomodulation, although a comprehensive investigation of its immunological role is lacking. Treatment-naive PDAC tumors with varying GATA6 expression yield distinct immunological landscapes. Tumors highly expressing GATA6 show reduced infiltration of immunosuppressive regulatory Tcells and M2 macrophages but increased infiltration of immune-stimulating, antigen-presenting, and activated Tcells. Our findings caution against solely relying on GATA6 for molecular subtyping in clinical trials and open avenues for exploring immune-based combination therapies.

Cognitive phenotypes in patients with relapsing-remitting multiple sclerosis with different disease duration, applying the international classification of cognitive disorders in MS (IC-CoDiMS)

Objective: Cognitive impairment is experienced by 40–70% of multiple sclerosis patients, with information processing speed and memory most affected. Until now, cognitive results classified patients as impaired and not impaired. With this dichotomous approach, it is difficult to identify, in a heterogeneous group of patients with cognitive impairment, which cognitive domain(s) are most altered. This study aims to identify cognitive phenotypes in a clinical cohort of adult patients with Relapsing-Remitting Multiple Sclerosis (RRMS) using the International Classification of Cognitive Disorders in MS (IC-CoDiMS) and to characterize their clinical features. Methods: Three hundred patients with RRMS underwent neuropsychological assessment with the Brief Repeatable Battery of Neuropsychological Tests (BRBN-T) and the Brief International Cognitive Multiple Sclerosis (BICAMS). Results: In our cohort, the mean age was 41.38 [11.48 SD] years, and 205 [68.3%] were women. At the −1 SD threshold, 49% were cognitively intact, 25% had uni-domain impairment, 17% had bi-domain impairment, and 9% had multi-domain impairment. Processing speed was the most frequent single-domain impairment, followed by memory and verbal fluency. At the −1.5 SD threshold, 74.7% were cognitively intact, 17% had uni-domain impairment, 6% had bi-domain impairment, had bi-domain impairment, and 3.0% had multi-domain impairment. Memory was the most frequent single-domain impairment, followed by processing speed and verbal fluency. Conclusions: This study corroborates the importance of determining cognitive phenotypes through taxonomy (IC-CoDiMS). In addition, it contributes to improving the classification of cognitive phenotypes in patients with RRMS to enhance the development of more effective treatments and cognitive interventions.

Relationship between the efficacy of sulfonylurea therapy and structural variants of the ABCC8 (rs757110) and KCNJ11 (rs5219) genes in patients with type 2 diabetes mellitus with different phenotypes

Recently, a large number of studies have been conducted to investigate the variability of response in the treatment of diabetes mellitus (T2DM), which make personalized medicine possible. However, the available data are insufficient to decide on the choice of sugar-lowering drugs in T2DM. The search for phenotypic and clinical factors of pharmacogenetic response to therapy continues. Aim of the study was to investigate the relationship between the effectiveness of sulfonylurea therapy and structural variants of the ABCC8 (rs757110) and KCNJ11 (rs5219) genes in T2DM patients with clinical phenotypes. Material and methods. A cross-sectional clinical and pharmacogenetic study of 1271 patients with T2DM receiving sulfonylurea monotherapy (SU) or in combination with metformin and divided into three clinical phenotypes: with insulin deficiency (insulinopenic phenotype), with obesity and moderate insulin resistance (classical phenotype) and with severe insulin resistance (insulin-resistant phenotype) was performed. A glycated hemoglobin content of less than 7% was considered as a good response to SU therapy. Results. Patients with a good response to SU therapy with a classic phenotype had a longer duration of diabetes compared to patients with absence of good response (5.68 ± 5.22 vs 9.12 ± 6.10 years, respectively, p < 0.001) and a later age at diagnosis of T2DM (54.05 ± 7.18 vs 49.37 ± 7.65 years, respectively, p < 0.001); patients with an insulin-resistant phenotype had T2DM established at an older age (54.35 ± 7.91 vs 50.10 ± 7.51 years, respectively, p = 0.001). Logistic regression analysis did not reveal a statistically significant association of the genotype of the structural variant of the ABCC8 gene (rs757110) with the effectiveness of SU in clinical phenotypes. The T/T genotype of the KCNJ11 gene (rs5219) was associated with a better response to SU therapy only in the group of T2DM patients with the classic phenotype (odds ratio 1.85, 95% confidence interval 1.05–3.25, p = 0.041). Conclusions. In patients with type 2 diabetes, predictors of a good response to SU with a classic phenotype were a later age of T2DM diagnosis, a shorter duration of diabetes, and the presence of the T/T genotype of the structural variant of the KCNJ11 gene (rs5219); with an insulin-resistant phenotype, a later age of T2DM diagnosis; with the insulinopenic phenotype no clinical and genetic predictors of the SU effectiveness were identified.

Cornelia de Lange syndrome type 4 with a novel heterozygous missense variant in RAD 21 gene

Cornelia de Lange syndrome (CdLS) is a congenital multisystemic disorder characterized by genetic heterogeneity. It presents with features such as growth and cognitive retardation, upper limb deformities, cardiac, ophthalmologic, and genitourinary anomalies, alongside distinctive facial characteristics. The CdLS phenotype represents a spectrum that includes both classic and non-classic phenotypes resulting from pathogenic variants in genes associated with cohesin functioning, including NIBPL, SMC1A, SMC3, RAD21, BRD4, HDAC8, and ANKRD11. Mutations in these genes manifest diverse clinical features, with RAD21 variants accounting for a small percentage of cohesinopathies in humans. RAD21-related cohesinopathy typically exhibits growth retardation, minor skeletal anomalies, and facial features overlapping with CdLS. However, cognitive involvement tends to be milder. Despite this, due to the limited number of reported cases with RAD21 mutations, establishing genotype-phenotype correlations remain challenging. We present the case of an 18-month-old boy exhibiting developmental delay and distinct morphological features including micro-brachycephaly, depressed nasal bridge, upturned nose, long philtrum, low-set ears, mesomelic limb dwarfism, and a complete endocardial cushion defect. Exome sequencing revealed a novel RAD21 variant in this individual.

0481 A Novel Phenotype Classification for Obstructive Sleep Apnea Based on Hypoxic Burden

0481 A Novel Phenotype Classification for Obstructive Sleep Apnea Based on Hypoxic Burden

Phenotypic Characterization of Patients with Polycystic Ovary Syndrome in a Population from the Ecuadorian Andes: A Cross-Sectional Study.

Background: Polycystic ovary syndrome (PCOS) is a highly prevalent endocrine-metabolic disorder in women of reproductive age. Diagnosis is based on the evidence-based international guideline 2018 and the Rotterdam Consensus to classify PCOS phenotypes. This study aims to characterize the biodemographic, clinical, metabolic, and reproductive variables and their relationship with PCOS phenotypes in a population from the Ecuadorian Andes. Methodology: A cross-sectional study was conducted with a non-random consecutive sample of 92 women who attended the outpatient gynecology and endocrinology clinic at the Hospital of the Technical University of Loja (UTPL)-Santa Inés, Loja, Ecuador, between January 2022 and July 2023. Descriptive statistics, mean calculations, standard deviation, parametric and nonparametric tests, odds ratios (OR), confidence intervals (CI), and p-values were employed. Results: The average age was 22 ± 3.4 years, with a predominantly mestizo, urban, single, highly educated, and medium-high socioeconomic level population. It was identified that phenotypes A + B are at a higher risk of developing oligomenorrhea and hypertriglyceridemia compared to phenotypes C + D, with statistically significant differences (p < 0.05). Furthermore, in terms of reproductive variables, phenotypes A + B exhibit a significantly higher frequency of elevated anti-Müllerian hormone (AMH) compared to phenotypes C + D, also with statistical significance (p < 0.05). Conclusions: The classical phenotypes A and B of PCOS are the most common in Ecuadorian Andean women and carry a higher risk of insulin resistance, anovulation, metabolic disorders, and elevated triglyceride levels compared to phenotypes C and D. Ethnic diversity and sociocultural habits influence the prevalence and clinical manifestations of these phenotypes.

A protocol for annotation of total body photography for machine learning to analyze skin phenotype and lesion classification.

Artificial Intelligence (AI) has proven effective in classifying skin cancers using dermoscopy images. In experimental settings, algorithms have outperformed expert dermatologists in classifying melanoma and keratinocyte cancers. However, clinical application is limited when algorithms are presented with 'untrained' or out-of-distribution lesion categories, often misclassifying benign lesions as malignant, or misclassifying malignant lesions as benign. Another limitation often raised is the lack of clinical context (e.g., medical history) used as input for the AI decision process. The increasing use of Total Body Photography (TBP) in clinical examinations presents new opportunities for AI to perform holistic analysis of the whole patient, rather than a single lesion. Currently there is a lack of existing literature or standards for image annotation of TBP, or on preserving patient privacy during the machine learning process. This protocol describes the methods for the acquisition of patient data, including TBP, medical history, and genetic risk factors, to create a comprehensive dataset for machine learning. 500 patients of various risk profiles will be recruited from two clinical sites (Australia and Spain), to undergo temporal total body imaging, complete surveys on sun behaviors and medical history, and provide a DNA sample. This patient-level metadata is applied to image datasets using DICOM labels. Anonymization and masking methods are applied to preserve patient privacy. A two-step annotation process is followed to label skin images for lesion detection and classification using deep learning models. Skin phenotype characteristics are extracted from images, including innate and facultative skin color, nevi distribution, and UV damage. Several algorithms will be developed relating to skin lesion detection, segmentation and classification, 3D mapping, change detection, and risk profiling. Simultaneously, explainable AI (XAI) methods will be incorporated to foster clinician and patient trust. Additionally, a publicly released dataset of anonymized annotated TBP images will be released for an international challenge to advance the development of new algorithms using this type of data. The anticipated results from this protocol are validated AI-based tools to provide holistic risk assessment for individual lesions, and risk stratification of patients to assist clinicians in monitoring for skin cancer.

Deep learning-based phenotype classification of three ark shells: Anadara kagoshimensis, Tegillarca granosa, and Anadara broughtonii

The rapid and accurate classification of aquatic products is crucial for ensuring food safety, production efficiency, and economic benefits. However, traditional manual methods for classifying ark shell species based on phenotype are time-consuming and inefficient, especially during peak seasons when the demand is high and labor is scarce. This study aimed to develop a deep learning model for the automated identification and classification of commercially important three ark shells (Tegillarca granosa, Anadara broughtonii, and Anadara kagoshimensis) from images. The ark shells were collected and identified using a polymerase chain reaction method developed in a previous study, and a total of 1,400 images were categorized into three species. Three convolutional neural network (CNN) models, Visual Geometry Group Network (VGGnet), Inception-Residual Network (ResNet), and SqueezeNet, were then applied to two different classification sets, Set-1 (four bivalve species) and Set-2 (three ark shell species). Our results showed that SqueezeNet demonstrated the highest accuracy during the training phase for both classification sets, whereas Inception-ResNet exhibited superior accuracy during the validation phase. Similar results were obtained after developing a third classification set (Set-3) to classify six categories by combining Set-1 and Set-2. Overall, the developed CNN-based classification model exhibited a performance comparable or superior to that presented in previous studies and can provide a theoretical basis for bivalve classification, thereby contributing to improved food safety, production efficiency, and economic benefits in the aquatic products industry.

Long-term characterization of cognitive phenotypes in children with seizures over 36 months

RationaleChildren with new-onset epilepsies often exhibit co-morbidities including cognitive dysfunction, which adversely affects academic performance. Application of unsupervised machine learning techniques has demonstrated the presence of discrete cognitive phenotypes at or near the time of diagnosis, but there is limited knowledge of their longitudinal trajectories. Here we investigate longitudinally the presence and progression of cognitive phenotypes and academic status in youth with new-onset seizures as sibling controls. Methods282 subjects (6–16 years) were recruited within 6 weeks of their first recognized seizure along with 167 unaffected siblings. Each child underwent a comprehensive neuropsychological assessment at baseline, 18 and 36 months later. Factor analysis of the neuropsychological tests revealed four underlying domains – language, processing speed, executive function, and verbal memory. Latent trajectory analysis of the mean factor scores over 36 months identified clusters with prototypical cognitive trajectories. ResultsThree unique phenotypic groups with distinct cognitive trajectories over the 36-month period were identified: Resilient, Average, and Impaired phenotypes. The Resilient phenotype exhibited the highest neuropsychological factor scores and academic performance that were all similar to controls; while the Impaired phenotype showed the polar opposite with the worst performances across all test metrics. These findings remained significant and stable over 36 months. Multivariate logistic regression indicated that age of onset, EEG, neurological examination, and sociodemographic disadvantage were associated with phenotype classification. ConclusionsThis study demonstrates the presence of diverse latent cognitive trajectory phenotypes over 36 months in youth with new-onset seizures that are associated with a stable neuropsychological and academic performance longitudinally.

Three-dimensional-based native alignment phenotype classification system: Description for use in planning for deformities during total knee arthroplasty

IntroductionTotal knee arthroplasty (TKA) is a complex surgical procedure that traditionally relies on two-dimensional radiographs for pre-operative planning. These radiographs may not capture the intricate details of individual knee anatomy, potentially limiting the precision of surgical interventions. With advancements in imaging technology, there is an opportunity to refine TKA outcomes. This study introduces the Native Alignment Phenotype classification system that is based on pre-operative 3-dimensional computed tomography (CT) scans, aiming to provide a more detailed understanding of knee deformities and their influence on characterizing knee osteoarthritis and planning for TKA procedures. MethodsThere were 1406 pre-operative non-weight-bearing CT scans analyzed by a single surgeon experienced with robotically-assisted total knee arthroplasties. These scans were converted into three-dimensional models, focusing on the coronal and sagittal planes. Intraoperatively, the robotic system was used to capture native coronal and sagittal deformities for each patient. These values were captured with the patient's leg held in a non-stress, extension pose. A new classification system, ‘The Native Alignment Phenotype’, was developed to categorize the specific differences between individual knees. ResultsThere were four primary knee malalignments identified: varus deformity; valgus deformity; and two deformities in the sagittal plane. These malalignments were further categorized based on the degrees of deviation, creating groups with 5° coronal and sagittal ranges. A total of 77 phenotypic alignment patterns were found based on the analyzed cohort. In the coronal plane, varus HKA deformity between 6 and 10° was the most common, with 36.9% of the cases, followed by varus HKA alignment, which was between 0 and 5°, representing 34.3% of the cases. In the sagittal plane, neutral and flexion contracture deformities between 0 and 5° were the most common, with 32.6% of the cases, followed by a fixed flexion contracture alignment, which was between 6 and 10°, representing 28.7% of the cases. When combining coronal and sagittal planes, the most common alignment was the varus between 0 and 5° with a flexion contracture between 0 and 5° (12.5% of cases), closely followed by the varus between 6 and 10° with a flexion contracture between 6 and 10° (12.4% of cases). ConclusionThe Native Alignment Phenotype classification system offers a nuanced understanding of knee deformities based on three-dimensional (CT scan) assessments, potentially leading to improved surgical outcomes in TKA. By leveraging the detailed data from the CT scans, this system provides a more comprehensive view of the knee's anatomy, emphasizing the importance of individualized, data-driven approaches in knee surgery.

Long-Term Triclocarban Exposure Induced Enterotoxicity by Triggering Intestinal AhR-Mediated Inflammation and Disrupting Microbial Community in Mice.

Exposure to triclocarban (TCC), a commonly used antibacterial agent, has been shown to induce significant intestine injuries and colonic inflammation in mice. However, the detailed mechanisms by which TCC exposure triggered enterotoxicity remain largely unclear. Herein, intestinal toxicity effects of long-term and chronic TCC exposure were investigated using a combination of histopathological assessments, metagenomics, targeted metabolomics, and biological assays. Mechanically, TCC exposure caused induction of intestinal aryl hydrocarbon receptor (AhR) and its transcriptional target cytochrome P4501A1 (Cyp1a1) leading to dysfunction of the gut barrier and disruption of the gut microbial community. A large number of lipopolysaccharides (LPS) are released from the gut lumen into blood circulation owing to the markedly increased permeability and gut leakage. Consequently, toll-like receptor-4 (TLR4) and NF-κB signaling pathways were activated by high levels of LPS. Simultaneously, classic macrophage phenotypes were switched by TCC, shown with marked upregulation of macrophage M1 and downregulation of macrophage M2 that was accompanied by striking upregulation of proinflammatory factors such as Il-1β, Il-6, Il-17, and Tnf-α in the intestinal lamina propria. These findings provide new evidence for the TCC-induced enterotoxicity.

Особенности патологии органа зрения при орфанном синдроме Pierson вследствие мутации гена LAMB2 у детей: обзор литературы и клиническое наблюдение

Pierson syndrome is an orphan disease with an autosomal recessive mode of inheritance, characterized by combined pathology of the organ of vision, kidneys and central nervous system. Pierson syndrome is caused by a mutation in the LAMB2 gene, mapped on chromosome 3p21, encoding β2-laminin. The pathology of the organ of vision in Pierson syndrome in children is quite diverse. They are characterized by the presence of cataract, posterior lenticonus, iris hypoplasia, glaucoma, retinal detachment, etc. Microcoria (pupil diameter less than 2 mm) is also often observed due to atrophy of the pupillary dilator. The final diagnosis of Pierson syndrome can only be made based on the results of a molecular genetic study. Due to the severe lifethreatening renal manifestations of Pierson syndrome, for a long time the description of ophthalmological pathology was rather superficial. However, considering the high frequency and clinical severity of vision pathology in Pierson syndrome, it seems especially important to study in detail the features of the ocular phenotype in this disease. We have described a clinical observation of a child with Pierson syndrome, who was diagnosed with a missense mutation of the LAMB2 gene, a previously undescribed variant (rs143405268) in exon 18. The disease manifested itself as a combined pathology of the organ of vision, kidneys and the central nervous system. The ocular phenotype of the probands represented by microcoria, diffuse congenital cataract, and later by secondary glaucoma (after cataract surgery). Features of the renal phenotype of the proband were revealed – a horseshoe-shaped kidney, proteinuria not reaching the level of nephrotic syndrome, with preserved renal function. The peculiarity of this case of a child with Pierson syndrome is the absence of nephrotic syndrome in combination with the classic ocular phenotype with a previously undescribed variant of the LAMB2 gene mutation identified. Key words: pierson syndrome, LAMB 2, congenital cataract, microcoria, secondary glaucoma

Abstract 6878: Images to insights: Transforming TME characterization into actionable spatial metrics with AI-powered multiplex immunofluorescence

Abstract Background: Precise and actionable assessment of cellular mechanisms and interactions within the tumor microenvironment (TME) is critical for advancing clinical trials, preclinical research, and informed decision-making in immuno-oncology. Multiplex immunofluorescence (mIF) assays have emerged as pivotal tools that can enable comprehensive visualization of protein expression and cellular interactions within the TME. Integration of advanced digital analysis tools and mIF assays, can revolutionize our ability to interpret the TME and usher these assays into high-throughput clinical settings. Methods: In this study, we processed tumor samples utilizing Akoya's PhenoCode Signature Immune Profile Human Protein Panel, which includes immune profiling markers CD8, CD68, CD3, CD20 and tumor marker PanCK. Following the Akoya Opal staining protocol, formalin-fixed paraffin-embedded (FFPE) samples were stained and scanned using the PhenoImager HT scanner. The images were then analyzed using Reveal’s AI-mIF digital assay on the ImageDx platform to generate contextual and spatial biomarker data. A deep learning-based cell segmentation model based on MaskRCNN architecture, signal preprocessing for noise and autofluorescence, and dynamic intensity normalization were used. Phenotype classification was precisely executed using dynamic thresholding or neural network classifiers as needed. Results were validated by establishing concordance with pathologist readouts. Results: The application of our methodology yielded highly concordant results when compared with manual counts, demonstrating the efficacy of our approach. The R2 correlation coefficient across all markers passed the minimum criteria of R2&amp;gt;0.85 with regression slopes between 0.7-1.3, indicating a robust alignment with traditional pathologist assessments. We showcase stratifying samples Desert, Excluded, or Inflamed phenotypes and other quantitative and spatial results such as interactions of immune cells and tumor within 30-um neighborhoods, and their dispersion and arrangement in the TME. Conclusion: The integration of Reveal Biosciences' ImageDx platform and RevealAI-mIF provides an ecosystem to host, visualize, and analyze Akoya’s immune profiling images. This collaborative effort has enabled the transformation of complex multiplex imaging data into distinct spatial phenomic endpoints, thereby introducing the potential to significantly advance biomarker discovery and contribution to translational and clinical research in the field of immuno-oncology. Citation Format: Misagh Naderi, Lazar Krstic, Djordje Cikic, Kristen Ruf, Igor Mihajlovic, Sinisa Todorovic, Claire Weston. Images to insights: Transforming TME characterization into actionable spatial metrics with AI-powered multiplex immunofluorescence [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6878.

Abstract 1577: Cancer associated fibroblast - tumor cell crosstalk enhances epithelial to mesenchymal transition and promotes a classical to basal switch in human pancreatic ductal adenocarcinoma

Abstract Introduction: Cancer associated fibroblasts (CAF) drive a complex tumor microenvironment (TME) in pancreatic ductal adenocarcinoma (PDAC) through incompletely understood interactions. Two major CAF subtypes, myofibroblastic (myCAF) and inflammatory (iCAF), dominate the tumor landscape and have been defined by gene expression profiles. There is significant plasticity between subtypes and alterations in CAF behavior likely drive changes in tumorigenicity and immune evasion. Additionally, PDAC tumor cells usually present with a predominant classical or basal phenotype that is associated with clinical outcomes. We hypothesized that CAF phenotype impacts tumor cell gene expression and alters the clinically-relevant tumor subtyping. Methods: We interrogated cellular crosstalk in the PDAC TME using a novel three-dimensional, patient-matched coculture system of patient-derived organoids (PDO) and CAFs, established from 12 patients undergoing surgical resection. Molecular characterization included bulk RNA sequencing, proteome profiling (107 factor), flow cytometry, multiplex RNA Scope, imaging mass cytometry (IMC), and RT-qPCR. Results: PDO gene expression analysis from bulk RNAseq demonstrated CAF coculture induced phenotypic shifts in the Moffitt subtype from a classical to a basal phenotype. Further, CAF coculture enhanced epithelial to mesenchymal transition (EMT) across all patients, even those with a baseline basal phenotype. We hypothesized this could be due to both cell-cell crosstalk and secreted factors. We therefore treated PDO with CAF-conditioned media (CM) and demonstrated that CM is sufficient to drive the observed classical-to-basal shift by RT-qPCR. Secretome analysis of CM derived from 3 CAF lines identified HGF, FGF19, and IL-8 as factors contributing to this transcriptional change. Next, we used spatial - omics (transcriptomics and proteomics) to examine the validity of these findings in source patient tissue. Using RNA Scope, we discovered distinct tumor neighborhoods expressing basal mRNAs KRT17 and KRT6a surrounded by CAFs compared to those expressing classical genes TFF1 and GATA6. We then used IMC to further explore CAF subtypes in patient tumors and found FAP+ myCAF regions corresponded to basal tumor regions, while CXCL12+ iCAFs were found in classical regions. Conclusions: PDAC tumor biology is likely driven by interactions between cancer cells and CAFs. Potential mediators of these interactions include both cell-cell contact and secreted factors. We introduce a novel approach combining in vitro coculture with tissue-based assays, demonstrating that modern culture techniques preserve cell type heterogeneity and plasticity. Together, these data demonstrate that a CAF-rich TME drives tumor cell plasticity not only towards EMT but also a more basal disease phenotype. Citation Format: Samantha Guinn, Brayan Perez, Jignasha Patel, Jae Lee, Daniel Zabransky, Won Ho, Elana J. Fertig, Emma Kartalia, Joseph Tandurella, Richard Burkhart, Jacquelyn W. Zimmerman, Elizabeth Jaffee. Cancer associated fibroblast - tumor cell crosstalk enhances epithelial to mesenchymal transition and promotes a classical to basal switch in human pancreatic ductal adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1577.

Abstract 4306: Label-free melanoma phenotype classification using AI-based morphological profiling

Abstract Melanoma plasticity and heterogeneity contribute to therapeutic resistance and mortality, and the ability of melanoma cells to switch from melanocytic to mesenchymal phenotypes results in increased invasion and metastasis. Investigating the role of these phenotypic states has been challenging due to limited marker availability for each cell state. High-parameter molecular methods such as RNAseq have produced more descriptive gene signatures of phenotypic states, but these methods are cell-destructive, labor-intensive, expensive, and can take days to weeks to obtain a readout. Different morphologies have been observed for melanocytic and mesenchymal phenotypic states in culture, thus we hypothesized that morphology could provide an orthogonal method to define and study phenotypes. We utilized the Deepcell platform to develop a classifier to predict the phenotype of melanoma cells as melanocytic or mesenchymal based on morphology alone. This method is free of labels and utilizes viable cells, overcoming the technical and practical limitations of traditional, marker-based methods. We used 20 patient-derived cell lines with known melanocytic or mesenchymal transcription scores to develop the ‘Melanoma Phenotype Classifier’ to phenotype melanoma cells based on morphology alone. A classifier accuracy of &amp;gt;88% was achieved, and morphological analysis of the images revealed distinct morphotypes for each phenotype, highlighting distinct morphological differences. To further link phenotypic state with multi-dimensional morphological profiles, we performed genetic and chemical perturbations known to shift the phenotypic state. The Melanoma Phenotype Classifier successfully predicted shifts in phenotypes driven by these perturbations. These results further demonstrate how phenotype is linked to distinct morphological changes that are detectable by artificial intelligence. Lastly, we applied the Melanoma Phenotype Classifier to dissociated melanoma biopsies, which revealed phenotypic heterogeneity that was confirmed by single cell RNASeq. This work establishes a link between cellular morphology and melanoma phenotype and lays the groundwork for the use of morphology as a label-free method of phenotyping viable melanoma cells. Citation Format: Evelyn Lattmann, Andreja Jovic, Julie Kim, Aizhan Tastanova, Tiffine Pham, Christian Corona, Kiran Saini, Zhouyang Lian, Senzeyu Zhang, Ryan Carelli, Kevin B. Jacobs, Manisha Ray, Vivian Lu, Stephane C. Boutet, Mahyar Salek, Maddison Masaeli, Mitchell P. Levesque. Label-free melanoma phenotype classification using AI-based morphological profiling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4306.

Diversity of CFTR variants across ancestries characterized using 454,727 UK biobank whole exome sequences

BackgroundLimited understanding of the diversity of variants in the cystic fibrosis transmembrane conductance regulator (CFTR) gene across ancestries hampers efforts to advance molecular diagnosis of cystic fibrosis (CF). The consequences pose a risk of delayed diagnoses and subsequently worsened health outcomes for patients. Therefore, characterizing the spectrum of CFTR variants across ancestries is critical for revolutionizing molecular diagnoses of CF.MethodsWe analyzed 454,727 UK Biobank (UKBB) whole-exome sequences to characterize the diversity of CFTR variants across ancestries. Using the PanUKBB classification, the participants were assigned into six major groups: African (AFR), American/American Admixed (AMR), Central South Asia (CSA), East Asian (EAS), European (EUR), and Middle East (MID). We segregated ancestry-specific CFTR variants, including those that are CF-causing or clinically relevant. The ages of certain CF-causing variants were determined and analyzed for selective pressure effects, and curated phenotype analysis was performed for participants with clinically relevant CFTR genotypes.ResultsWe detected over 4000 CFTR variants, including novel ancestry-specific variants, across six ancestries. Europeans had the most unique CFTR variants [n = 2212], while the American group had the least unique variants [n = 23]. F508del was the most prevalent CF-causing variant found in all ancestries, except in EAS, where V520F was the most prevalent. Common EAS variants such as 3600G > A, V456A, and V520, which appeared approximately 270, 215, and 338 generations ago, respectively, did not show evidence of selective pressure. Sixteen participants had two CF-causing variants, with two being diagnosed with CF. We found 154 participants harboring a CF-causing and varying clinical consequences (VCC) variant. Phenotype analysis performed for participants with multiple clinically relevant variants returned significant associations with CF and its pulmonary phenotypes [Bonferroni-adjusted p < 0.05].ConclusionsWe leveraged the UKBB database to comprehensively characterize the broad spectrum of CFTR variants across ancestries. The detection of over 4000 CFTR variants, including several ancestry-specific and uncharacterized CFTR variants, warrants the need for further characterization of their functional and clinical relevance. Overall, the presentation of classical CF phenotypes seen in non-CF diagnosed participants with more than one CF-causing variant indicates that they may benefit from current CFTR modulator therapies.

Tibia-first, gap-balanced patient-specific alignment restores bony phenotypes and joint line obliquity in a great majority of varus and straight knees and normalises valgus and severe varus deformities.

The present study focuses on testing the capability of a restricted tibia-first, gap-balanced patient-specific alignment technique (PSA) to restore bony morphology and phenotypes. Three-hundred and sixty-seven patients were treated with navigated total knee arthroplastyand tibia-first gap-balanced PSA technique. Boundaries for medial proximal tibial angle were 86°-92°, mechanical lateral distal femoral angle 86°-92°, and hip-knee-ankle angle175°-183°. Knees were classified by coronal plane alignment of the knee (CPAK), with subsequent analyses comparing pre- and postoperative distributions. Phenotype classification within CPAK groups assessed pre- and postoperative distributions. Preoperatively, the largest CPAK group was type II (30.8%), followed by type I (20.5%) and type V (17.8%). Postoperatively, type II remained the largest group (39%), followed by type V (30%). All groups with varus/valgus deformities (I, III, IV and VI) became smaller. While in straight legs (II, IV), the CPAK was restored in more than 70%-75%, in varus groups (I, IV) in 40%-50%and in valgus (III and VI) in 5%-18%. The joint line obliquity remained the same in the majority of knees (straight >75%; varus 63%-80%; valgus VI 95%), with the exception of CPAK III (40%). The phenotype analysis showed for straight legs a phenotype restoration of 85%, for varus 94% and for valgus 37%. Joint line convergence anglewas reduced significantly in all groups from 1.8°-4.3° preoperatively to 0.6°-1.2° postoperatively. PSA restores bony phenotypes and joint line obliquity in the majority of straight and varus knees, while most of the valgus and extreme varus knees are normalised. Level III, retrospective cohort study.

On-the-fly Raman microscopy guaranteeing the accuracy of discrimination

Accelerating the measurement for discrimination of samples, such as classification of cell phenotype, is crucial when faced with significant time and cost constraints. Spontaneous Raman microscopy offers label-free, rich chemical information but suffers from long acquisition time due to extremely small scattering cross-sections. One possible approach to accelerate the measurement is by measuring necessary parts with a suitable number of illumination points. However, how to design these points during measurement remains a challenge. To address this, we developed an imaging technique based on a reinforcement learning in machine learning (ML). This ML approach adaptively feeds back "optimal" illumination pattern during the measurement to detect the existence of specific characteristics of interest, allowing faster measurements while guaranteeing discrimination accuracy. Using a set of Raman images of human follicular thyroid and follicular thyroid carcinoma cells, we showed that our technique requires 3,333 to 31,683 times smaller number of illuminations for discriminating the phenotypes than raster scanning. To quantitatively evaluate the number of illuminations depending on the requisite discrimination accuracy, we prepared a set of polymer bead mixture samples to model anomalous and normal tissues. We then applied a home-built programmable-illumination microscope equipped with our algorithm, and confirmed that the system can discriminate the sample conditions with 104 to 4,350 times smaller number of illuminations compared to standard point illumination Raman microscopy. The proposed algorithm can be applied to other types of microscopy that can control measurement condition on the fly, offering an approach for the acceleration of accurate measurements in various applications including medical diagnosis.

Active transfer learning for audiogram estimation.

Computational audiology (CA) has grown over the last few years with the improvement of computing power and the growth of machine learning (ML) models. There are today several audiogram databases which have been used to improve the accuracy of CA models as well as reduce testing time and diagnostic complexity. However, these CA models have mainly been trained on single populations. This study integrated contextual and prior knowledge from audiogram databases of multiple populations as informative priors to estimate audiograms more precisely using two mechanisms: (1) a mapping function drawn from feature-based homogeneous Transfer Learning (TL) also known as Domain Adaptation (DA) and (2) Active Learning (Uncertainty Sampling) using a stream-based query mechanism. Simulations of the Active Transfer Learning (ATL) model were tested against a traditional adaptive staircase method akin to the Hughson-Westlake (HW) method for the left ear at frequencies kHz, resulting in accuracy and reliability improvements. ATL improved HW tests from a mean of 41.3 sound stimuli presentations and reliability of dB down to dB. Integrating multiple databases also resulted in classifying the audiograms into 18 phenotypes, which means that with increasing data-driven CA, higher precision is achievable, and a possible re-conceptualisation of the notion of phenotype classifications might be required. The study contributes to CA in identifying an ATL mechanism to leverage existing audiogram databases and CA models across different population groups. Further studies can be done for other psychophysical phenomena using ATL.

Clinical and imaging outcomes of different phenotypes of axial spondyloarthritis: 5-year analysis of the DESIR cohort

ObjectivesTo compare the long-term outcomes of three phenotypes of axial SpA (axSpA). MethodsPatients with a clinical diagnosis of axSpA from the DESIR cohort were grouped into three phenotypes at baseline: ‘Pure axSpA’ (‘Axial’), ‘axSpA with peripheral signs’ (‘IBP+Peripheral’) and ‘axSpA at risk’ (‘At risk’) by latent class analysis. Clinical and imaging data were collected up to 5 years. Clinical outcomes, measured in each visit, included disability (BASFI) and quality of life (QoL; SF36). Imaging outcomes included inflammatory and structural lesions on MRI and radiographs of spine and SIJ. The association between phenotype membership at baseline and each outcome was tested in multivariable GEE models. ResultsIn total, 576 patients with axSpA were included. ‘At risk’ patients had worse disability and QoL than ‘Axial’ patients over time. For instance, ‘At risk’ patients had on average 0.4 more points in BASFI over time than ‘Axial’ patients [β (95 % CI): 0.4 (0.2; 0.7)]. This difference was mostly noted in female patients who were HLA-B27 positive. In addition, the difference between the ‘At risk’ and ‘Axial’ phenotypes was higher in patients receiving bDMARDs than in those not (β=0.6 vs 0.5), since BASFI improved more in ‘Axial’ (∆BASFI: -1.3) than in ‘At risk’ (∆BASFI: -0.9) treated patients. There were no differences in disability and QoL between ‘Axial’ and ‘IBP+Peripheral’ patients. Imaging outcomes were worse in the ‘Axial’ phenotype than in the others over time. ConclusionPatients with ‘axSpA at risk’ show worse self-reported outcomes over time and are less likely to benefit from anti-inflammatory treatment than those with a classical axSpA phenotype.