CKCatherine King

Catherine King

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Catherine King has a total of 66 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    1277 Identification of clinically relevant spatial tissue phenotypes in large-scale multiplex immunofluorescence data via unsupervised graph learning in non-small cell lung cancer

    by Robert Egger, Andrew Fisher, Michael Drage, Jimena Trillo-Tinoco, John Abel, Andrew Browne, Deepta Rajan, Tai Wang, Jake Conway, Catherine King, et al.

    Background Multiplex immunofluorescence (mIF) allows simultaneous spatial interrogation of multiple cell- and tissue-based biomarkers from patient cohorts at scale using whole-slide images (WSI). Identification of spatially-derived insights is limited by conventional approaches that reduce spatial data into human-derived feature sets (e.g., nearest neighbor), necessitating new methods for surveying spatial patterns in full. Here, we utilized a graph neural network (GNN) to identify clinically-relevant, spatially-defined tissue phenotypes with distinct immunogenic profiles in non-small cell lung cancer (NSCLC) and compare this method to traditional approaches. Methods To identify relevant cancer, stromal, and immune cell types, mIF for cytokeratin, CD8, FoxP3, myeloperoxidase, CD68/CD163, and fibroblast activation protein-A was performed with a DAPI counterstain on clinical NSCLC specimens (N=165). From the same cohort, we extracted bulk mRNAseq and proteomic signatures. mIF images were acquired and spectrally unmixed using the Phenoptic platform and Inform software (Akoya). A convolutional neural network was trained to segment regions of cancer epithelium, stroma, and necrosis, while a pretrained network segmented all cells (HALO-AI, Indica Labs). Cells were converted into graphs, and a GNN autoencoder was trained to discover tissue patterns defined by spatial arrangement and mIF cell/tissue phenotypes. Hierarchical clustering identified patient subsets based on tissue patterns, and Cox proportional hazard models assessed overall survival (OS) (figure 1). Results Segmentation and post-processing of 165 mIF WSIs resulted in graphs describing 243,646 cells/WSI (mean; range: 3,279-1,552,934). Training the GNN autoencoder on this dataset revealed three tissue phenotypes: (1) cancer epithelium, (2) cancer stroma and epithelial-stromal interface (ESI) with high inflammation, and (3) cancer stroma and ESI with low inflammation. Samples were then classified based on the relative abundance of these phenotypes. While conventional analyses (e.g., nearest neighbor) failed to identify subgroups with prognostic value, our model revealed prolonged OS (HR=0.46 [0.22, 0.96]) in patients with abundant phenotype 2. Furthermore, patients in this subgroup displayed significant overexpression of the MHC-I antigen presentation pathway and downregulation of a general immune exclusion signature, potentially increasing intrinsic anti-tumor immunity to improve prognosis. Conclusions We developed a novel, unsupervised deep-learning-based GNN for analysis of WSI mIF data in NSCLC. This method identified interpretable tissue phenotypes with clinically-relevant differences in anti-tumor immunity and prognosis. These data show the utility of unsupervised spatial deep-learning methods, compared to traditional approaches, for data-driven discovery of complex patterns in large-scale multiplex images. Ethics Approval This study was performed in accordance with the Bristol Myers Squibb Bioethics policy (https://www.bms.com/about-us/responsibility/position-on-key-issues/bioethics-policy-statement.html) and adhered to the World Medical Association Declaration of Helsinki for Human Research. Study workflow.From left to right: Graphs are constructed from cell locations, morphology, mIF and tissue features. An unsupervised GNN learns to identify characteristic tissue phenotypes defined by these features and the spatial arrangement of cells. Finally, relative abundance of the different phenotypes is used to stratify patients into groups. Notably, these groups are associated with differences in overall survival

  • Precision Medicine With Leflunomide: Consideration of the DHODH Haplotype and Plasma Teriflunomide Concentration and Modification of Outcomes in Patients With Rheumatoid Arthritis

    by Michael D. Wiese, Ashley M. Hopkins, Catherine King, Mihir D. Wechalekar, Anita Lee, Llewellyn Spargo, Robert Metcalf, Leah McWilliams, Catherine Hill, Leslie G. Cleland, et al.

    Objective: Leflunomide is a commonly used disease‐modifying drug in the treatment of rheumatoid arthritis (RA). Its effects are mediated via inhibition of dihydroorotate dehydrogenase (DHODH) by its active metabolite teriflunomide, and the pharmacokinetics of teriflunomide are highly variable. Our objective was to examine the association between the DHODH haplotype and plasma teriflunomide concentration with response to leflunomide in patients with RA where leflunomide was added to an existing disease‐modifying drug regimen after failure to achieve an adequate response with conventional triple therapy.Methods: Patients with RA who were taking, or were about to initiate, leflunomide were included. Participant characteristics, including the DHODH haplotype, were determined. Up to 5 plasma samples were collected after leflunomide was initiated for assays of total and free teriflunomide concentration. Disease activity was determined via the 28‐joint Disease Activity Score (DAS28). The association between DAS28 scores and patient covariates was determined by linear mixed‐effects modeling.Results: A total of 67 patients were included in the study. The DAS28 score after initiation of leflunomide was associated with the baseline DAS28 score (β = 0.70, P < 0.001) and was higher in those who carried the DHODH haplotype 2 (β = 0.56. P = 0.01) and did not carry the shared epitope (β = 0.56, P = 0.013). As total and free plasma teriflunomide concentration increased, the DAS28 score was significantly lower (P < 0.001 and P = 0.001, respectively). When considering threshold concentrations, teriflunomide concentrations >16 mg/liter were associated with a DAS28 score that was 0.33 lower, and when free teriflunomide concentration was >35 µg/liter, the DAS28 score was 0.32 lower.Conclusion: Teriflunomide concentration and carriage of the DHODH haplotype 2 are associated with response to leflunomide in patients with RA, and a total plasma teriflunomide concentration of at least 16 mg/liter is needed to maximize the likelihood of response.


  • Association between major depressive disorder and multiple disease outcomes: a phenome-wide Mendelian randomisation study in the UK Biobank

    by Anwar Mulugeta, Ang Zhou, Catherine King, Elina Hyppönen

    Depression affects all aspects of an individual’s life but evidence relating to the causal effects on health is limited. We used information from 337,536 UK Biobank participants and performed hypothesis-free phenome-wide association analyses between major depressive disorder (MDD) genetic risk score (GRS) and 925 disease outcomes. GRS–disease outcome associations passing the multiple-testing corrected significance threshold (P < 1.9 × 10−3) were followed by Mendelian randomisation (MR) analyses to test for causality. MDD GRS was associated with 22 distinct diseases in the phenome-wide discovery stage, with the strongest signal observed for MDD diagnosis and related co-morbidities including anxiety and sleep disorders. In inverse-variance weighted MR analyses, MDD was associated with several inflammatory and haemorrhagic gastrointestinal diseases, including oesophagitis (OR 1.32, 95% CI 1.18–1.48), non-infectious gastroenteritis (OR 1.25, 95% CI 1.06–1.48), gastrointestinal haemorrhage (OR 1.26, 95% CI 1.11–1.43) and intestinal E.coli infections (OR 3.24, 95% CI 1.74–6.02). Signals were also observed for symptoms/disorders of the urinary system (OR 1.36, 95% CI 1.19–1.56), asthma (OR 1.23, 95% CI 1.06–1.44), and painful respiration (OR 1.28, 95% CI 1.14–1.44). MDD was associated with disorders of lipid metabolism (OR 1.22, 95% CI 1.12–1.34) and ischaemic heart disease (OR 1.30, 95% CI 1.15–1.47). Sensitivity analyses excluding pleiotropic variants provided consistent associations. Our study indicates a causal link between MDD and a broad range of diseases, suggesting a notable burden of co-morbidity. Early detection and management of MDD is important, and treatment strategies should be selected to also minimise the risk of related co-morbidities.


  • Response to interferon-beta treatment in multiple sclerosis patients: a genome-wide association study

    by S Mahurkar, M Moldovan, V Suppiah, M Sorosina, F Clarelli, G Liberatore, S Malhotra, X Montalban, A Antigüedad, M Krupa, et al.

    Up to 50% of multiple sclerosis (MS) patients do not respond to interferon-beta (IFN-β) treatment and determination of response requires lengthy clinical follow-up of up to 2 years. Response predictive genetic markers would significantly improve disease management. We aimed to identify IFN-β treatment response genetic marker(s) by performing a two-stage genome-wide association study (GWAS). The GWAS was carried out using data from 151 Australian MS patients from the ANZgene/WTCCC2 MS susceptibility GWAS (responder (R)=51, intermediate responders=24 and non-responders (NR)=76). Of the single-nucleotide polymorphisms (SNP) that were validated in an independent group of 479 IFN-β-treated MS patients from Australia, Spain and Italy (R=273 and NR=206), eight showed evidence of association with treatment response. Among the replicated associations, the strongest was observed for FHIT (Fragile Histidine Triad; combined P-value 6.74 × 10−6) and followed by variants in GAPVD1 (GTPase activating protein and VPS9 domains 1; combined P-value 5.83 × 10−5) and near ZNF697 (combined P-value 8.15 × 10−5).


  • Lack of Evidence for Genomic Instability in Autistic Children as Measured by the Cytokinesis‐Block Micronucleus Cytome Assay

    by Penelope A.E. Main, Philip Thomas, Manya T. Angley, Robyn Young, Adrian Esterman, Catherine E. King, Michael F. Fenech

    Autism spectrum disorders are a set of neurodevelopmental disorders that are highly hereditable. Increased genomic instability has been observed in other heritable paediatric neurobiological disorders; therefore, the aim of our study was to test the hypothesis that DNA damage is increased in children with autism and that B vitamin status may explain variations in genome integrity between autistic and normal children. We compared 35 children with autism, 27 of their siblings without autism and 25 age‐ and gender‐matched community controls for genomic stability using the cytokinesis‐block micronucleus cytome (CBMN‐cyt) assay, B vitamins and homocysteine, as well as autism‐related behaviours. It was found that there were no differences in CBMN‐cyt biomarkers between the three groups. Vitamin B2 was significantly raised in children with autism and their siblings compared with controls (P = 0.027 and P = 0.016 respectively) but there was no difference in other B vitamins or homocysteine. In conclusion, although replication using a larger cohort is needed, it appears unlikely that genomic instability is a feature of the aetiology of autism. We cannot rule out in utero effects or other types of DNA damage not measured by the CBMN‐cyt assay. Autism Res 2015, 8: 94–104. © 2014 International Society for Autism Research, Wiley Periodicals, Inc.


  • Nineteen years of management for Phoenicopteriformes in European Association of Zoos and Aquaria institutions: the Fabulous Flamingo Surveys and strategies to increase reproduction in captivity

    by C. E. King, A. Bračko

    Flamingos are estimated to be present in at least two‐thirds of European Association of Zoos and Aquaria (EAZA) institutions. In total, 8837 flamingos in 168 institutions, including 164 EAZA zoos and four Wildfowl & Wetlands Trust centres were reported as at 1 January 2010. Around 93% (8214) of the flamingos reported were the three Phoenicopterus spp: Greater flamingo Phoenicopterus roseus, Caribbean flamingo Phoenicopterus ruber and Chilean flamingo Phoenicopterus chilensis. These three species are included in the EAZA Ciconiiformes and Phoenicopteriformes Regional Collection Plan (RCP), along with the Lesser flamingo Phoeniconaias minor. All four species are managed in the lowest RCP category, ‘Monitor by TAG’, as efforts currently focus on broad‐scale issues at the zoo/group level. Answers to a survey demonstrated that EAZA zoos would like to hold many more flamingos than are currently available. The Taxon Advisory Group has adopted 11 strategies to improve the conditions and breeding of flamingos in captivity. For example, to improve breeding results zoos are encouraged to have only a single flamingo species in an enclosure, to balance sex ratios in groups and to keep full‐flighted flamingos. Health problems as well as predator and trauma mortality are also discussed in this paper. The Fabulous Flamingo Survey is being repeated at 5 year intervals in order to monitor progress and adjust strategies.