A unidirectional mapping of ICD-8 to ICD-10 codes, for harmonized longitudinal analysis of diseases

Periodic revisions of the international classification of diseases (ICD) ensure that the classification reflects new practices and knowledge; however, this complicates retrospective research as diagnoses are coded in different versions. For longitudinal disease trajectory studies, a crosswalk is an essential tool and a comprehensive mapping between ICD-8 and ICD-10 has until now been lacking. In this study, we map all ICD-8 morbidity codes to ICD-10 in the expanded Danish ICD version. We mapped ICD-8 codes to ICD-10, using a many-to-one system inspired by general equivalence mappings such that each ICD-8 code maps to a single ICD-10 code. Each ICD-8 code was manually and unidirectionally mapped to a single ICD-10 code based on medical setting and context. Each match was assigned a score (1 of 4 levels) reflecting the quality of the match and, if applicable, a "flag" signalling choices made in the mapping. We provide the first complete mapping of the 8596 ICD-8 morbidity codes to ICD-10 codes. All Danish ICD-8 codes representing diseases were mapped and 5106 (59.4%) achieved the highest consistency score. Only 334 (3.9%) of the ICD-8 codes received the lowest mapping consistency score. The mapping provides a scaffold for translation of ICD-8 to ICD-10, which enable longitudinal disease studies back to and 1969 in Denmark and to 1965 internationally with further adaption.

Dynamic and explainable machine learning prediction of mortality in patients in the intensive care unit: a retrospective study of high-frequency data in electronic patient records

BACKGROUND

Many mortality prediction models have been developed for patients in intensive care units (ICUs); most are based on data available at ICU admission. We investigated whether machine learning methods using analyses of time-series data improved mortality prognostication for patients in the ICU by providing real-time predictions of 90-day mortality. In addition, we examined to what extent such a dynamic model could be made interpretable by quantifying and visualising the features that drive the predictions at different timepoints.

METHODS

Based on the Simplified Acute Physiology Score (SAPS) III variables, we trained a machine learning model on longitudinal data from patients admitted to four ICUs in the Capital Region, Denmark, between 2011 and 2016. We included all patients older than 16 years of age, with an ICU stay lasting more than 1 h, and who had a Danish civil registration number to enable 90-day follow-up. We leveraged static data and physiological time-series data from electronic health records and the Danish National Patient Registry. A recurrent neural network was trained with a temporal resolution of 1 h. The model was internally validated using the holdout method with 20% of the training dataset and externally validated using previously unseen data from a fifth hospital in Denmark. Its performance was assessed with the Matthews correlation coefficient (MCC) and area under the receiver operating characteristic curve (AUROC) as metrics, using bootstrapping with 1000 samples with replacement to construct 95% CIs. A Shapley additive explanations algorithm was applied to the prediction model to obtain explanations of the features that drive patient-specific predictions, and the contributions of each of the 44 features in the model were analysed and compared with the variables in the original SAPS III model.

FINDINGS

From a dataset containing 15 615 ICU admissions of 12 616 patients, we included 14 190 admissions of 11 492 patients in our analysis. Overall, 90-day mortality was 33·1% (3802 patients). The deep learning model showed a predictive performance on the holdout testing dataset that improved over the timecourse of an ICU stay: MCC 0·29 (95% CI 0·25-0·33) and AUROC 0·73 (0·71-0·74) at admission, 0·43 (0·40-0·47) and 0·82 (0·80-0·84) after 24 h, 0·50 (0·46-0·53) and 0·85 (0·84-0·87) after 72 h, and 0·57 (0·54-0·60) and 0·88 (0·87-0·89) at the time of discharge. The model exhibited good calibration properties. These results were validated in an external validation cohort of 5827 patients with 6748 admissions: MCC 0·29 (95% CI 0·27-0·32) and AUROC 0·75 (0·73-0·76) at admission, 0·41 (0·39-0·44) and 0·80 (0·79-0·81) after 24 h, 0·46 (0·43-0·48) and 0·82 (0·81-0·83) after 72 h, and 0·47 (0·44-0·49) and 0·83 (0·82-0·84) at the time of discharge.

INTERPRETATION

The prediction of 90-day mortality improved with 1-h sampling intervals during the ICU stay. The dynamic risk prediction can also be explained for an individual patient, visualising the features contributing to the prediction at any point in time. This explanation allows the clinician to determine whether there are elements in the current patient state and care that are potentially actionable, thus making the model suitable for further validation as a clinical tool.

FUNDING

Novo Nordisk Foundation and the Innovation Fund Denmark.

miRandola 2017: a curated knowledge base of non-invasive biomarkers

miRandola (http://mirandola.iit.cnr.it/) is a database of extracellular non-coding RNAs (ncRNAs) that was initially published in 2012, foreseeing the relevance of ncRNAs as non-invasive biomarkers. An increasing amount of experimental evidence shows that ncRNAs are frequently dysregulated in diseases. Further, ncRNAs have been discovered in different extracellular forms, such as exosomes, which circulate in human body fluids. Thus, miRandola 2017 is an effort to update and collect the accumulating information on extracellular ncRNAs that is spread across scientific publications and different databases. Data are manually curated from 314 articles that describe miRNAs, long non-coding RNAs and circular RNAs. Fourteen organisms are now included in the database, and associations of ncRNAs with 25 drugs, 47 sample types and 197 diseases. miRandola also classifies extracellular RNAs based on their extracellular form: Argonaute2 protein, exosome, microvesicle, microparticle, membrane vesicle, high density lipoprotein and circulating. We also implemented a new web interface to improve the user experience.

Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records

BACKGROUND

Intensive-care units (ICUs) treat the most critically ill patients, which is complicated by the heterogeneity of the diseases that they encounter. Severity scores based mainly on acute physiology measures collected at ICU admission are used to predict mortality, but are non-specific, and predictions for individual patients can be inaccurate. We investigated whether inclusion of long-term disease history before ICU admission improves mortality predictions.

METHODS

Registry data for long-term disease histories for more than 230 000 Danish ICU patients were used in a neural network to develop an ICU mortality prediction model. Long-term disease histories and acute physiology measures were aggregated to predict mortality risk for patients for whom both registry and ICU electronic patient record data were available. We compared mortality predictions with admission scores on the Simplified Acute Physiology Score (SAPS) II, the Acute Physiologic Assessment and Chronic Health Evaluation (APACHE) II, and the best available multimorbidity score, the Multimorbidity Index. An external validation set from an additional hospital was acquired after model construction to confirm the validity of our model. During initial model development data were split into a training set (85%) and an independent test set (15%), and a five-fold cross-validation was done during training to avoid overfitting. Neural networks were trained for datasets with disease history of 1 month, 3 months, 6 months, 1 year, 2·5 years, 5 years, 7·5 years, 10 years, and 23 years before ICU admission.

FINDINGS

Mortality predictions with a model based solely on disease history outperformed the Multimorbidity Index (Matthews correlation coefficient 0·265 vs 0·065), and performed similarly to SAPS II and APACHE II (Matthews correlation coefficient with disease history, age, and sex 0·326 vs 0·347 and 0·300 for SAPS II and APACHE II, respectively). Diagnoses up to 10 years before ICU admission affected current mortality prediction. Aggregation of previous disease history and acute physiology measures in a neural network yielded the most precise predictions of in-hospital mortality (Matthews correlation coefficient 0·391 for in-hospital mortality compared with 0·347 with SAPS II and 0·300 with APACHE II). These results for the aggregated model were validated in an external independent dataset of 1528 patients (Matthews correlation coefficient for prediction of in-hospital mortality 0·341).

INTERPRETATION

Longitudinal disease-spectrum-wide data available before ICU admission are useful for mortality prediction. Disease history can be used to differentiate mortality risk between patients with similar vital signs with more precision than SAPS II and APACHE II scores. Machine learning models can be deconvoluted to generate novel understandings of how ICU patient features from long-term and short-term events interact with each other. Explainable machine learning models are key in clinical settings, and our results emphasise how to progress towards the transformation of advanced models into actionable, transparent, and trustworthy clinical tools.

FUNDING

Novo Nordisk Foundation and Innovation Fund Denmark.

The impact of the protein interactome on the syntenic structure of mammalian genomes

Conserved synteny denotes evolutionary preserved gene order across species. It is not well understood to which degree functional relationships between genes are preserved in syntenic blocks. Here we investigate whether protein-coding genes conserved in mammalian syntenic blocks encode gene products that serve the common functional purpose of interacting at protein level, i.e. connectivity. High connectivity among protein-protein interactions (PPIs) was only moderately associated with conserved synteny on a genome-wide scale. However, we observed a smaller subset of 3.6% of all syntenic blocks with high-confidence PPIs that had significantly higher connectivity than expected by random. Additionally, syntenic blocks with high-confidence PPIs contained significantly more chromatin loops than the remaining blocks, indicating functional preservation among these syntenic blocks. Conserved synteny is typically defined by sequence similarity. In this study, we also examined whether a functional relationship, here PPI connectivity, can identify syntenic blocks independently of orthology. While orthology-based syntenic blocks with high-confident PPIs and the connectivity-based syntenic blocks largely overlapped, the connectivity-based approach identified additional syntenic blocks that were not found by conventional sequence-based methods alone. Additionally, the connectivity-based approach enabled identification of potential orthologous genes between species. Our analyses demonstrate that subsets of syntenic blocks are associated with highly connected proteins, and that PPI connectivity can be used to detect conserved synteny even if sequence conservation drifts beyond what orthology algorithms normally can identify.

Sequencing and de novo assembly of 150 genomes from Denmark as a population reference

Hundreds of thousands of human genomes are now being sequenced to characterize genetic variation and use this information to augment association mapping studies of complex disorders and other phenotypic traits. Genetic variation is identified mainly by mapping short reads to the reference genome or by performing local assembly. However, these approaches are biased against discovery of structural variants and variation in the more complex parts of the genome. Hence, large-scale de novo assembly is needed. Here we show that it is possible to construct excellent de novo assemblies from high-coverage sequencing with mate-pair libraries extending up to 20 kilobases. We report de novo assemblies of 150 individuals (50 trios) from the GenomeDenmark project. The quality of these assemblies is similar to those obtained using the more expensive long-read technology. We use the assemblies to identify a rich set of structural variants including many novel insertions and demonstrate how this variant catalogue enables further deciphering of known association mapping signals. We leverage the assemblies to provide 100 completely resolved major histocompatibility complex haplotypes and to resolve major parts of the Y chromosome. Our study provides a regional reference genome that we expect will improve the power of future association mapping studies and hence pave the way for precision medicine initiatives, which now are being launched in many countries including Denmark.

Klinefelter syndrome comorbidities linked to increased X chromosome gene dosage and altered protein interactome activity

Klinefelter syndrome (KS) (47,XXY) is the most common male sex chromosome aneuploidy. Diagnosis and clinical supervision remain a challenge due to varying phenotypic presentation and insufficient characterization of the syndrome. Here we combine health data-driven epidemiology and molecular level systems biology to improve the understanding of KS and the molecular interplay influencing its comorbidities. In total, 78 overrepresented KS comorbidities were identified using in- and out-patient registry data from the entire Danish population covering 6.8 million individuals. The comorbidities extracted included both clinically well-known (e.g. infertility and osteoporosis) and still less established KS comorbidities (e.g. pituitary gland hypofunction and dental caries). Several systems biology approaches were applied to identify key molecular players underlying KS comorbidities: Identification of co-expressed modules as well as central hubs and gene dosage perturbed protein complexes in a KS comorbidity network build from known disease proteins and their protein–protein interactions. The systems biology approaches together pointed to novel aspects of KS disease phenotypes including perturbed Jak-STAT pathway, dysregulated genes important for disturbed immune system (IL4), energy balance (POMC and LEP) and erythropoietin signalling in KS. We present an extended epidemiological study that links KS comorbidities to the molecular level and identify potential causal players in the disease biology underlying the identified comorbidities.

MicroRNAs, Regulatory Networks, and Comorbidities: Decoding Complex Systems

MicroRNAs (miRNAs) are small noncoding RNAs involved in the posttranscriptional regulation of messenger RNAs (mRNAs). Each miRNA targets a specific set of mRNAs. Upon binding the miRNA inhibits mRNA translation or facilitate mRNA degradation. miRNAs are frequently deregulated in several pathologies including cancer and cardiovascular diseases. Since miRNAs have a crucial role in fine-tuning the expression of their targets, they have been proposed as biomarkers of disease progression and prognostication.In this chapter we discuss different approaches for computational predictions of miRNA targets based on sequence complementarity and integration of expression data. In the last section of the chapter we discuss new opportunities in the study of miRNA regulatory networks in the context of temporal disease progression and comorbidities.

Novel genes involved in pathophysiology of gonadotropin-dependent adrenal tumors in mice

Specific inbred strains and transgenic inhibin-α Simian Virus 40 T antigen (inhα/Tag) mice are genetically susceptible to gonadectomy-induced adrenocortical neoplasias. We identified altered gene expression in prepubertally gonadectomized (GDX) inhα/Tag and wild-type (WT) mice. Besides earlier reported Gata4 and Lhcgr, we found up-regulated Esr1, Prlr-rs1, and down-regulated Grb10, Mmp24, Sgcd, Rerg, Gnas, Nfatc2, Gnrhr, Igf2 in inhα/Tag adrenal tumors. Sex-steroidogenic enzyme genes expression (Srd5a1, Cyp19a1) was up-regulated in tumors, but adrenal-specific steroidogenic enzyme (Cyp21a1, Cyp11b1, Cyp11b2) down-regulated. We localized novel Lhcgr transcripts in adrenal cortex parenchyma and in non-steroidogenic A cells, in GDX WT and in intact WT mice. We identified up-regulated Esr1 as a potential novel biomarker of gonadectomy-induced adrenocortical tumors in inhα/Tag mice presenting with an inverted adrenal-to-gonadal steroidogenic gene expression profile. A putative normal adrenal remodeling or tumor suppressor role of the down-regulated genes (e.g. Grb10, Rerg, Gnas, and Nfatc2) in the tumors remains to be addressed.

Chromosome-wise Protein Interaction Patterns and Their Impact on Functional Implications of Large-Scale Genomic Aberrations

Gene copy-number changes influence phenotypes through gene-dosage alteration and subsequent changes of protein complex stoichiometry. Human trisomies where gene copy numbers are increased uniformly over entire chromosomes provide generic cases for studying these relationships. In most trisomies, gene and protein level alterations have fatal consequences. We used genome-wide protein-protein interaction data to identify chromosome-specific patterns of protein interactions. We found that some chromosomes encode proteins that interact infrequently with each other, chromosome 21 in particular. We combined the protein interaction data with transcriptome data from human brain tissue to investigate how this pattern of global interactions may affect cellular function. We identified highly connected proteins that also had coordinated gene expression. These proteins were associated with important neurological functions affecting the characteristic phenotypes for Down syndrome and have previously been validated in mouse knockout experiments. Our approach is general and applicable to other gene-dosage changes, such as arm-level amplifications in cancer.

The potential role of Alu Y in the development of resistance to SN38 (Irinotecan) or oxaliplatin in colorectal cancer

BACKGROUND

Irinotecan (SN38) and oxaliplatin are chemotherapeutic agents used in the treatment of colorectal cancer. However, the frequent development of resistance to these drugs represents a considerable challenge in the clinic. Alus as retrotransposons comprise 11% of the human genome. Genomic toxicity induced by carcinogens or drugs can reactivate Alus by altering DNA methylation. Whether or not reactivation of Alus occurs in SN38 and oxaliplatin resistance remains unknown.

RESULTS

We applied reduced representation bisulfite sequencing (RRBS) to investigate the DNA methylome in SN38 or oxaliplatin resistant colorectal cancer cell line models. Moreover, we extended the RRBS analysis to tumor tissue from 14 patients with colorectal cancer who either did or did not benefit from capecitabine + oxaliplatin treatment. For the clinical samples, we applied a concept of 'DNA methylation entropy' to estimate the diversity of DNA methylation states of the identified resistance phenotype-associated methylation loci observed in the cell line models. We identified different loci being characteristic for the different resistant cell lines. Interestingly, 53% of the identified loci were Alu sequences- especially the Alu Y subfamily. Furthermore, we identified an enrichment of Alu Y sequences that likely results from increased integration of new copies of Alu Y sequence in the drug-resistant cell lines. In the clinical samples, SOX1 and other SOX gene family members were shown to display variable DNA methylation states in their gene regions. The Alu Y sequences showed remarkable variation in DNA methylation states across the clinical samples.

CONCLUSION

Our findings imply a crucial role of Alu Y in colorectal cancer drug resistance. Our study underscores the complexity of colorectal cancer aggravated by mobility of Alu elements and stresses the importance of personalized strategies, using a systematic and dynamic view, for effective cancer therapy.

Novel variation and de novo mutation rates in population-wide de novo assembled Danish trios

Building a population-specific catalogue of single nucleotide variants (SNVs), indels and structural variants (SVs) with frequencies, termed a national pan-genome, is critical for further advancing clinical and public health genetics in large cohorts. Here we report a Danish pan-genome obtained from sequencing 10 trios to high depth (50 × ). We report 536k novel SNVs and 283k novel short indels from mapping approaches and develop a population-wide de novo assembly approach to identify 132k novel indels larger than 10 nucleotides with low false discovery rates. We identify a higher proportion of indels and SVs than previous efforts showing the merits of high coverage and de novo assembly approaches. In addition, we use trio information to identify de novo mutations and use a probabilistic method to provide direct estimates of 1.27e−8 and 1.5e−9 per nucleotide per generation for SNVs and indels, respectively.

The generation of a national pan-genome, a population-specific catalogue of genetic variation, may advance the impact of clinical genetics studies. Here the Besenbacher et al. carry out deep sequencing and de novo assembly of 10 parent–child trios to generate a Danish pan-genome that provides insight into structural variation, de novo mutation rates and variant calling.

The transcriptome of corona radiata cells from individual MІІ oocytes that after ICSI developed to embryos selected for transfer: PCOS women compared to healthy women

Background

Corona radiata cells (CRCs) refer to the fraction of cumulus cells just adjacent to the oocyte. The CRCs are closely connected to the oocyte throughout maturation and their gene expression profiles might reflect oocyte quality. Polycystic ovary syndrome (PCOS) is a common cause of infertility. It is controversial whether PCOS associate with diminished oocyte quality. The purpose of this study was to compare individual human CRC samples between PCOS patients and controls.

Methods

All patients were stimulated by the long gonadotropin-releasing hormone (GnRH) agonist protocol. The CRC samples originated from individual oocytes developing into embryos selected for transfer. CRCs were isolated in a two-step denudation procedure, separating outer cumulus cells from the inner CRCs. Extracted RNA was amplified and transcriptome profiling was performed with Human Agilent® arrays.

Results

The transcriptomes of CRCs showed no individual genes with significant differential expression between PCOS and controls, but gene set enrichment analysis identified several cell cycle- and DNA replication pathways overexpressed in PCOS CRCs (FDR < 0.05). Five of the genes contributing to the up-regulated cell cycle pathways in the PCOS CRCs were selected for qRT-PCR validation in ten PCOS and ten control CRC samples. qRT-PCR confirmed significant up-regulation in PCOS CRCs of cell cycle progression genes HIST1H4C (FC = 2.7), UBE2C (FC = 2.6) and cell cycle related transcription factor E2F4 (FC = 2.5).

Conclusion

The overexpression of cell cycle-related genes and cell cycle pathways in PCOS CRCs could indicate a disturbed or delayed final maturation and differentiation of the CRCs in response to the human chorionic gonadotropin (hCG) surge. However, this had no effect on the in vitro development of the corresponding embryos. Future studies are needed to clarify whether the up-regulated cell cycle pathways in PCOS CRCs have any clinical implications.

Electronic supplementary material

The online version of this article (doi:10.1186/s13048-014-0110-6) contains supplementary material, which is available to authorized users.

P4-01-17: TIMP-1 Over-Expression Confers Resistance of MCF-7 Breast Cancer Cells to Fulvestrant

Background

Endocrine resistance represents a major challenge in the management of estrogen receptor (ER) positive breast cancer. Currently no predictive biomarkers for endocrine resistance in ERpositive breast cancer patients are in clinical use.

In a clinical study, patients with metastatic breast cancer and high levels of serum Tissue Inhibitor of Metalloproteinases-1 (TIMP-1) had less benefit from endocrine therapy than patients with a lower level of serum TIMP-1 [1].

Therefore, we evaluated the association between TIMP-1 and response to endocrine therapy using an in vitro approach.

We have previously presented initial results on TIMP-1 and response to endocrine therapy [2].

Materials and Methods: MCF-7 cells were stably transfected with pcDNA3.1(Hyg)-TIMP-1 plasmid, and a panel of 11 subclones with different expression levels of TIMP-1 was generated. TIMP-1 expression levels were confirmed using enzyme-linked immunosorbent assay (ELISA).

Four subclones with high or low TIMP-1 expression were analyzed for the growth response to estrogen, 4-hydroxytamoxifen and fulvestrant. These four subclones were analyzed for protein expression by western blotting. All subclones were analyzed by whole human genome oligo microarrays 4×44K for determination of gene expression levels. Data were analyzed using the limma R/Bioconductor package. Paired-end Solexa sequencing was applied to selected subclones with high and low TIMP-1 levels to identify transcriptomic changes.

Results: High expression of TIMP-1 was associated with resistance to fulvestrant, whereas growth response to either estrogen or 4-hydroxytamoxifen was independent of TIMP-1 expression levels. High expression of TIMP-1 protein and mRNA was associated with undetectable levels of progesterone receptor (PgR) protein and mRNA whereas ER protein and mRNA levels were unaffected by TIMP-1. To characterize the potential role of TIMP-1 in estrogen signaling we analyzed the expression of reported estrogen-responsive genes and no general change was observed. We identified genes that correlated positively or negatively to TIMP-1 expression. Among the identified genes was PgR, which is a direct target for ER.

Conclusion: Our data suggest that a high expression of TIMP-1 in vitro is associated with resistance to fulvestrant but not to 4-hydroxytamoxifen. Estrogen-regulated genes are not generally affected by changes in TIMP-1 expression levels and therefore TIMP-1 appears to affect endocrine resistance through other mechanisms than globally regulating ER signaling. However, high expression of TIMP-1 is associated with loss of PgR and this may be related to the resistance towards fulvestrant.

References:

[1] Lipton, A et al: Serum TIMP-1 and Response to the Aromatase Inhibitor Letrozole Versus Tamoxifen in Metastatic Breast Cancer. J Clin Oncol; 2008; 26;(16); 2653–8

[2] Effect of TIMP-1 Overexpression on Endocrine Sensitivity of MCF-7 ER-positive Human Breast Cancer Cells In Vitro. Cancer Res; 2009; 69(24 suppl); abstract nr 2029

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-01-17.