Predicting Hypoxia Using Machine Learning: Systematic Review.

Background: Hypoxia is an important risk factor and indicator for the declining health of inpatients. Predicting future hypoxic events using machine learning is a prospective area of study to facilitate time-critical interventions to counter patient health deterioration. Objective: This systematic review aims to summarize and compare previous efforts to predict hypoxic events in the hospital setting using machine learning with respect to their methodology, predictive performance, and assessed population. Methods: A systematic literature search was performed using Web of Science, Ovid with Embase and MEDLINE, and Google Scholar. Studies that investigated hypoxia or hypoxemia of hospitalized patients using machine learning models were considered. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool. Results: After screening, a total of 12 papers were eligible for analysis, from which 32 models were extracted. The included studies showed a variety of population, methodology, and outcome definition. Comparability was further limited due to unclear or high risk of bias for most studies (10/12, 83%). The overall predictive performance ranged from moderate to high. Based on classification metrics, deep learning models performed similar to or outperformed conventional machine learning models within the same studies. Models using only prior peripheral oxygen saturation as a clinical variable showed better performance than models based on multiple variables, with most of these studies (2/3, 67%) using a long short-term memory algorithm. Conclusions: Machine learning models provide the potential to accurately predict the occurrence of hypoxic events based on retrospective data. The heterogeneity of the studies and limited generalizability of their results highlight the need for further validation studies to assess their predictive performance.

Individualized targeted treatment in a case of a rare TFG::ROS1 fusion positive inflammatory myofibroblastic tumor (IMT).

BACKGROUND: Inflammatory myofibroblastic tumor (IMTs) are rare mesenchymal neoplasms with slow growth. Resection is considered as therapeutic standard, with chemotherapy being insufficiently effective in advanced disease. ALK translocations are present in 50% of cases, ROS1 fusions (YWHAE::ROS1, TFG::ROS1) are extremely rare. Here, we present a case with TFG::ROS1 fusion and highlight the significance of molecular tumor boards (MTBs) in clinical precision oncology for post-last-line therapy. CASE PRESENTATION: A 32-year-old woman presented with IMT diagnosed at age 27 for biopsy and treatment evaluation. Previous treatments included multiple resections and systemic therapy with vinblastine, cyclophosphamide, and methotrexate. A computed tomography scan showed extensive tumor infiltration of the psoas muscles and the posterior abdomen. Next generation sequencing revealed an actionable ROS1 fusion (TFG::ROS1) with breakpoints at exon 4/35 including the kinase domain and activating the RAS-pathway. TFG, the Trk-fused gene, exerts functions such as intracellular trafficking and exhibits high sequence homology between species. Based on single reports about efficacy of ROS1-targeting in ROS1 translocation positive IMTs the patient was started on crizotinib, an ATP-competitive small molecule c-MET, ALK and ROS1-inhibitor. With a follow-up of more than 9 months, the patient continues to show a profound response with major tumor regression, improved quality of life and no evidence for severe adverse events. CONCLUSION: This case underscores the importance of the availability of modern molecular diagnostics and interdisciplinarity in precision oncology to identify rare, disease-defining genotypes that make an otherwise difficult-to-treat disease targetable.

[NGS-based molecular genetics of leukemia-a powerful and decentralized approach]. / NGS-basierte Molekulargenetik der Leukämie ­ ein leistungsfähiger und dezentraler Lösungsansatz.

The diagnosis of acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS), originally based on morphological assessment alone, has to bring together more and more disciplines. Today, modern AML/MDS diagnostics rely on cytomorphology, cytochemistry, immunophenotyping, cytogenetics, and molecular genetics. Only the integration of all these methods allows a comprehensive and complementary characterization of each case, which is a prerequisite for optimal AML/MDS diagnosis and treatment. In the following, we present why multidisciplinary and local diagnosis is essential today and will become even more important in the future, especially in the context of precision medicine. We present our idea and strategy implemented at Augsburg University Hospital, which has realized multidisciplinary diagnostics in AML/MDS in an interdisciplinary and decentralized approach. In particular, this includes the recent technical advances that molecular genetics provides with modern methods. The enormous amount of data generated by these techniques represents a major challenge, but also a unique opportunity. We will reflect on how this increase in knowledge can be integrated into routine practice to lead the way for personalized medicine in AML/MDS to improve patient care.

Predictive value of DNA methylation patterns in AML patients treated with an azacytidine containing induction regimen.

BACKGROUND: Acute myeloid leukemia (AML) is a heterogeneous disease with a poor prognosis. Dysregulation of the epigenetic machinery is a significant contributor to disease development. Some AML patients benefit from treatment with hypomethylating agents (HMAs), but no predictive biomarkers for therapy response exist. Here, we investigated whether unbiased genome-wide assessment of pre-treatment DNA-methylation profiles in AML bone marrow blasts can help to identify patients who will achieve a remission after an azacytidine-containing induction regimen. RESULTS: A total of n = 155 patients with newly diagnosed AML treated in the AMLSG 12-09 trial were randomly assigned to a screening and a refinement and validation cohort. The cohorts were divided according to azacytidine-containing induction regimens and response status. Methylation status was assessed for 664,227 500-bp-regions using methyl-CpG immunoprecipitation-seq, resulting in 1755 differentially methylated regions (DMRs). Top regions were distilled and included genes such as WNT10A and GATA3. 80% of regions identified as a hit were represented on HumanMethlyation 450k Bead Chips. Quantitative methylation analysis confirmed 90% of these regions (36 of 40 DMRs). A classifier was trained using penalized logistic regression and fivefold cross validation containing 17 CpGs. Validation based on mass spectra generated by MALDI-TOF failed (AUC 0.59). However, discriminative ability was maintained by adding neighboring CpGs. A recomposed classifier with 12 CpGs resulted in an AUC of 0.77. When evaluated in the non-azacytidine containing group, the AUC was 0.76. CONCLUSIONS: Our analysis evaluated the value of a whole genome methyl-CpG screening assay for the identification of informative methylation changes. We also compared the informative content and discriminatory power of regions and single CpGs for predicting response to therapy. The relevance of the identified DMRs is supported by their association with key regulatory processes of oncogenic transformation and support the idea of relevant DMRs being enriched at distinct loci rather than evenly distribution across the genome. Predictive response to therapy could be established but lacked specificity for treatment with azacytidine. Our results suggest that a predictive epigenotype carries its methylation information at a complex, genome-wide level, that is confined to regions, rather than to single CpGs. With increasing application of combinatorial regimens, response prediction may become even more complicated.

Progressive multifocal leukoencephalopathy associated with chemotherapy induced lymphocytopenia in solid tumors - case report of an underestimated complication.

Background: JC virus reactivation causing progressive multifocal leukoencephalopathy (PML) occurs preferentially in human immunodeficiency virus (HIV) positive individuals or patients suffering from hematologic neoplasms due to impaired viral control. Reactivation in patients suffering from solid malignancies is rarely described in published literature. Case Presentation: Here we describe a case of PML in a male patient suffering from esophageal cancer who underwent neoadjuvant radiochemotherapy and surgical resection in curative intent resulting in complete tumor remission. The radiochemotherapy regimen contained carboplatin and paclitaxel (CROSS protocol). Since therapy onset, the patient presented with persistent and progredient leukopenia and lymphopenia in absence of otherwise known risk factors for PML. Symptom onset, which comprised aphasia, word finding disorder, and paresis, was apparent 7 months after therapy initiation. There was no relief in symptoms despite standard of care PML directed supportive therapy. The patient died two months after therapy onset. Conclusion: PML is a very rare event in solid tumors without obvious states of immununosuppression and thus harbors the risk of unawareness. The reported patient suffered from lymphopenia, associated with systemic therapy, but was an otherwise immunocompetent individual. In case of neurologic impairment in patients suffering from leukopenia, PML must be considered - even in the absence of hematologic neoplasia or HIV infection.

DNA methylation epitypes highlight underlying developmental and disease pathways in acute myeloid leukemia.

Acute myeloid leukemia (AML) is a molecularly complex disease characterized by heterogeneous tumor genetic profiles and involving numerous pathogenic mechanisms and pathways. Integration of molecular data types across multiple patient cohorts may advance current genetic approaches for improved subclassification and understanding of the biology of the disease. Here, we analyzed genome-wide DNA methylation in 649 AML patients using Illumina arrays and identified a configuration of 13 subtypes (termed "epitypes") using unbiased clustering. Integration of genetic data revealed that most epitypes were associated with a certain recurrent mutation (or combination) in a majority of patients, yet other epitypes were largely independent. Epitypes showed developmental blockage at discrete stages of myeloid differentiation, revealing epitypes that retain arrested hematopoietic stem-cell-like phenotypes. Detailed analyses of DNA methylation patterns identified unique patterns of aberrant hyper- and hypomethylation among epitypes, with variable involvement of transcription factors influencing promoter, enhancer, and repressed regions. Patients in epitypes with stem-cell-like methylation features showed inferior overall survival along with up-regulated stem cell gene expression signatures. We further identified a DNA methylation signature involving STAT motifs associated with FLT3-ITD mutations. Finally, DNA methylation signatures were stable at relapse for the large majority of patients, and rare epitype switching accompanied loss of the dominant epitype mutations and reversion to stem-cell-like methylation patterns. These results show that DNA methylation-based classification integrates important molecular features of AML to reveal the diverse pathogenic and biological aspects of the disease.

Netboost: Boosting-Supported Network Analysis Improves High-Dimensional Omics Prediction in Acute Myeloid Leukemia and Huntington's Disease.

State-of-the art selection methods fail to identify weak but cumulative effects of features found in many high-dimensional omics datasets. Nevertheless, these features play an important role in certain diseases. We present Netboost, a three-step dimension reduction technique. First, a boosting-based filter is combined with the topological overlap measure to identify the essential edges of the network. Second, sparse hierarchical clustering is applied on the selected edges to identify modules and finally module information is aggregated by the first principal components. We demonstrate the application of the newly developed Netboost in combination with CoxBoost for survival prediction of DNA methylation and gene expression data from 180 acute myeloid leukemia (AML) patients and show, based on cross-validated prediction error curve estimates, its prediction superiority over variable selection on the full dataset as well as over an alternative clustering approach. The identified signature related to chromatin modifying enzymes was replicated in an independent dataset, the phase II AMLSG 12-09 study. In a second application we combine Netboost with Random Forest classification and improve the disease classification error in RNA-sequencing data of Huntington's disease mice. Netboost is a freely available Bioconductor R package for dimension reduction and hypothesis generation in high-dimensional omics applications.