M-ClustEHR: A multimodal clustering approach for electronic health records.

Sepsis refers to a potentially life-threatening situation where the immune system of the human body has an extreme response to an infection. In the presence of underlying comorbidities, the situation can become even worse and result in death. Employing unsupervised machine learning techniques, such as clustering, can assist in providing a better understanding of patient phenotypes by unveiling subgroups characterized by distinct sepsis progression and treatment patterns. More concretely, this study introduces M-ClustEHR, a clustering approach that utilizes medical data of multiple modalities by employing a multimodal autoencoder for learning comprehensive sepsis patient representations. M-ClustEHR consistently outperforms traditional clustering approaches in terms of several internal clustering performance metrics, as well as cluster stability in identifying phenotypes in the sepsis cohort. The unveiled patterns, supported by existing medical literature and clinicians, highlight the importance of multimodal clustering for advancing personalized sepsis care.

[Overview of immune-related side effects from immune checkpoint inhibitors. Part 1: Gastrointestinal, lung and kidney toxicity]. / Immunrelaterade biverkningar i tarm, lever, lungor och njurar.

In the past decade, immunotherapy with checkpoint inhibitors has revolutionized the field of oncology. Checkpoint inhibitors have been approved for several types of cancer and thousands of patients in Sweden now receive oncological immunotherapy annually. Immune-related side effects are common and can occur in almost any organ. These side effects are different from those that occur with traditional oncological treatments. The side effects are usually mild, but can be serious and even lethal. In a short time, health care providers have had to readjust to be able to handle these side effects. Early and correct diagnosis of immune-related side effects, proper management and a multidisciplinary approach is crucial. Here, we give an overview of the presentation, diagnosis and treatment of immune-related side effects, with emphasis of those occurring in gastrointestinal organs, lungs and kidneys.

Lack of cytoplasmic ERK activation is an independent adverse prognostic factor in primary cutaneous melanoma.

The aim of this study was to estimate the impact on survival of NRAS and BRAF mutations and activation of Akt and extracellular signal-regulated kinase (ERK) in primary melanomas. A cohort of 57 primary cutaneous T1-2 melanoma tumors was analyzed. Mutation frequency for both genes was 61% (NRAS 26% and BRAF 39%). In a univariate analysis, shorter overall survival was associated with the presence of ulceration (P=0.001) and BRAF exon 15 mutations (P=0.005) as well as the absence of nuclear activation of Akt (P=0.022) and of cytoplasmic activation of ERK (P=0.003). Unexpectedly, ulceration was a significant adverse prognostic factor only in melanomas with BRAF mutations, whereas there was no effect of ulceration on overall survival in tumors with wild-type BRAF. A multivariate analysis showed that significant independent adverse survival prognostic markers were absence of cytoplasmic activation of ERK (P=0.007) and ulceration (P=0.008), whereas BRAF exon 15 mutation status showed a nonsignificant trend (P=0.066). The absence of cytoplasmic ERK activation in poor prognosis T1-2 melanomas may be associated with activation of some other uncharacterized pathway leading to tumor progression and adverse outcome. Immunohistochemical analysis of cytoplasmic phosphorylated ERK could be used as a prognostic marker in primary melanomas if confirmed in another data set.