Risk of cancer in patients with thyroid disease and venous thromboembolism

Association between thyroid disorders and extra-thyroidal cancers, a review

Thyroid hormone has been shown to have both tumor-promoting and tumor-suppressing actions, which has led to significant debate over its involvement in the development of cancer. Proliferation, apoptosis, invasiveness, and angiogenesis are all aspects of cancer that are affected by the thyroid hormones T3 and T4, according to research conducted in animal models and in vitro experiments. The effects of thyroid hormones on cancer cells are mediated by many non-genomic mechanisms, one of which involves the activation of the plasma membrane receptor integrin αvβ3. Typically, abnormal amounts of thyroid hormones are linked to a higher occurrence of cancer. Both benign and malignant thyroid disorders were found to be associated with an increased risk of extra-thyroidal malignancies, specifically colon, breast, prostate, melanoma, and lung cancers. The purpose of this review was to shed light on this link to define which types of cancer are sensitive to thyroid hormones and, as a result, are anticipated to respond favorably to treatment of the thyroid hormone axis.

Clinical explainable machine learning models for early identification of patients at risk of hospital-acquired urinary tract infection

BACKGROUND

Machine learning (ML) models for early identification of patients at risk of hospital-acquired urinary tract infections (HA-UTI) may enable timely and targeted preventive and therapeutic strategies. However, clinicians are often challenged in the interpretation of the predictive outcomes provided by the ML models, which often reach different performances.

AIM

To train ML models for predicting patients at risk of HA-UTI using available data from electronic health records at the time of hospital admission. We focused on the performance of different ML models and clinical explainability.

METHODS

This retrospective study investigated patient data representing 138.560 hospital admissions in the North Denmark Region from 01.01.2017 to 31.12.2018. We extracted 51 health socio-demographic and clinical features in a full dataset and used the χ² test in addition to expert knowledge for feature selection, resulting in two reduced datasets. Seven different ML models were trained and compared between the three datasets. We applied the SHapley Additive exPlanation (SHAP) method to support population- and patient-level explainability.

FINDINGS

The best-performing ML model was a neural network based on the full dataset, reaching an area under the curve (AUC) of 0.758. The neural network was also the best-performing ML model based on the reduced datasets, reaching an AUC of 0.746. Clinical explainability was demonstrated with a SHAP summary- and forceplot.

CONCLUSION

Within 24h of hospital admission, the ML models were able to identify patients at risk of developing HA-UTI, providing new opportunities to develop efficient strategies for the prevention of HA-UTI. Using SHAP, we demonstrate how risk predictions can be explained at individual patient level and for the patient population in general.

Thrombosis Risk Associated with Head and Neck Cancer: A Review

Venous thromboembolism (VTE) is a common complication for cancer patients. VTE-associated risk varies according to the type of tumor disease. Head and neck cancer is a common cancer worldwide, and most tumors are squamous cell carcinomas due to tobacco and alcohol abuse. The risk of VTE associated with head and neck (H&N) cancer is considered empirically low, but despite the high incidence of H&N cancer, few data are available on this cancer; thus, it is difficult to state the risk of VTE. Our review aims to clarify this situation and tries to assess the real VTE risk associated with H&N cancer. We report that most clinical studies have concluded that there is a very low thrombosis risk associated with H&N cancer. Even with the biases that often exist, this clinical review seems to confirm that the risk of VTE was empirically hypothesized. Furthermore, we highlight that H&N cancer has all the biological features of a cancer associated with a high thrombosis risk, including a strong expression of procoagulant proteins, modified thrombosis/fibrinolysis mechanisms, and secretions of procoagulant microparticles and procoagulant cytokines. Thus, this is a paradoxical situation, and some undiscovered mechanisms that could explain this clinical biological ambivalence might exist.