An Evaluation of the Efficacy of Machine Learning in Predicting Thyrotoxicosis and Hypothyroidism: A Comparative Assessment of Biochemical Test Parameters Used in Different Health Checkups.

Objective This study assessed the efficacy of machine learning in predicting thyrotoxicosis and hypothyroidism [thyroid-stimulating hormone (TSH) >10.0 mIU/L] by leveraging age and sex as variables and integrating biochemical test parameters used by the Japan Society of Health Evaluation and Promotion (JHEP) and the Japan Society of Ningen Dock (JND). Subjects and Methods Our study included 20,653 untreated patients with Graves' disease, 3,435 untreated patients with painless thyroiditis, 4,266 healthy individuals, and 18,937 untreated patients with Hashimoto's thyroiditis. Machine learning was conducted using Prediction One on three distinct datasets: the Ito dataset (age, sex, and 30 blood tests and biochemical test data), the JHEP dataset (age, sex, and TP, T-Bil, AST, ALT, γGTP, ALP, CRE, UA, and T-Cho test data), and the JND dataset (age, sex, and AST, ALT, γGTP, CRE, and UA test data). Results The results for distinguishing thyrotoxicosis patients from the healthy control group showed that the JHEP dataset yielded substantial discriminative capacity with an area under the curve (AUC) of 0.966, sensitivity of 92.2%, specificity of 89.1%, and accuracy of 91.7%. The JND dataset displayed similar robustness, with an AUC of 0.948, sensitivity of 92.0%, specificity of 81.3%, and accuracy of 90.4%. Differentiating hypothyroid patients from the healthy control group yielded similarly robust performances, with the JHEP dataset yielding AUC, sensitivity, specificity, and accuracy values of 0.864, 84.2%, 72.1%, and 77.4%, respectively, and the JND dataset yielding values of 0.840, 83.2%, 67.2%, and 74.3%, respectively. Conclusions Machine learning is a potent screening tool for thyrotoxicosis and hypothyroidism.

Does Age or Sex Relate to Severity or Treatment Prognosis in Graves' Disease?

Background: The prognosis of Graves' disease (GD) is reportedly related to sex, age, and genetic factors, although there is no consensus. The objective of this study was to investigate the relationship between severity and prognosis of GD and sex or age. Methods: Subjects were patients newly diagnosed with GD between January 2005 and June 2019, and medical records were retrospectively reviewed. Patients diagnosed between January 2009 and December 2010 and followed up for at least 12 months were enrolled. Patients were divided into nine age-stratified groups. Remission was defined as maintenance of a euthyroid state for more than one year after withdrawal of antithyroid drugs (ATDs). Results: Participants comprised 21,633 patients (3954 males, 17,679 females). Initial free triiodothyronine (fT3) and free thyroxine (fT4) levels significantly decreased with increasing age, including after sex stratification. fT4 was significantly higher in males than females aged 20-39 years. In 2191 patients treated with ATDs alone, median durations until remission were 37.7 and 30.6 months in males and females, respectively. Remission and recurrence were observed in 1391 patients (204 males, 1187 females) and 262 patients (37 males, 225 females), respectively. By Kaplan-Meier analyses, males required a significantly longer time to achieve remission than females (p < 0.0001), although there were no significant age-related differences (p = 0.08). Cox proportional hazard modeling showed a 41% higher hazard ratio (HR) for remission in females than males (adjusted HRs [aHR] confidence interval [CI] = 1.41 [1.21-1.64]), and each additional 10 years of age had a 14% lower rate of recurrence (age [per 10-year increase], aHR [CI] = 0.86 [0.78-0.94]); no significant relationship between recurrence rate and sex was identified. Conclusions: Severity of hyperthyroidism in GD was significantly higher in males in their 20s and 30s, declining with advancing age in both sexes. Females were more likely to achieve remission than males, and younger patients had a higher risk of recurrence, although recurrence was unrelated to sex.

Assessment of Adenosine Triphosphatase Phospholipid Transporting 8B1 (ATP8B1) Function in Patients With Cholestasis With ATP8B1 Deficiency by Using Peripheral Blood Monocyte-Derived Macrophages.

Adenosine triphosphatase phospholipid transporting 8B1 (ATP8B1) deficiency, an ultrarare autosomal recessive liver disease, includes severe and mild clinical forms, referred to as progressive familial intrahepatic cholestasis type 1 (PFIC1) and benign recurrent intrahepatic cholestasis type 1 (BRIC1), respectively. There is currently no practical method for determining PFIC1 or BRIC1 at an early disease course phase. Herein, we assessed the feasibility of developing a diagnostic method for PFIC1 and BRIC1. A nationwide Japanese survey conducted since 2015 identified 25 patients with cholestasis with ATP8B1 mutations, 15 of whom agreed to participate in the study. Patients were divided for analysis into PFIC1 (n = 10) or BRIC1 (n = 5) based on their disease course. An in vitro mutagenesis assay to evaluate pathogenicity of ATP8B1 mutations suggested that residual ATP8B1 function in the patients could be used to identify clinical course. To assess their ATP8B1 function more simply, human peripheral blood monocyte-derived macrophages (HMDMs) were prepared from each patient and elicited into a subset of alternatively activated macrophages (M2c) by interleukin-10 (IL-10). This was based on our previous finding that ATP8B1 contributes to polarization of HMDMs into M2c. Flow cytometric analysis showed that expression of M2c-related surface markers cluster of differentiation (CD)14 and CD163 were 2.3-fold and 2.1-fold lower (95% confidence interval, 2.0-2.5 for CD14 and 1.7-2.4 for CD163), respectively, in patients with IL-10-treated HMDMs from PFIC1 compared with BRIC1. Conclusion: CD14 and CD163 expression levels in IL-10-treated HMDMs may facilitate diagnosis of PFIC1 or BRIC1 in patients with ATP8B1 deficiency.

Urinalysis by combination of the dipstick test and urine protein-creatinine ratio (UPCR) assessment can prevent unnecessary lenvatinib interruption in patients with thyroid cancer.

BACKGROUND: Proteinuria induced by lenvatinib is a class effect that occurs secondary to VEGFR suppression. Withholding of lenvatinib is required in cases with severe proteinuria. Urine protein-creatinine ratio (UPCR, g/gCre) has recently attracted attention as an alternative to 24-h urine collection for assessing proteinuria. The aim of this study was to examine the correlation between the results of proteinuria assessed by the dipstick test and UPCR, and to investigate the influence of proteinuria grading with UPCR on lenvatinib dose adjustment compared to that with only the dipstick test. METHOD: Three hundred and ten urine samples from 63 patients with advanced thyroid cancer under treatment with lenvatinib, which were tested by both the dipstick test and UPCR were analyzed. Lenvatinib was withheld when there was evidence of CTCAE grade 3 proteinuria, and restarted when it resolved. The frequency of proteinuria, correlation between the results of the dipstick test and UPCR test, and the effect of dose withholding in cases with results of 3 + in the dipstick test were calculated. RESULTS: Proteinuria was seen in 56 (88.9%) patients. Of the 154 dipstick 3 + samples, only 56 (36.4%) were judged as more than 3.5 g/gCre by UPCR (grade 3 proteinuria), although none of the 1 + and only 3.7% of 2 + samples were judged as grade 3 proteinuria. We were able to prevent unnecessary lenvatinib interruption due to proteinuria in 63.6% of dipstick 3 + samples by assessment of UPCR. CONCLUSIONS: Urinalysis by combination of the dipstick test and UPCR assessment might be a better strategy for preventing unnecessary interruption of lenvatinib.

Thyroid remnant ablation using 1,110 MBq of I-131 after total thyroidectomy: regulatory considerations on release of patients after unsealed radioiodine therapy.

OBJECTIVES: This study was undertaken to measure the radiation exposure level of caregivers following outpatient NaI (I-131) 1,110 MBq therapy for remnant thyroid ablation after total thyroidectomy in patients with differentiated thyroid cancer, and to evaluate the influence of activities of daily living on radiation exposure level, with the goal of proposing an optimum method of I-131 therapy. METHODS: The study included 37 patients with differentiated thyroid cancer, who had undergone total thyroidectomy and received outpatient based remnant thyroid ablation using NaI (I-131) 1,110 MBq, who were satisfying the following requirements: (1) patients who have no evidence of distant metastases, (2) whose living environments were appropriate for outpatient I-131 (1,110 MBq) therapy, and (3) patients who gave written informed consent. The dose rate at a distance of 1 m from the body surface of the patient at the moment of release was measured using survey meters of the GM type or ionization chamber type. The dose level for the caregiver was measured with a personal dosimeter in all cases. RESULTS: The dose rate at a distance of 1 m from the patient's body surface 1 h after I-131 administration was in the range of 29-115 µSv/h (mean 63.8 µSv/h). The 7-day cumulative effective dose of caregivers was 0.11 ± 0.08 mSv, on an average, in 34 dosimeters. In 31 of 34 dosimeters, cumulative effective dose of caregivers was below 0.2 mSv. Dose levels exceeding 0.2 mSv were recorded in 3 cases (0.21, 0.35 and 0.43 mSv in one case each). These results suggest that the exposure level of family members (caregiver and others) was minimal and is lower than the radiation levels affecting human environments. CONCLUSION: Outpatient-based remnant thyroid ablation with I-131 (1,110 MBq) performed after total thyroidectomy in patients with differentiated thyroid cancer is safe if applied in accordance with the appropriate supervision and guidance by experts with certain qualifications.