Primary biliary cholangitis (PBC) is a common condition that usually shows a progressive course towards cirrhosis without adequate treatment. Growth differentiation factor 15 (GDF15) plays multiple roles in various pathological conditions. The overall role of circulating GDF15 in cirrhotic PBC requires further investigation. Twenty patients with cirrhotic PBC, 26 with non‐cirrhotic PBC, and 10 healthy subjects were enrolled between 2014 and 2018, and the serum levels of GDF15 were measured via enzyme immunoassay. The correlations between serum GDF15, weight, biochemical parameters, and the prognosis were analysed. Serum levels of GDF15 were significantly higher in cirrhotic PBC patients than in non‐cirrhotic PBC patients or healthy controls (p = 0.009 and p < 0.001, respectively). The circulating GDF15 levels strongly correlated with weight changes (r = −0.541, p = 0.0138), albumin (r = −0.775, p < 0.0001), direct bilirubin (r = −0.786, p < 0.0001), total bile acids (r = 0.585, p = 0.007), and C‐reactive protein (r = 0.718, p = 0.0005). Moreover, circulating GDF15 levels strongly correlated with the Mayo risk score (r = 0.685, p = 0.0009) and Model for End‐stage Liver Disease score (r = 0.687, p = 0.0008). Determined by the area under the receiver operating characteristic curves, the overall diagnostic accuracies of GDF15 were as follows: cirrhosis = 0.725 (>3646.55 pg/mL, sensitivity: 70.0%, specificity: 69.2%), decompensated cirrhosis = 0.956 (>4073.30 pg/mL, sensitivity: 84.62%, specificity: 100%), and cirrhotic biochemical non‐responders = 0.835 (>3479.20 pg/mL, sensitivity: 71.43%, specificity: 92.31%). GDF15 may be a useful and integrated biochemical marker to evaluate not only the disease severity and prognosis but also the nutrition and response to treatment of cirrhotic PBC patients, and its overall performance is satisfactory. Therapy targeting GDF15 is likely to benefit cirrhotic PBC patients and is worth further research.

Background. Primary biliary cholangitis‐autoimmune hepatitis overlap syndrome (PBC‐AIH OS), which exhibits features between autoimmune hepatitis and cholestasis, is a common condition and usually shows a progressive course toward cirrhosis and liver failure without adequate treatment. Synthesis of bile acids (BAs) plays an important role in liver injury in cholestasis, and the process is regulated by fibroblast growth factor 19 (FGF19). The overall role of circulating FGF19 in BA synthesis and PBC‐AIH OS requires further investigation. Methods. We analyzed BA synthesis and correlated clinical parameters with serum BAs and FGF19 in 35 patients with PBC‐AIH OS. Serum concentrations of 7alpha‐hydroxycholest‐4‐en‐3‐one (C4) were used to quantify the synthesis of BA directly. Results. Serum FGF19 levels were higher, while C4 levels were substantially lower in PBC‐AIH OS patients than those in healthy controls. Circulating FGF19 levels strongly correlated with C4 (r = −0.695, p < 0.0001), direct bilirubin (r = 0.598, p = 0.0001), and total bile acids (r = 0.595, p = 0.002). Moreover, circulating FGF19 levels strongly correlated with the model for end‐stage liver disease score (r = 0.574, p = 0.0005) and Mayo risk score (r = 0.578, p = 0.001). Conclusions. Serum FGF19 is significantly increased in patients with PBC‐AIH OS, while BA synthesis is suppressed. Circulating FGF19 primarily controls the regulation of BA synthesis in response to cholestasis and under cholestatic conditions. Therefore, modulation of circulating FGF19 could provide a promising targeted therapy for patients with PBC‐AIH OS.

Background: Breast cancer (BC) metastasis is the common cause of high mortality. Conventional prognostic criteria cannot accurately predict the BC metastasis risk. The machine learning technologies can overcome the disadvantage of conventional models.Aim: We developed a model to predict BC metastasis using the random survival forest (RSF) method.Methods: Based on demographic data and routine clinical data, we used RSF‐recursive feature elimination to identify the predictive variables and developed a model to predict metastasis using RSF method. The area under the receiver operating characteristic curve (AUROC) and Kaplan–Meier survival (KM) analyses were plotted to validate the predictive effect when C‐index was plotted to assess the discrimination and Brier scores was plotted to assess the calibration of the predictive model.Results: We developed a metastasis prediction model comprising three variables (pathological stage, aspartate aminotransferase, and neutrophil count) selected by RSF‐recursive feature elimination. The model was reliable and stable when assessed by the AUROC (0.932 in training set and 0.905 in validation set) and KM survival analyses (p < .0001). The C‐indexes (0.959) and Brier score (0.097) also validated the good predictive ability of this model.Conclusions: This model relies on routine data and examination indicators in real‐time clinical practice and exhibits an accurate prediction performance without increasing the cost for patients. Using this model, clinicians can facilitate risk communication and provide precise and efficient individualized therapy to patients with breast cancer.

Background: Breast cancer survivorship requires attention to treatment-related changes in body composition, nutritional status, and quality of life. Bioelectrical impedance phase angle is influenced by tissue electrical properties and fluid distribution and may provide complementary whole-body and segmental information. Objective: This study described perioperative body-composition and phase-angle measures in women undergoing breast cancer surgery after neoadjuvant chemotherapy or treated through a surgery-first pathway. It also estimated T0-to-T2 changes and documented the availability of selected longitudinal variables. Methods: BC-NUTRITION (Body Composition and Nutrition Trajectories in Breast Cancer) is a prospective 2-cohort observational study with 5 planned assessments from admission to 6 months after discharge. The present exploratory pilot analysis included 30 women in cohort A, assessed after neoadjuvant chemotherapy, and 65 women in cohort B, who were treated through a surgery-first pathway. T1 measurements were displayed descriptively; inferential analyses used T0 values and participant-level T0-to-T2 changes. T0 between-cohort comparisons used Welch 2-sample t tests, within-cohort changes used paired t tests, and between-cohort differences in change used Welch tests. Benjamini-Hochberg q values were calculated across the 13 between-cohort change comparisons. Results: All 95 participants had complete data for the 13 selected longitudinal variables at T0, T1, and T2. At T0, cohort A had lower albumin, absolute lymphocyte count, prognostic nutritional index, and whole-body 50-kHz phase angle. The whole-body phase-angle means were 4.64° (SD 0.66) and 5.23° (SD 0.62), with a difference of −0.59° (95% CI −0.88° to −0.31°; P <.001). Cohort A had a higher whole-body extracellular water to total body water ratio (ECW/TBW) than cohort B (0.3890 vs 0.3815; P <.001). Ipsilateral upper-limb 50-kHz phase angle changed by −0.50° in cohort A and −0.51° in cohort B between T0 and T2. The between-cohort difference was 0.01° (95% CI −0.17° to 0.19°; P =.93). The corresponding difference in whole-body phase-angle change was 0.14° (95% CI −0.02° to 0.30°; P =.09). No q value for the 13 between-cohort change comparisons was below 0.05. Conclusions: The treatment pathways differed in selected laboratory and whole-body bioelectrical measures at T0. The mean upper-limb phase angle ipsilateral to the recorded tumor side decreased in both cohorts between T0 and T2. All five prespecified perioperative feasibility thresholds were met in the available T0-to-T2 data. Assessment acceptability and T3/T4 follow-up were not evaluated. Definitive interpretation requires standardized acquisition, a prechemotherapy baseline, relevant clinical covariates, and outpatient follow-up.