Development and evaluation of machine learning models for predicting large-for-gestational-age newborns in women exposed to radiation prior to pregnancy.

INTRODUCTION: The correlation between radiation exposure before pregnancy and abnormal birth weight has been previously proven. However, for large-for-gestational-age (LGA) babies in women exposed to radiation before becoming pregnant, there is no prediction model yet. MATERIAL AND METHODS: The data were collected from the National Free Preconception Health Examination Project in China. A sum of 455 neonates (42 SGA births and 423 non-LGA births) were included. A training set (n = 319) and a test set (n = 136) were created from the dataset at random. To develop prediction models for LGA neonates, conventional logistic regression (LR) method and six machine learning methods were used in this study. Recursive feature elimination approach was performed by choosing 10 features which made a big contribution to the prediction models. And the Shapley Additive Explanation model was applied to interpret the most important characteristics that affected forecast outputs. RESULTS: The random forest (RF) model had the highest average area under the receiver-operating-characteristic curve (AUC) for predicting LGA in the test set (0.843, 95% confidence interval [CI]: 0.714-0.974). Except for the logistic regression model (AUC: 0.603, 95%CI: 0.440-0.767), other models' AUCs displayed well. Thereinto, the RF algorithm's final prediction model using 10 characteristics achieved an average AUC of 0.821 (95% CI: 0.693-0.949). CONCLUSION: The prediction model based on machine learning might be a promising tool for the prenatal prediction of LGA births in women with radiation exposure before pregnancy.

Clinical characteristics and prognosis of patients with newly diagnosed primary central nervous system lymphoma: a multicentre retrospective analysis.

Bruton's tyrosine kinase inhibitors (BTKi) exhibit superior efficacy in relapsed/refractory primary central nervous system lymphoma (PCNSL), but few studies have evaluated patients with newly diagnosed PCNSL, and even fewer studies have evaluated differences in efficacy between treatment with BTKi and traditional chemotherapy. This study retrospectively analyzed the clinical characteristics of 86 patients with PCNSL and identified predictors of poor prognosis for overall survival (OS). After excluding patients who only received palliative care, 82 patients were evaluated for efficacy and survival. According to the induction regimen, patients were divided into the traditional chemotherapy, BTKi combination therapy, and radiotherapy groups; the objective response rates (ORR) of the three groups were 71.4%, 96.2%, and 71.4% (P = 0.037), respectively. Both median progression-free survival and median duration of remission showed statistically significant differences (P = 0.019 and P = 0.030, respectively). The median OS of the BTKi-containing therapy group was also longer than that of the traditional chemotherapy group (not reached versus 47.8 (32.5-63.1) months, P = 0.038).Seventy-one patients who achieved an ORR were further analyzed, and achieved an ORR after four cycles of treatment and maintenance therapy had prolonged OS (P = 0.003 and P = 0.043, respectively). In conclusion, survival, and prognosis of patients with newly diagnosed PCNSL are influenced by the treatment regimen, with the BTKi-containing regimen showing great potential.

Comparison of the Efficacy Among Nilotinib, Dasatinib, Flumatinib and Imatinib in Newly Diagnosed Chronic-Phase Chronic Myeloid Leukemia Patients: A Real-World Multi-Center Retrospective Study.

BACKGROUND: There are limited data comprehensively comparing therapy responses and outcomes among nilotinib, dasatinib, flumatinib and imatinib for newly diagnosed chronic-phase chronic myeloid leukemia in a real-world setting. PATIENTS AND METHODS: Data from patients with chronic-phase CML receiving initial a second-generation tyrosine-kinase inhibitor (2G-TKI, nilotinib, dasatinib or flumatinib) or imatinib therapy from 77 Chinese centers were retrospectively interrogated. Propensity-score matching (PSM) analyses were performed to to compare therapy responses and outcomes among these 4 TKIs. RESULTS: 2,496 patients receiving initial nilotinib (n = 512), dasatinib (n = 134), flumatinib (n = 411) or imatinib (n = 1,439) therapy were retrospectively interrogated in this study. PSM analyses indicated that patients receiving initial nilotinib, dasatinib or flumatinib therapy had comparable cytogenetic and molecular responses (p = .28-.91) and survival outcomes including failure-free survival (FFS, p = .28-.43), progression-free survival (PFS, p = .19-.93) and overall survival (OS) (p values = .76-.78) but had significantly higher cumulative incidences of cytogenetic and molecular responses (all p values < .001) and higher probabilities of FFS (p < .001-.01) than those receiving imatinib therapy, despite comparable PFS (p = .18-.89) and OS (p = .23-.30). CONCLUSION: Nilotinib, dasatinib and flumatinib had comparable efficacy, and significantly higher therapy responses and higher FFS rates than imatinib in newly diagnosed CML patients. However, there were no significant differences in PFS and OS among these 4 TKIs. These real-world data may provide additional evidence for routine clinical assessments to identify more appropriate therapies.