Machine learning models for outcome prediction of Chinese uveal melanoma patients: A 15-year follow-up study.
Ontology highlight
ABSTRACT:
SUBMITTER: Chen YN
PROVIDER: S-EPMC8923127 | biostudies-literature | 2022 Mar
REPOSITORIES: biostudies-literature
Publications
Machine learning models for outcome prediction of Chinese uveal melanoma patients: A 15-year follow-up study.
Chen Yu-Ning YN Wang Yi-Ning YN Chen Meng-Xi MX Zhang Kai K Chen Rong-Tian RT Fang Rui R Wang Heng H Zhang Hai-Han HH Huang Yi-Ning YN Feng Yu Y Luo Jing-Ting JT Lan Yin-Jun YJ Liu Yue-Ming YM Li Yang Y Wei Wen-Bin WB
Cancer communications (London, England) 20220109 3
Similar Datasets
Project description:PurposeMRI is increasingly used in the diagnosis and therapy planning of uveal melanoma (UM). In this prospective cohort study, we assessed the radiological characteristics, in terms of anatomical and functional imaging, of UM after ruthenium-106 plaque brachytherapy or proton beam therapy (PBT) and compared them to conventional ultrasound.MethodsTwenty-six UM patients were evaluated before and 3, 6 and 12 months after brachytherapy (n = 13) or PBT (n = 13). Tumour prominences were compared between ultrasound and MRI. On diffusion-weighted imaging, the apparent diffusion value (ADC), and on perfusion-weighted imaging (PWI), the time-intensity curves (TIC), relative peak intensity and outflow percentages were determined. Values were compared between treatments and with baseline.ResultsPre-treatment prominences were comparable between MRI and ultrasound (mean absolute difference 0.51 mm, p = 0.46), but larger differences were observed post-treatment (e.g. 3 months: 0.9 mm (p = 0.02)). Pre-treatment PWI metrics were comparable between treatment groups. After treatment, brachytherapy patients showed favourable changes on PWI (e.g. 67% outflow reduction at 3 months, p < 0.01). After PBT, significant perfusion changes were observed at a later timepoint (e.g. 38% outflow reduction at 6 months, p = 0.01). No consistent ADC changes were observed after either treatment, e.g. a 0.11 × 10-3mm2/s increase 12 months after treatment (p = 0.15).ConclusionMR-based follow-up is valuable for PBT-treated patients as favourable perfusion changes, including a reduction in outflow, can be detected before a reduction in size is apparent on ultrasound. For brachytherapy, a follow-up MRI is of less value as already 3 months post-treatment a significant size reduction can be measured on ultrasound.
| S-EPMC10227806 | biostudies-literature
Project description:Survival is generally poor for Chinese patients with advanced melanoma because of high rates of acral and mucosal melanoma and limited therapeutic options. The first analysis of the phase 1b KEYNOTE-151 study showed second-line pembrolizumab was well tolerated and had clinically meaningful antitumor activity in Chinese patients with advanced melanoma. Three-year follow-up is presented. Eligible patients were of Chinese descent and had unresectable stage III/IV melanoma that progressed after first-line therapy. Patients received pembrolizumab 2 mg/kg every 3 weeks for ≤35 cycles. Primary end points were safety and objective response rate (ORR). Secondary end points included duration of response (DOR), progression-free survival (PFS), and overall survival (OS). Response was assessed per RECIST v1.1 by blinded independent central review. Subgroup analyses were conducted by melanoma subtype and BRAF and PD-L1 status (acral melanoma only). 103 patients were enrolled; median follow-up duration (time from first dose to data cutoff [July 13, 2020]) was 44.6 months (IQR, 39.1-46.2). Any-grade treatment-related adverse events (TRAEs) occurred in 85.4% of patients, and grade 3/4 TRAEs in 12.6%. No grade 5 TRAEs occurred. Three patients discontinued pembrolizumab because of TRAEs (immune-mediated hepatitis, pneumonia, and arthritis). Immune-mediated AEs and infusion reactions occurred in 34.0% (grade 3/4, 2.9%). ORR was 17.6% (95% CI, 10.8-26.4; 1 complete response/17 partial responses), and median DOR was 13.8 months (range, 2.7-37.4+). Median PFS was 2.8 months (95% CI, 2.7-3.5) and 36-month PFS rate was 5.0%. Median OS was 13.2 months (95% CI, 10.4-16.5) and 36-month OS rate was 22.3%. Median OS for patients with known melanoma subtype was 14.8 months for acral, 13.5 months for nonacral cutaneous, and 7.4 months for mucosal melanoma. Among the acral subgroup, median OS was 22.8 months for PD-L1-positive disease, 8.4 months for PD-L1-negative disease, 18.5 months for BRAF wild-type disease, and 5.8 months for BRAF-mutant disease. Over 3 years' follow-up, second-line pembrolizumab continued to show manageable safety, clinically meaningful antitumor activity, and durable responses in Chinese patients with advanced melanoma. Subgroup analysis suggested particular benefit in PD-L1-positive and BRAF wild-type acral melanoma, although small subgroup sizes preclude definitive conclusions.Clinical trial registrationhttps://clinicaltrials.gov, identifier NCT02821000.
| S-EPMC9593700 | biostudies-literature
Project description:IntroductionUveal melanoma (UM) is the most common intraocular malignancy in adults. Plaque brachytherapy remains the dominant eyeball-conserving therapy for UM. Tumor regression in UM after plaque brachytherapy has been reported as a valuable prognostic factor. The present study aimed to develop an accurate machine-learning model to predict the 4-year risk of metastasis and death in UM based on ocular ultrasound data.Material and methodsA total of 454 patients with UM were enrolled in this retrospective, single-center study. All patients were followed up for at least 4 years after plaque brachytherapy and underwent ophthalmologic evaluations before the therapy. B-scan ultrasonography was used to measure the basal diameters and thickness of tumors preoperatively and postoperatively. Random Forest (RF) algorithm was used to construct two prediction models: whether a patient will survive for more than 4 years and whether the tumor will develop metastasis within 4 years after treatment.ResultsOur predictive model achieved an area under the receiver operating characteristic curve (AUC) of 0.708 for predicting death using only a one-time follow-up record. Including the data from two additional follow-ups increased the AUC of the model to 0.883. We attained AUCs of 0.730 and 0.846 with data from one and three-time follow-up, respectively, for predicting metastasis. The model found that the amount of postoperative follow-up data significantly improved death and metastasis prediction accuracy. Furthermore, we divided tumor treatment response into four patterns. The D(decrease)/S(stable) patterns are associated with a significantly better prognosis than the I(increase)/O(other) patterns.ConclusionsThe present study developed an RF model to predict the risk of metastasis and death from UM within 4 years based on ultrasound follow-up records following plaque brachytherapy. We intend to further validate our model in prospective datasets, enabling us to implement timely and efficient treatments.
| S-EPMC8816318 | biostudies-literature
Project description:Uveal melanoma (UM) patients face a significant risk of distant metastasis, closely tied to a poor prognosis. Despite this, there is a dearth of research utilizing big data to predict UM distant metastasis. This study leveraged machine learning methods on the Surveillance, Epidemiology, and End Results (SEER) database to forecast the risk probability of distant metastasis. Therefore, the information on UM patients from the SEER database (2000-2020) was split into a 7:3 ratio training set and an internal test set based on distant metastasis presence. Univariate and multivariate logistic regression analyses assessed distant metastasis risk factors. Six machine learning methods constructed a predictive model post-feature variable selection. The model evaluation identified the multilayer perceptron (MLP) as optimal. Shapley additive explanations (SHAP) interpreted the chosen model. A web-based calculator personalized risk probabilities for UM patients. The results show that nine feature variables contributed to the machine learning model. The MLP model demonstrated superior predictive accuracy (Precision = 0.788; ROC AUC = 0.876; PR AUC = 0.788). Grade recode, age, primary site, time from diagnosis to treatment initiation, and total number of malignant tumors were identified as distant metastasis risk factors. Diagnostic method, laterality, rural-urban continuum code, and radiation recode emerged as protective factors. The developed web calculator utilizes the MLP model for personalized risk assessments. In conclusion, the MLP machine learning model emerges as the optimal tool for predicting distant metastasis in UM patients. This model facilitates personalized risk assessments, empowering early and tailored treatment strategies.
| S-EPMC11462970 | biostudies-literature
Project description:BackgroundAccording to the Global Burden of Disease Study 2017, smoking is one of the leading four risk factors contributing to deaths in China. We aimed to evaluate the associations of smoking with all-cause mortality in a Chinese rural population.MethodsMale participants over age 45 (n = 5367) from a large familial aggregation study in rural China, were included in the current analyses. A total of 528 former smokers and 3849 current smokers accounted for 10 and 71.7% of the cohort, respectively. Generalized Estimating Equations were used to evaluate the association between baseline smoking status and mortality, adjusting for pertinent covariates.ResultsThere were 579 recorded deaths during the 15-year follow-up. Current smokers (odds ratio [OR],1.60; 95% CI,1.23-2.08) had higher all-cause mortality risks than nonsmokers. Relative to nonsmokers, current smokers of more than 40 pack-years ([OR],1.85; 95% CI,1.33-2.56) had a higher all-cause mortality risk. Compared to nonsmokers, current smokers who started smoking before age 20 ([OR],1.91; 95% CI,1.43-2.54) had a higher all-cause mortality risk, and former smokers in the lower pack-year group who quit after age 41 (median) ([OR],3.19; 95% CI,1.83-5.56) also had a higher risk of death after adjustment. Furthermore, former smokers who were also former drinkers had the highest significant risk of mortality than never smokers or drinkers. (P for interaction = 0.034).ConclusionsThis study provides evidence that current smokers and former smokers have a higher mortality risk than nonsmokers and would benefit from cessation at a younger age.
| S-EPMC8034075 | biostudies-literature
Project description:IntroductionThe worldwide prevalence of diabetes has been increasing for decades; diabetes can lead to serious health problems and even death, but the effects of maintaining low fasting blood glucose (FBG) remain controversial. The purpose of this study was to investigate the relationship between FBG levels and all-cause mortality in a long-term follow-up cohort and to find a relatively safe range of FBG levels.MethodsThis study included 17,902 adults from a community-based cohort study in rural China who were prospectively followed from 2003 to 2018. Generalized estimating equations were used to evaluate the association between FBG and all-cause mortality, adjusting for pertinent covariates and auto-correlations among siblings.ResultsA total of 1053 (5.9%) deaths occurred during 15 years of follow-up. There was a significant U-shaped association between all-cause mortality and FBG. Compared with the reference group (FBG of 5.6 - < 6.1 mmol/l), the risk of death among individuals with FBG levels < 5.6 mmol/l significantly increased by 38% (OR 1.34; 95% CI 1.13-1.59), while the risk of death among individuals with FBG ≥ 6.1 mmol/l or participants with a self-reported history of diabetes significantly increased by 51% (OR 1.49; 95% CI 1.20-1.85). Additionally, the U-shaped association remained steady in any stratification of risk factors.ConclusionOur study showed a significant U-shaped relationship between FBG levels and risk of all-cause mortality in this rural Chinese population. When FBG was within the range of 5.6 - < 6.1 mmol/l, the risk of all-cause mortality was the lowest.
| S-EPMC7547918 | biostudies-literature
Project description:ObjectiveThis study aims at comparing treatment outcome and tooth survival of root canal-filled teeth following manual vs rotary instrumentation techniques over a 5-year period.MethodsThis was a single-centre study conducted as a follow-up to a short-term parallel-arm randomised controlled noninferiority trial in which root canal treatment was performed on teeth using either rotary or manual instrumentation. Patients were monitored at post 6-month, 1-year, 4-year, and 5-year review periods by blinded evaluators. Treatment outcome was categorised as favourable, uncertain, and unfavourable (employing European Society of Endodontology categorisation based on strict clinical and radiographic criteria), and 5-year tooth survival was determined by assessing whether tooth was in situ in the oral cavity or extracted. The Kaplan-Meier method and log rank test evaluated tooth survival. P value <.05 was considered statistically significant.ResultsNinety of 120 treated teeth were assessed in 37 men and 40 women with mean age of 30.6 ± 10.99 years. Treatment outcome was significantly more favourable in the rotary group compared to the manual group at post 6-month (P = .021) and 1-year (P = .043) review periods. The differences in favourable outcome (P = .498) and tooth survival (P = .296) between the 2 groups were, however, not significant at the 5-year review period.ConclusionsThe rotary instrumentation technique was shown to be more effective in resolving clinical symptoms and promoting periapical healing after the post 6-month and 1-year review compared to the manual instrumentation technique; however, both groups had similar favourable outcomes and survival rates after an extended 5-year review period.
| S-EPMC9676548 | biostudies-literature
Project description:Although a broad array of neurocognitive dysfunctions are associated with ADHD, it is unknown whether these dysfunctions play a role in the course of ADHD symptoms. The present longitudinal study investigated whether neurocognitive functions assessed at study-entry (mean age = 11.5 years, SD = 2.7) predicted ADHD symptom severity and overall functioning 6 years later (mean age = 17.4 years, 82.6 % = male) in a carefully phenotyped large sample of 226 Caucasian participants from 182 families diagnosed with ADHD-combined type. Outcome measures were dimensional measures of ADHD symptom severity and the Kiddie-Global Assessment Scale (K-GAS) for overall functioning. Predictors were derived from component scores for 8 domains of neurocognitive functioning: working memory, motor inhibition, cognitive inhibition, reaction time variability, timing, information processing speed, motor control, intelligence. Effects of age, gender, and pharmacological treatment were considered. Results showed that better working memory predicted lower ADHD symptom severity (R 2 = 3.0 %), and less reaction time variability predicted better overall functioning (higher K-GAS-score, R 2 = 5.6 %). Predictors were still significant with baseline behavior included in the models. The role of neurocognitive functioning in the long term outcome of ADHD behavior is discussed.
| S-EPMC5241361 | biostudies-literature
Project description:BackgroundWe previously reported that a median of 5.6 years of intensive as compared with standard glucose lowering in 1791 military veterans with type 2 diabetes resulted in a risk of major cardiovascular events that was significantly lower (by 17%) after a total of 10 years of combined intervention and observational follow-up. We now report the full 15-year follow-up.MethodsWe observationally followed enrolled participants (complete cohort) after the conclusion of the original clinical trial by using central databases to identify cardiovascular events, hospitalizations, and deaths. Participants were asked whether they would be willing to provide additional data by means of surveys and chart reviews (survey cohort). The prespecified primary outcome was a composite of major cardiovascular events, including nonfatal myocardial infarction, nonfatal stroke, new or worsening congestive heart failure, amputation for ischemic gangrene, and death from cardiovascular causes. Death from any cause was a prespecified secondary outcome.ResultsThere were 1655 participants in the complete cohort and 1391 in the survey cohort. During the trial (which originally enrolled 1791 participants), the separation of the glycated hemoglobin curves between the intensive-therapy group (892 participants) and the standard-therapy group (899 participants) averaged 1.5 percentage points, and this difference declined to 0.2 to 0.3 percentage points by 3 years after the trial ended. Over a period of 15 years of follow-up (active treatment plus post-trial observation), the risks of major cardiovascular events or death were not lower in the intensive-therapy group than in the standard-therapy group (hazard ratio for primary outcome, 0.91; 95% confidence interval [CI], 0.78 to 1.06; P = 0.23; hazard ratio for death, 1.02; 95% CI, 0.88 to 1.18). The risk of major cardiovascular disease outcomes was reduced, however, during an extended interval of separation of the glycated hemoglobin curves (hazard ratio, 0.83; 95% CI, 0.70 to 0.99), but this benefit did not continue after equalization of the glycated hemoglobin levels (hazard ratio, 1.26; 95% CI, 0.90 to 1.75).ConclusionsParticipants with type 2 diabetes who had been randomly assigned to intensive glucose control for 5.6 years had a lower risk of cardiovascular events than those who received standard therapy only during the prolonged period in which the glycated hemoglobin curves were separated. There was no evidence of a legacy effect or a mortality benefit with intensive glucose control. (Funded by the VA Cooperative Studies Program; VADT ClinicalTrials.gov number, NCT00032487.).
| S-EPMC6706253 | biostudies-literature
Project description:ObjectiveTo develop machine learning (ML) models, using pre and intraoperative surgical parameters, for predicting trabeculectomy outcomes in the eyes of patients with juvenile-onset primary open-angle glaucoma (JOAG) undergoing primary surgery.SubjectsThe study included 207 JOAG patients from a single center who met the following criteria: diagnosed between 10 and 40 years of age, with an IOP of >22 mmHg in the eyes on two or more occasions, open angle on gonioscopy in both eyes, with glaucomatous optic neuropathy, and requiring a trabeculectomy for IOP control. Only the patients with a minimum 5-year follow-up after surgery were included in the study.MethodsA successful surgical outcome was defined as IOP ≤18 mmHg (criterion A) or 50% reduction in IOP from baseline (criterion B) 5 years after trabeculectomy. Feature selection techniques were used to select the most important contributory parameters, and tenfold cross-validation was used to evaluate model performance. The ML models were evaluated, compared, and prioritized based on their accuracy, sensitivity, specificity, Matthew correlation coefficient (MCC) index, and mean area under the receiver operating characteristic curve (AUROC). The prioritized models were further optimized by tuning the hyperparameters, and feature contributions were evaluated. In addition, an unbiased relationship analysis among the parameters was performed for clinical utility.ResultsAge at diagnosis, preoperative baseline IOP, duration of preoperative medical treatment, Tenon's thickness, scleral fistulation technique, and intraoperative mitomycin C (MMC) use, were identified as the main contributing parameters for developing efficient models. The three models developed for a consensus-based outcome to predict trabeculectomy success showed an accuracy of >86%, sensitivity of >90%, and specificity of >74%, using tenfold cross-validation. The use of intraoperative MMC and a punch for scleral fistulation compared to the traditional excision with scissors were significantly associated with long-term success of trabeculectomy.ConclusionOptimizing surgical parameters by using these ML models might reduce surgical failures associated with trabeculectomy and provide more realistic expectations regarding surgical outcomes in young patients.
| S-EPMC11329837 | biostudies-literature