Adjuvant chemotherapy for isolated resectable colorectal lung metastasis: A retrospective study using inverse probability treatment weighting propensity analysis

BACKGROUND The benefit of adjuvant chemotherapy(ACT)for patients with no evidence of disease after pulmonary metastasis resection(PM)from colorectal cancer(CRC)remains controversial.AIM To assess the efficacy of ACT in patients after PM resection for CRC.METHODS This study included 96 patients who underwent pulmonary metastasectomy for CRC at a single institution between April 2008 and July 2023.The primary end-point was overall survival(OS);secondary endpoints included cancer-specific survival(CSS)and disease-free survival(DFS).An inverse probability of treat-ment-weighting(IPTW)analysis was conducted to address indication bias.Sur-vival outcomes compared using Kaplan-Meier curves,log-rank test,Cox regre-ssion and confirmed by propensity score-matching(PSM).RESULTS With a median follow-up of 27.5 months(range,18.3-50.4 months),the 5-year OS,CSS and DFS were 72.0%,74.4%and 51.3%,respectively.ACT had no significant effect on OS after PM resection from CRC[original cohort:P=0.08;IPTW:P=0.15].No differences were observed for CSS(P=0.12)and DFS(P=0.68)between the ACT and non-ACT groups.Multivariate analysis showed no association of ACT with better survival,while sublobar resection(HR=0.45;95%CI:0.20-1.00,P=0.049)and longer disease-free interval(HR=0.45;95%CI:0.20-0.98,P=0.044)were associated with improved survival.CONCLUSION ACT does not improve survival after PM resection for CRC.Further well-designed randomized controlled trials are needed to determine the optimal ACT regimen and duration.

Length of hospital stay and mortality associated with burns from assault:a retrospective study with inverse probability weighting analysis

Background:Burns resulting from assaults account for considerable morbidity and mortality among patients with burn injuries around the world.However,it is still unclear whether unfavorable clinical outcomes are associated primarily with the severity of the injuries.To elucidate the direct relationship between burns resulting from assaults and mortality and/or length of hospital stays,we performed this study with the hypothesis that burns from assault would be independently associated with fewer hospital-free days than would burns from other causes,regardless of the severity of burn injuries.Methods:We conducted a retrospective cohort study,using a city-wide burn registry(1996–2017)accounting for 14 burn centers in Tokyo,Japan.Patients who arrived within 24 hours after injury were included,and those with self-inflicted burn injuries were excluded.Patients were divided into two groups according to mechanism of burns(assault vs.accident),and the number of hospitalfree days until day 30 after injury(a composite of in-hospital death and hospital length of stay)was compared between the groups.To estimate the probability that an injury would be classified as an assault,we calculated propensity scores,using multivariate logistic regression analyses adjusted for known outcome predictors.We also performed an inverse probability weighting(IPW)analysis to compare adjusted numbers of hospital-free days.Results:Of 7419 patients in the registry with burn injuries during the study period,5119 patients were included in this study.Of these,113(2.2%)were injured as a result of assault;they had significantly fewer hospital-free days than did those with burns caused by accident(18[27]vs.24[20]days;coefficient=−3.4[−5.5 to−1.3]days;p=0.001).IPW analyses similarly revealed the independent association between assault burn injury and fewer hospital-free days(adjusted coefficient=−0.6[−1.0 to−0.1]days;p=0.009).Conclusions:Burn from assault was independently associated with fewer hospital-free days,regardless of the severity of burn injuries.The pathophysiological mechanism underlying the relationship should be further studied in a prospective observational study.

Effectiveness of antibiotic prophylaxis for acute esophageal variceal bleeding in patients with band ligation: A large observational study

BACKGROUND Esophageal variceal bleeding is a severe complication associated with liver cirrhosis and typically necessitates endoscopic hemostasis.The current standard treatment is endoscopic variceal ligation(EVL),and Western guidelines recom-mend antibiotic prophylaxis following hemostasis.However,given the impro-vements in prognosis for variceal bleeding due to advancements in the management of bleeding and treatments of liver cirrhosis and the global concerns regarding the emergence of multidrug-resistant bacteria,there is a need to reassess the use of routine antibiotic prophylaxis after hemostasis.AIM To evaluate the effectiveness of antibiotic prophylaxis in patients treated for EVL.METHODS We conducted a 13-year observational study using the Tokushukai medical database across 46 hospitals.Patients were divided into the prophylaxis group(received antibiotics on admission or the next day)and the non-prophylaxis group(did not receive antibiotics within one day of admission).The primary outcome was composed of 6-wk mortality,4-wk rebleeding,and 4-wk spontaneous bacterial peritonitis(SBP).The secondary outcomes were each individual result and in-hospital mortality.A logistic regression with inverse probability of treatment weighting was used.A subgroup analysis was conducted based on the Child-Pugh classification to determine its influence on the primary outcome measures,while sensitivity analyses for antibiotic type and duration were also performed.RESULTS Among 980 patients,790 were included(prophylaxis:232,non-prophylaxis:558).Most patients were males under the age of 65 years with a median Child-Pugh score of 8.The composite primary outcomes occurred in 11.2%of patients in the prophylaxis group and 9.5%in the non-prophylaxis group.No significant differences in outcomes were observed between the groups(adjusted odds ratio,1.11;95%confidence interval,0.61-1.99;P=0.74).Individual outcomes such as 6-wk mortality,4-wk rebleeding,4-wk onset of SBP,and in-hospital mortality were not significantly different between the groups.The primary outcome did not differ between the Child-Pugh subgroups.Similar results were observed in the sensitivity analyses.CONCLUSION No significant benefit to antibiotic prophylaxis for esophageal variceal bleeding treated with EVL was detected in this study.Global reassessment of routine antibiotic prophylaxis is imperative.

Predicting portal venous anomalies by left-sided gallbladder or rightsided ligamentum teres hepatis: A large scale, propensity scorematched study

BACKGROUND Right-sided ligamentum teres(RSLT)is often associated with portal venous anomalies(PVA)and is regarded as a concerning feature for hepatobiliary intervention.Most studies consider RSLT to be one of the causes of left-sided gallbladder(LGB),leading to the hypothesis that LGB must always be present with RSLT.However,some cases have shown that right-sided gallbladder(RGB)can also be present in livers with RSLT.AIM To highlight the rare variation that RSLT may not come with LGB and to determine whether ligamentum teres(LT)or gallbladder location is reliable to predict PVA.METHODS This study retrospectively assessed 8552 contrast-enhanced abdominal computed tomography examinations from 2018 to 2021[4483 men,4069 women;mean age,59.5±16.2(SD)years].We defined the surrogate outcome as major PVAs.The cases were divided into 4 subgroups according to gallbladder and LT locations.On one hand,we analyzed PVA prevalence by LT locations using gallbladder location as a controlled variable(n=36).On the other hand,we controlled LT location and computed PVA prevalence by gallbladder locations(n=34).Finally,we investigated LT location as an independent factor of PVA by using propensity score matching(PSM)and inverse probability of treatment weighting(IPTW).RESULTS We found 9 cases of RSLT present with RGB.Among the LGB cases,RSLT is associated with significantly higher PVA prevalence than typical LT[80.0%vs 18.2%,P=0.001;OR=18,95%confidence interval(CI):2.92-110.96].When RSLT is present,we found no statistically significant difference in PVA prevalence for RGB and LGB cases(88.9%vs 80.0%,P>0.99).Both PSM and IPTW yielded balanced cohorts in demographics and gallbladder locations.The RSLT group had a significantly higher PVA prevalence after adjusted by PSM(77.3%vs 4.5%,P<0.001;OR=16.27,95%CI:2.25-117.53)and IPTW(82.5%vs 4.7%,P<0.001).CONCLUSION RSLT doesn't consistently coexist with LGB.RSLT can predict PVA independently while the gallbladder location does not serve as a sufficient predictor.

An evaluation of treatments and survival rates for pancreatic adenocarcinoma through survival analysis with inverse probability of treatment weighting:a population-based study

Objective:This study conducted inverse probability of treatment weighting(IPTW)survival analysis to examine survival in pancreatic adenocarcinoma patients.Methods:In this population-based study,data from the Surveillance,Epidemiology,and End Results program of the United States were analyzed to identify patients diagnosed with adenocarcinoma of the pancreas 2004 to 2014.Differences in survival rates were examined among patients who underwent pancreatectomy alone,radiotherapy alone,and those who had pancreatectomy plus adjuvant radiotherapy.Kaplan-Meier estimates and Cox proportional hazards models with the IPTW were performed to determine the effect of different treatments on overall and cancer-specific survival.This study was approved by the Ethics Review Board of Weifang Medical University.Results:A total of 8191 patients were included,with 3409 taking pancreatectomy only,2865 taking radiotherapy only,and 1917 taking pancreatectomy plus adjuvant radiotherapy.Patients who received surgery plus adjuvant radiotherapy had statistically a higher survival rate than those who received the other 2 treatments.Survival analysis with the IPTW for the 3 different groups showed that the difference in median overall survival time among these patient groups was significant.Conclusion:Using IPTW survival analysis,the present study shows that surgery with adjuvant radiotherapy is significantly associated with improved overall and cancer-specific survival among patients with pancreatic adenocarcinoma.

CBPS-Based Inference in Nonlinear Regression Models with Missing Data

In this article, to improve the doubly robust estimator, the nonlinear regression models with missing responses are studied. Based on the covariate balancing propensity score (CBPS), estimators for the regression coefficients and the population mean are obtained. It is proved that the proposed estimators are asymptotically normal. In simulation studies, the proposed estimators show improved performance relative to usual augmented inverse probability weighted estimators.

Testing conditional independence with data missing at random

It is known that conditional independence is a quite basic assumption in many fields of statistics. How to test its validity is of great importance and has been extensively studied by the literature. Nevertheless, all of the existing methods focus on the case that data are fully observed, but none of them seems having taken into account of the scenario when missing data are present. Motivated by this, this paper develops two testing statistics to handle such a situation relying on the idea of inverse probability weighted and augmented inverse probability weighted techniques. The asymptotic distributions of the proposed statistics are also derived under the null hypothesis. The simulation studies indicate that both testing statistics perform well in terms of size and power.

Is ypTNM staging a comparable predictor as pTNM staging for survival in non-metastatic rectal cancer after preoperative chemoradiation therapy?

Abstract:Background:The pTNM staging system is widely recognized as the most effective prognostic indicator for cancer.The latest update of this staging system introduced a new pathological staging system(ypTNM)for patients receiving neoadjuvant chemoradiotherapy(NACRT).However,whether the prognostic value of the ypTNM staging system for rectal cancer is similar to that of the pTNM staging system remains unclear.This study was conducted to compare the ypTNM and pTNM staging systems in terms of their prognostic value for patients with nonmetastatic rectal cancer undergoing proctectomy.Material and Methods:This study was conducted at a large teaching hospital.Between January 2014 and December 2022,542 patients with rectal cancer were analyzed(median follow-up period,60 months;range,6–105 months).Of them,258 and 284 were included in the pTNM and ypTNM groups,respectively.Inverse probability of treatment weighting(IPTW)was performed to account for the effects of confounders.Cox proportional-hazards regression was performed for the between-group comparison of overall survival(OS).Results:The crude model revealed that OS was similar between the two groups(p=0.607).After performing IPTW,we found that patients with the same ypTNM-and pTNM-classified stages had similar overall survival(hazard ratio=1.15;95%CI=0.76–1.73;p=0.5074).Conclusions:For patients with rectal cancer who have received preoperative NACRT,the prognostic value of ypTNM staging appears to be similar to that of pTNM staging,mostly because of the downstaging effect of neoadjuvant chemoradiotherapy。

Association of food intake with a risk of metabolic dysfunction-associated fatty liver disease:a cross-sectional study

Background:Metabolic dysfunction-associated fatty liver disease(MAFLD)is a common liver disease,the risk of which can be increased by poor diet.The objective of this study was to evaluate the associations between food items and MAFLD,and to propose reasonable dietary recommendations for the prevention of MAFLD.Methods:Physical examination data were collected from April 2015 through August 2017 at Nanping First Hospital(n=3,563).Dietary intakes were assessed using a semi-quantitative food frequency questionnaire.The association between food intake and the risk of MAFLD was assessed by using the inverse probability weighted propensity score.Results:Beverages(soft drinks and sugar-sweetened beverages)and instant noodles were positively associated with MAFLD risk,adjusting for smoking,drinking,tea intake,and weekly hours of physical activity[adjusted odds ratio(ORadjusted):1.568;P=0.044;ORadjusted:4.363;P=0.001].Milk,tubers,and vegetables were negatively associated with MAFLD risk(ORadjusted:0.912;P=0.002;ORadjusted:0.633;P=0.007;ORadjusted:0.962;P=0.028).In subgroup analysis,the results showed that women[odds ratio(OR):0.341,95%confidence interval(CI):0.172–0.676]had a significantly lower risk of MAFLD through consuming more tubers than men(OR:0.732,95%CI:0.564–0.951).Conclusions:These findings suggest that reducing consumption of beverages(soft drinks and sugar-sweetened beverages)and instant noodles,and consuming more milk,vegetables,and tubers may reduce the risk of MAFLD.

Empirical Likelihood for Response Differences in Two Linear Regression Models with Missing Data

Oonsider two linear models Xi = U＇β ＋ ei, Yj = V1/2y ＋ ηj with response variables missing at random. In this paper, we assume that X, Y are missing at random （MAR） and use the inverse probability weighted imputation to produce ＇complete＇ data sets for X and Y. Based on these data sets, we construct an empirical likelihood （EL） statistic for the difference of X and Y （denoted as A）, and show that the EL statistic has the limiting distribution of X~, which is used to construct a confidence interval for A. Results of a simulation study on the finite sample performance of EL-based confidence intervals on A are reported.

Quasi-likelihood estimation of average treatment effects based on model information

In this paper, the estimation of average treatment effects is considered when we have the model information of the conditional mean and conditional variance for the responses given the covariates. The quasi-likelihood method adapted to treatment effects data is developed to estimate the parameters in the conditional mean and conditional variance models. Based on the model information, we define three estimators by imputation, regression and inverse probability weighted methods. All the estimators are shown asymptotically normal. Our simulation results show that by using the model information, the substantial efficiency gains are obtained which are comparable with the existing estimators.

Regression Estimation for Longitudinal Data with Nonignorable Intermittent Nonresponse and Dropout

We mainly focus on regression estimation in a longitudinal study with nonignorable intermittent nonresponse and dropout.To handle the identifiability issue,we take a time-independent covariate as nonresponse instrument which is independent of nonresponse propensity conditioned on other covariates and responses to ensure the identifiability of nonresponse propensity.The nonresponse propensity is assumed to be a parametric model,and the corresponding parameters are estimated by using the generalized method of moments approach.Then the marginal response means are estimated by inverse probability weighting method.Furthermore,to improve the robustness of estimators,we derive an augmented inverse probability weighting estimator which is shown to be consistent and asymptotically normally distributed.Simulation studies and a real-data analysis show that the proposed approach yields highly efficient estimators.

Robust dynamic risk prediction with longitudinal studies

Providing accurate and dynamic age-specific risk prediction is a crucial step in precision medicine.In this manuscript,we introduce an approach for estimating theτ-year age-specific absolute riskdirectly via a flexible varying coefficient model.The approach facilitates the utilisation of predictors varying over an individual’s lifetime.By using a nonparametric inverse probability weightedkernel estimating equation,the age-specific effects of risk factors are estimated without requiring the specification of the functional form.The approach allows borrowing information acrossindividuals of similar ages,and therefore provides a practical solution for situations where the longitudinal information is only measured sparsely.We evaluate the performance of the proposedestimation and inference procedures with numerical studies,and make comparisons with existingmethods in the literature.We illustrate the performance of our proposed approach by developinga dynamic prediction model using data from the Framingham Study.

An equivalence result for moment equations when data are missing at random

We consider general statistical models defined by moment equations when data are missing atrandom. Using the inverse probability weighting, such a model is shown to be equivalent with amodel for the observed variables only, augmented by a moment condition defined by the missing mechanism. Our framework covers a large class of parametric and semiparametric modelswhere we allow for missing responses, missing covariates and any combination of them. Theequivalence result is stated under minimal technical conditions and sheds new light on variousaspects of interest in the missing data literature, as for instance the efficiency bounds and theconstruction of the efficient estimators, the restricted estimators and the imputation.

Deep advantage learning for optimal dynamic treatment regime

Recently deep learning has successfully achieved state-of-the-art performance on many difficulttasks. Deep neural networks allow for model flexibility and process features without the needof domain knowledge. Advantage learning (A-learning) is a popular method in dynamic treatment regime (DTR). It models the advantage function, which is of direct relevance to optimaltreatment decision. No assumptions on baseline function are made. However, there is a paucityof literature on deep A-learning. In this paper, we present a deep A-learning approach to estimate optimal DTR. We use an inverse probability weighting method to estimate the differencebetween potential outcomes. Parameter sharing of convolutional neural networks (CNN) greatlyreduces the amount of parameters in neural networks, which allows for high scalability. Convexified convolutional neural networks (CCNN) relax the constraints of CNN for optimisation purpose.Different architectures of CNN and CCNN are implemented for contrast function estimation.Both simulation results and application to the STAR*D (Sequenced Treatment Alternatives toRelieve Depression) trial indicate that the proposed methods outperform penalised least squareestimator.

Estimating Differences in Restricted Mean Lifetime Using Additive Hazards Models under Dependent Censoring

In epidemiological and clinical studies,the restricted mean lifetime is often of direct interest quantity.The differences of this quantity can be used as a basis of comparing several treatment groups with respect to their survival times.When the factor of interest is not randomized and lifetimes are subject to both dependent and independent censoring,the imbalances in confounding factors need to be accounted.We use the mixture of additive hazards model and inverse probability of censoring weighting method to estimate the differences of restricted mean lifetime.The average causal effect is then obtained by averaging the differences in fitted values based on the additive hazards models.The asymptotic properties of the proposed method are also derived and simulation studies are conducted to demonstrate their finite-sample performance.An application to the primary biliary cirrhosis(PBC)data is illustrated.

An Additive-Multiplicative Cox-Aalen Subdistribution Hazard Model for Competing Risks Data

This paper proposes a flexible additive-multiplicative Cox-Aalen hazard model which allows time-varying covariate effects for the subdistribution in a competing risks study.Weigh ted estimating equation approaches under an covariates-dependent adjusted weight by fitting the Cox proportional hazard model for the censoring distribution are established for inference on the model parametric and nonparametric components.In addition,large number properties are presented and the finite sample behavior of the proposed estimators is evaluated through simulation studies,estimators from the proposed method perform satisfactorily on reduction of the bias.The authors apply our model to a competing risks data set from a tamoxifen trail for breast cancer study.