With the continuous development of genomics and precision medicine,targeted therapy and immunotherapy targeting biomarkers have ush-ered in a new era of anti-tumor therapy.However,due to the heterogeneity of tumor cells and the variability of tumor microenvironment,there are still significant differences in response to the same drug even in patient populations with the same bio-marker enrichment.By combining omics data with drug sensitivity algorithms,the response of anti-tu-mor drugs can be predicted and transformed into personalized diagnosis and treatment strategies re-quired for precision medicine,which is expected to improve the effectiveness of anti-tumor drugs in clinical treatment.Currently,machine learning is one of the commonly used modeling algorithms for predicting the response of anti-tumor drugs.How-ever,due to differences in input data and algorithm construction methods,there is currently a lack of comprehensive literature review in this field.There-fore,this article provides a review of machine learning algorithms for predicting anti-tumor drug responses,summarizing publicly available cell ge-nome characterization datasets,machine learning algorithms,and evaluation indicators in drug re-sponse prediction,as well as the current situation and challenges faced in clinical applications,in or-der to provide methodological references for the main research problems and potential solutions of machine learning algorithms in the field of drug re-sponse prediction.

With the continuous development of genomics and precision medicine,targeted therapy and immunotherapy targeting biomarkers have ush-ered in a new era of anti-tumor therapy.However,due to the heterogeneity of tumor cells and the variability of tumor microenvironment,there are still significant differences in response to the same drug even in patient populations with the same bio-marker enrichment.By combining omics data with drug sensitivity algorithms,the response of anti-tu-mor drugs can be predicted and transformed into personalized diagnosis and treatment strategies re-quired for precision medicine,which is expected to improve the effectiveness of anti-tumor drugs in clinical treatment.Currently,machine learning is one of the commonly used modeling algorithms for predicting the response of anti-tumor drugs.How-ever,due to differences in input data and algorithm construction methods,there is currently a lack of comprehensive literature review in this field.There-fore,this article provides a review of machine learning algorithms for predicting anti-tumor drug responses,summarizing publicly available cell ge-nome characterization datasets,machine learning algorithms,and evaluation indicators in drug re-sponse prediction,as well as the current situation and challenges faced in clinical applications,in or-der to provide methodological references for the main research problems and potential solutions of machine learning algorithms in the field of drug re-sponse prediction.

The intention to treat (ITT) principle is widely applied in the randomized controlled trials (RCTs), which is based on the intention to treat a subject (i.e. the planned treatment regimen) rather than the actual treatment given. With the development and progress of clinical trials, ITT principle needs to be reinterpreted. In the standard definition of ITT principle, the intentional grouping of subjects does not specifically refer to randomization, but rather to planned treatment regimen. In non-randomized clinical trials, that subjects agree to be included in a certain treatment regimen is also a type of intention. ITT principles can be applied to evaluate the effectiveness of treatment regimens under patient/doctor determined intentionality in single-arm clinical trials and non-randomized RWS (real world study, RWS). In the practical application, the grouping strategy based on ITT principle should be considered comprehensively according to the research purpose.

The widespread availability real-world study (RWS) offers valuable insights into disease treatment, disease management, and socio-economic status in routine practice, as well as cautionary tales and methodological challenges such as the discovery of sample heterogeneity and bias of data and its correction. This paper summarizes the common bias and its control in the process of design, implementation and analysis for RWS in order to promote the standardization and rationality of the implementation of RWS.

Objective: To evaluate the efficacy and safety of a domestic human plasma derived coagulation Factor Ⅸ concentrate (pd-FⅨ) in patients with hemophilia B. Methods: The study was a multicenter, open-label and single-arm study. The efficacy of pd-F Ⅸ was evaluated by objective performance criteria. The doses of pd-FⅨ were calculated according to the bleeding symptom and disease severity. The infusion efficiency of pd-FⅨ and improvement of bleeding symptoms were measured at 30 minutes and (24±4) h after the first infusion, respectively. Adverse events were recorded. Viral infection and FⅨ inhibitor were detected 90 d after the first infusion. Results: All 36 subjects with hemophilia B were enrolled in the study. The median age of these patients was 31 years old and the median injection doses were 4 (1-17) times. The hemostatic effect of 27/36 (75.00%) and 9/36 (25.00%) acute bleeding events were rated as "excellent" and "better" , respectively. The recovery rate was 111.92% (65.55%-194.28%) at 30 minutes after infusion of FⅨ. There was no adverse event related to FⅨ. No reactivation of HBV, HCV or HIV and FⅨ inhibitor was detected at 90-104 d after the first FⅨ infusion. Conclusion: This domestically made human plasma derived FⅨ concentrate is safe and effective in the treatment of acute bleeding in patients with hemophilia B. Clinical trial registration China food and Durg Administration, 2016L08027.

Missing data is a common but unavoidable issue in clinical trials. It not only lowers the trial power, but brings the bias to the trial results. Therefore, on one hand, the missing data handling methods are employed in data analysis. On the other hand, it is vital to prevent the missing data in the trials. Prevention of missing data should take the first place. From the perspective of data, firstly, some measures should be taken at the stages of protocol design, data collection and data check to enhance the patients' compliance and reduce the unnecessary missing data. Secondly, the causes of confirmed missing data in the trials should be notified and recorded in detail, which are very important to determine the mechanism of missing data and choose the suitable missing data handling methods, e.g., last observation carried forward (LOCF); multiple imputation (MI); mixed-effect model repeated measure (MMRM), etc.

Testing of hypothesis was affected by statistical analysis set division which was an important data management work before data base lock-in. Objective division of statistical analysis set under blinding was the guarantee of scientific trial conclusion. All the subjects having accepted at least once trial treatment after randomization should be concluded in safety set. Full analysis set should be close to the intention-to-treat as far as possible. Per protocol set division was the most difficult to control in blinded examination because of more subjectivity than the other two. The objectivity of statistical analysis set division must be guaranteed by the accurate raw data, the comprehensive data check and the scientific discussion, all of which were the strict requirement of data management. Proper division of statistical analysis set objectively and scientifically is an important approach to improve the data management quality.

Objective To evaluate the effect of ilaprazole enteric tablets on intragastric pH in duodenal ulcer patients. Methods A randomized, double blind, positive controlled clinical trial was carried out. A total of forty-two patients with duodenal ulcer were randomized into low dose ilaprazole group (5 mg/d), medium dose ilaprazole group (10 mg/d), high dose ilaprazole group(20 mg/d) and omeprazole group(20 mg/d). An ambulatory 24 hour intragastric pH study was performed at the fifth treatment day. Fraction time pH above 3, 4 or 5, median values of 24 hour diurnal pH and 12 hour nocturnal pH, the percentage of patients with total time pH above 3, 4 or 5 at least for 18 hours were evaluated. Results There were no significant differences of fraction time pH above 3 or 4, median values of 24 hour diurnal pH and 12 hour nocturnal pH and the percentage of patients with total time pH above 3, 4 or 5 at least for 18 hours among all the groups with different doses of ilaprazole and the omeprazole group. The fraction time pH above 5 in medium and high dose ilaprazole groups were (87.96 ± 12. 29)% and (89.86±15. 18)% respectively, which was higher than that in low dose ilaprazole group [(67. 17± 30. 16)%] and omeprazole group[(76. 14 ± 16. 75)%], P <0. 05. Conclusion Ilaprazole has a strong effect on intragastric acid control with a dose dependent trend.

Objective To evaluate the therapeutic efficacy and safety of transjugular intrahepatic portosystemic shunt (TIPS) for the treatment of portal hypertension of patients with hepatocellular carcinoma.Methods Ninety-five portal hypertension patients with hepatic carcinoma were enrolled.TIPS was performed in 63 patients and the other 32 patients received support medical care.The data referred to survival time of the 95 patients after treatment was collected by follow-up visit.The informations about success rate of TIPS,hepatic encephalopathy,rebleeding and causes of death were assessed.The Kaplan-Meier method was used to compare the survival time between two groups.The association of survival time with Child-Pugh classification and model for end-stage liver disease (MELD) score was analyzed.Results The success rate of TIPS was 97.8% with reduction of mean portal vein pressure of 13.6 cmH2O(1 cmH2O=0.098 kPa).The incidence of hepatic encephalopathy was 20.6% and rebleeding was 26.3% six months after TIPS treatment.Fifty-six patients treated with TIPS died at the end of follow-up.Twelve of which were died of variceal bleeding complicated with portal hypertension.The median survival time of TIPS group (3.67 months) was significantly longer than that of control group (1 month). Moreover, the median survival time in patients with low MELD score (≤13) was significantly longer than that in those with high MELD seore (＞13, x2=4.71,P=0.03). Whereas the median survival time was decreasing from Child-Pugh A to C(x2=15.6,P=0.00). Conclusions TIPS is one of effective and safe therapeutic methods to control portal hypertension. However, liver function is an important factor for selcetion of TIPS.

Objective To determine the appropriate methods to sample size adjustment for adaptive design through simulations by comparing sample size adjustment of blinding internal pilot design with that of unwinding one in clinical trials. Methods Compare type I error and the power of blinding and unblinding internal pilot design through Monte Carlo simulation. Results Either type I error rates or the powers are not substantially distinct in two lands of settings. Conclusion Blinding sample size adjustment is more preferable.