Online causal inference with application to near real-time post-market vaccine safety surveillance.

Streaming data routinely generated by social networks, mobile or web applications, e-commerce, and electronic health records present new opportunities to monitor the impact of an intervention on an outcome via causal inference methods. However, most existing causal inference methods have been focused on and applied to static data, that is, a fixed data set in which observations are pooled and stored before performing statistical analysis. There is thus a pressing need to turn static causal inference into online causal learning to support near real-time monitoring of treatment effects. In this paper, we present a framework for online estimation and inference of treatment effects that can incorporate new information as it becomes available without revisiting prior observations. We show that, under mild regularity conditions, the proposed online estimator is asymptotically equivalent to the offline oracle estimator obtained by pooling all data. Our proposal is motivated by the need for near real-time vaccine effectiveness and safety monitoring, and our proposed method is applied to a case study on COVID-19 vaccine safety surveillance.

Premarket Evidence and Postmarketing Requirements for Real-Time Oncology Review Indication Approvals.

This cross-sectional study evaluates the use of the US Food and Drug Administration's Real-Time Oncology Review (RTOR) program in confirming the effectiveness of cancer drugs.

Risk of acute pancreatitis among new users of empagliflozin compared to sulfonylureas in patients with type 2 diabetes: A post-authorization safety study.

PURPOSE: This study was undertaken to evaluate the potential risk of acute pancreatitis with empagliflozin in patients with type 2 diabetes (T2D) newly initiating empagliflozin. METHODS: Data from two large US claims databases were analyzed in an observational study of patients with T2D receiving metformin who were newly prescribed empagliflozin versus sulfonylurea (SU). Because dipeptidyl peptidase-4 inhibitors and glucagon-like peptide-1 receptor agonists have been associated with the risk of acute pancreatitis in some studies, patients on these agents were excluded. Using pooled analyses of data from the two databases (2014-2021), patients initiating empagliflozin were matched 1:1 within database to patients initiating SU using propensity scores (PS) that incorporated relevant demographic and clinical characteristics. Prespecified sensitivity analyses were performed for design parameters. RESULTS: The analyses identified 72 661 new users of empagliflozin and 422 018 new users of SUs, with both patient groups on concurrent metformin therapy. Baseline characteristics within treatment groups appeared to be similar across the 72 621 matched pairs. After mean follow-up of ~6 months, incidence rates of acute pancreatitis in the pooled matched cohort were 10.30 (95% confidence interval [CI] 9.29-11.39) events per 1000 patient-years (PY) for empagliflozin and 11.65 (95% CI 10.59-12.77) events per 1000 PY for SUs. On a background of metformin, patients newly initiating empagliflozin did not have an increased risk of acute pancreatitis compared with those initiating an SU (pooled PS matched hazard ratio 0.88 [0.76-1.02]) across 75621.42 PY of follow-up. CONCLUSIONS: The results of this voluntary post-approval safety study provide additional evidence that the use of empagliflozin for the treatment of T2D is not associated with an increased risk of acute pancreatitis.

Postmarket safety communications on drugs approved in Japan: A 25-year analysis.

Drug safety communications (DSCs) are essential tools for communicating important postmarket serious drug safety information to healthcare professionals and patients. Previous studies characterized DSCs issued by the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA); however, knowledge about the activities of the Pharmaceuticals and Medical Devices Agency (PMDA)/the Ministry of Health, Labor and Welfare (MHLW) is limited. This study characterized DSCs by the PMDA/MHLW in comparison with previously reported DSCs by the FDA and the EMA. We retrospectively analyzed 37 DSCs of 41 adverse drug reactions (ADRs) for 33 drugs in Japan from 1997 to 2022. Most DSCs were related to non-oncology drugs (30/37, 81.1%), and the median (interquartile range) time from approval to DSC issuance was 19 (10-51) months. Notably, the regulatory review reports and the latest labels before DSC issuance did not describe 16/28 (57.1%) and 12/37 (32.4%) of the ADRs related to DSCs, respectively. Most DSCs resulted in label revisions (36/37, 97.3%) and seven drugs were eventually withdrawn. Some DSC characteristics are similar among the PMDA/MHLW, the FDA, and the EMA; however, the number, contents, and range of new safety issues addressed by DSCs differ among the three jurisdictions. Our study emphasized the importance of continuous efforts to gather postmarket drug safety information because substantial ADRs that led to DSCs were recognized after approval and were associated with critical label revisions and withdrawals. Future studies are required to address global challenges for regulatory harmonization of safety-related regulatory actions.

Risk of hypertension in erenumab-treated patients with migraine: Analyses of clinical trial and postmarketing data.

OBJECTIVE: To assess the risk of hypertension in patients with migraine who received erenumab in clinical trials and in the postmarketing setting. BACKGROUND: Erenumab is a monoclonal antibody for migraine prevention that targets the calcitonin gene-related peptide (CGRP) receptor. Hypertension is a theoretical risk for inhibitors of the CGRP pathway. Although no evidence of an association between erenumab treatment and hypertension was observed during the clinical development program, adverse events (AEs) of hypertension have been identified in the postmarketing setting. METHODS: Safety data from four phase 2 and phase 3 clinical trials were used to perform a pooled analysis of hypertension AEs in patients with migraine receiving erenumab. Postmarketing AEs of hypertension were identified from the Amgen Global Safety database from May 17, 2018, through January 31, 2020. RESULTS: In the pooled analysis of clinical trials, hypertension AEs (placebo, 9/1043 [0.9%]; erenumab 70 mg, 7/893 [0.8%]; erenumab 140 mg, 1/507 [0.2%]) and percentage of patients initiating medication to treat hypertension (12/1043 [1.2%], 7/893 [0.8%], 1/507 [0.2%], respectively) were similar across treatment groups. A total of 362 AEs of hypertension were identified from the postmarketing setting, 26.2% (95/362) of which were serious, >245,000 patient-years of exposure. The exposure-adjusted incidence of hypertension was 0.144 per 100 patient-years. CONCLUSIONS: Clinical trials did not demonstrate an increased risk of hypertension with erenumab compared with placebo, and AE rates of hypertension reported with erenumab in the postmarketing setting were generally low. Additional data are needed to fully characterize the extent to which hypertension is a risk associated with erenumab.

Reporting of Death in US Food and Drug Administration Medical Device Adverse Event Reports in Categories Other Than Death.

Importance: In the US, most postmarket medical device safety data are obtained through adverse event reports that are submitted to the US Food and Drug Administration (FDA)'s Manufacturer and User Facility Device Experience (MAUDE) database. Adverse event reports are classified by the reporter as injury, malfunction, death, or other. If the device may have caused or contributed to a death, or if the cause of death is unknown, the FDA requires that the adverse event be reported as a death. Objective: To determine the percentage of medical device adverse event reports submitted to the MAUDE database that were not classified as death even though the patient died. Design, Setting, and Participants: In this study, a natural language processing algorithm was applied to the MAUDE database, followed by manual text review, to identify reports in the injury, malfunction, other or missing categories that included at least 1 term that suggested a patient death, such as patient died or patient expired, from December 31, 1991, to April 30, 2020, for any medical device. Exposures: Manual review of a random sample of 1000 adverse event reports not classified as death and of selected reports for 62 terms that are associated with deaths but were not classified as death. Main Outcomes and Measures: Percentage of adverse event reports in which the patient was said to have died in the narrative section of the report but the reporter classified the report in a category other than death. Results: The terms in the natural language processing algorithm identified 290 141 reports in which a serious injury or death was reported. Of these, 151 145 (52.1%) were classified by the reporter as death and 47.9% were classified as malfunction, injury, other, or missing. For the overall sample, the percentage of reports with deaths that were not classified as deaths was 23% (95% CI, 20%-25%), suggesting that approximately 31 552 reports in our sample had deaths that were classified in other categories. The overall percentage of missed deaths, defined as the percentage of deaths that were classified in other categories, was 17% (95% CI, 16%-19%). Conclusions and Relevance: Many of the findings of this study suggest that many medical device adverse event reports in the FDA's MAUDE database that involved a patient death are classified in categories other than death. As the FDA only routinely reviews all adverse events that are reported as patient deaths, improving the accuracy of adverse event reporting may enhance patient safety.

Effectiveness and Overall Safety of NutropinAq® for Growth Hormone Deficiency and Other Paediatric Growth Hormone Disorders: Completion of the International Cooperative Growth Study, NutropinAq® European Registry (iNCGS).

Objective: The International Cooperative Growth Study, NutropinAq® European Registry (iNCGS) (NCT00455728) monitored long-term safety and effectiveness of recombinant human growth hormone (rhGH; NutropinAq® [somatropin]) in paediatric growth disorders. Methods: Open-label, non-interventional, post-marketing surveillance study recruiting children with growth disorders. Endpoints included gain in height standard deviation score (SDS), adult height, and occurrence of adverse events (AEs). Results: 2792 patients were enrolled. 2082 patients (74.6%) had growth hormone deficiency (GHD), which was isolated idiopathic in 1825 patients (87.7%). Non-GHD diagnoses included Turner syndrome (TS) (n=199), chronic renal insufficiency (CRI) (n=10), other non-GHD (n=498), and missing data for three participants. Improvements from baseline height SDS occurred at all time points to Month 132, and in all subgroups by disease aetiology. At Month 12, mean (95% CI) change in height SDS by aetiology was: idiopathic GHD 0.63 (0.61;0.66), organic GHD 0.71 (0.62;0.80), TS 0.59 (0.53; 0.65), CRI 0.54 (-0.49;1.56), and other non-GHD 0.64 (0.59;0.69). Mean height ( ± SD) at the last visit among the 235 patients with adult or near-adult height recorded was 154.0 cm ( ± 8.0) for girls and 166.7 cm ( ± 8.0) for boys. The most frequent biological and clinical non-serious drug-related AEs were increased insulin-like growth factor concentrations (314 events) and injection site haematoma (99 events). Serious AEs related to rhGH according to investigators were reported (n=30); the most frequent were scoliosis (4 events), epiphysiolysis (3 events), and strabismus (2 events). Conclusions: There was an improvement in mean height SDS in all aetiology subgroups after rhGH treatment. No new safety concerns were identified.

Unexpected beneficial effects of drugs: an analysis of cases in the Dutch spontaneous reporting system.

INTRODUCTION: Drug use is inherently related to both beneficial effects on health as well as the occurrence of risks. The beneficial effects may be related to efficacy, the treatment range of a product, or even to user-friendliness of a product. However, in addition to the occurrence of adverse drug reactions, a drug can also have an unexpected beneficial effect on a patient's health, not related to the indication for which the drug was used. The aim of this article is to characterize the reports of unexpected beneficial effects of drugs in the Dutch spontaneous reporting system. METHODS: A descriptive analysis was used to gain insight in number of reports and drug classes responsible for unexpected beneficial effects of drugs. Grouping of positive side effects into classes was done by a conventional qualitative content analysis of the cases. RESULTS: Four hundred nine reports which described unexpected beneficial effects of drugs were included, which mentioned 451 associations between suspected drugs and unexpected beneficial effects. There were 147 drug classes on the 4th ATC level involved. Content analysis of the reports gave rise to 22 categories of unexpected beneficial effects of drugs, including one "other category". DISCUSSION AND CONCLUSION: The analysis showed a diverse spectrum of reported reactions and drugs with some categories of unexpected beneficial effects of drugs mentioned multiple times for certain drug classes on the 4th ATC level. Most of these findings are consistent with the existing literature and knowledge on the pharmacological mechanism of the drugs in question. Coding harmonization would make it possible to study these effects in international databases.

Estimating Baseline Incidence of Conditions Potentially Associated with Vaccine Adverse Events: a Call for Surveillance System Using the Korean National Health Insurance Claims Data.

BACKGROUND: Vaccines against coronavirus disease 2019 (COVID-19) are raising concerns about vaccine safety, particularly in the context of large-scale immunization. To address public concerns, we measured the baseline incidence rates of major conditions potentially related to vaccine-related adverse events (VAEs). We aimed to provide a basis for evaluating VAEs and verifying causality. METHODS: Conditions of interest were selected from the US Vaccine Adverse Event Reporting System Table of Reportable Events and a recent report from a European consortium on vaccine surveillance. We used the National Health Insurance Service database in Korea to identify the monthly numbers of cases with these conditions. Data from January 2006 to June 2020 were included. Prediction models were constructed from the observed incidences using an autoregressive integrated moving average. We predicted the incidences of the conditions and their respective 95% confidence intervals (CIs) for January through December 2021. In addition, subgroup analysis for the expected vaccination population was conducted. RESULTS: Mean values (95% CIs) of the predicted monthly incidence of vasovagal syncope, anaphylaxis, brachial neuritis, acute disseminated encephalomyelitis, Bell's palsy, Guillain-Barré syndrome, encephalopathy, optic neuritis, transverse myelitis, immune thrombocytopenic purpura, and systemic lupus erythematosus in 2021 were 23.89 (19.81-27.98), 4.72 (3.83-5.61), 57.62 (51.37-63.88), 0.03 (0.01-0.04), 8.58 (7.90-9.26), 0.26 (0.18-0.34), 2.13 (1.42-2.83), 1.65 (1.17-2.13), 0.19 (0.14-0.25), 0.75 (0.61-0.90), and 3.40 (2.79-4.01) cases per 100,000 respectively. The majority of the conditions showed an increasing trend with seasonal variations in their incidences. CONCLUSION: We measured the incidence of a total of 11 conditions that could potentially be associated with VAEs to predict the monthly incidence in 2021. In Korea, conditions that could potentially be related to VAEs occur on a regular basis, and an increasing trend is observed with seasonality.

Analysis of completeness for spontaneous reporting of disease-modifying therapies in multiple sclerosis.

Introduction: Considering the need for effective postmarketing surveillance of disease-modifying therapies (DMTs) in multiple sclerosis (MS), we analyzed the potential of the spontaneous reports for safety signal detection, verifying the completeness of the reports in the FDA Adverse Event Reporting System (FAERS).Methods: All reports with DMTs for MS considered the primary suspect cause of ADRs and registered between January 2004 and June 2019 were selected. The vigiGrade completeness score was applied and reports with a score greater than 0.80 were considered well documented. Descriptive statistical analysis and comparisons of well-documented reports by DMTs were performed.Results: A total of 297,926 reports were analyzed. The lowest completeness rates were observed for type of report (13.5%), dose (62.7%), and time from treatment start to the ADR (79.0%). Overall, 80.8% of reports were classified as well documented and those related to natalizumab had the highest proportion (92.4%, p < 0.001), while the lowest was observed for reports sent in 2017 (53.1%, p < 0.001) and for teriflunomide (48.5%, p < 0.001).Conclusions: The high proportion of well-documented reports for DMTs indicates that they can be a valuable source for safety signal detection. A more careful analysis should be performed for data from the groups identified with low completeness to avoid the disclosure of spurious results.

Maternal and neonatal data collection systems in low- and middle-income countries for maternal vaccines active safety surveillance systems: A scoping review.

BACKGROUND: Most post-licensure vaccine pharmacovigilance in low- and middle-income countries (LMICs) are passive reporting systems. These have limited utility for maternal immunization pharmacovigilance in LMIC settings and need to be supplemented with active surveillance. Our study's main objective was to identify existing perinatal data collection systems in LMICs that collect individual information on maternal and neonatal health outcomes and could be developed to inform active safety surveillance of novel vaccines for use during pregnancy. METHODS: A scoping review was performed following the Arksey and O'Malley six-stage approach. We included studies describing electronic or mixed paper-electronic data collection systems in LMICs, including research networks, electronic medical records, and custom software platforms for health information systems. Medline PubMed, EMBASE, Global Health, Cochrane Library, LILACS, Bibliography of Asian Studies (BAS), and CINAHL were searched through August 2019. We also searched grey literature including through Google and websites of existing relevant perinatal data collection systems, as well as contacted authors of key studies and experts in the field to validate the information and identify additional sources of relevant unpublished information. RESULTS: A total of 11,817 records were identified. The full texts of 264 records describing 96 data collection systems were assessed for eligibility. Eight perinatal data collection systems met our inclusion criteria: Global Network's Maternal Newborn Health Registry, International Network for the Demographic Evaluation of Populations and their Health; Perinatal Informatic System; Pregnancy Exposure Registry & Birth Defects Surveillance; SmartCare; Open Medical Record System; Open Smart Register Platform and District Health Information Software 2. These selected systems were qualitatively characterized according to seven different domains: governance; system design; system management; data management; data sources, outcomes and data quality. CONCLUSION: This review provides a list of active maternal and neonatal data collection systems in LMICs and their characteristics as well as their outreach, strengths, and limitations. Findings could potentially help further understand where to obtain population-based high-quality information on outcomes to inform the conduct of maternal immunization active vaccine safety surveillance activities and research in LMICs.

QT Prolongation Risk Assessment in Oncology: Lessons Learned From Small-Molecule New Drug Applications Approved During 2011-2019.

The International Conference on Harmonisation (ICH) E14 guidance provides recommendations to assess the potential of a drug to delay cardiac repolarization (QT prolongation), including general guidelines for cases in which a conventional thorough QT study (TQT) might not be feasible. These guidelines have been updated through the ICH question-and-answer process, with the last revision in 2015. We conducted a comprehensive analysis of QT prolongation evaluation of small-molecule new drug applications (NDAs) approved in oncology between 2011 and 2019 to extract learning experience. The following information was analysed: (1) methods to assess QT prolongation, (2) electrocardiogram data collection, (3) QT-related label language, and (4) postmarketing requirements. Overall, every NDA included a QT assessment. The concentration-QTc modeling approach (studies in which QT was not the primary objective) was the most common approach (59%), followed by the TQT and the dedicated QT studies (20% and 21%, respectively). The quality and quantity of the QT assessments were different across NDAs, which suggested relatively large flexibility in the designs and approaches to characterizing QT liability. The QT-related label language reflected the QT results, but also the safety events and the study design limitations because of the oncology settings. There was no delay in approval because of less robust QTc studies as long as the benefit-to-risk ratio of the drug was acceptable, and the implications were reflected in the label. This work offers a structured understanding of the QT evaluation criteria by the Food and Drug Administration and can assist in planning QT prolongation assessments in oncology settings.

A Broad Safety Assessment of the 9-Valent Human Papillomavirus Vaccine.

Parents indicate that safety is their top concern about human papillomavirus (HPV) vaccination. A data-mining method not requiring prespecification of health outcome(s) or postexposure period(s) of potentially increased risk can be used to identify possible associations between an exposure and any of thousands of medically attended health outcomes; this method was applied to data on the 9-valent HPV vaccine (HPV9) to detect potential safety problems. Data on 9- to 26-year-olds who had received HPV9 vaccine between November 4, 2016, and August 5, 2018, inclusive, were extracted from the MarketScan database and analyzed for statistically significant clustering of incident diagnoses within the hierarchy of diagnoses coded using the International Classification of Diseases and temporally within the 1 year after vaccination, using the self-controlled tree-temporal scan statistic and TreeScan software. Only 56 days of postvaccination enrollment was required; subsequent follow-up was censored at disenrollment. Multiple testing was adjusted for. The analysis included 493,089 doses of HPV9. Almost all signals resulted from temporal confounding, not unexpected with a 1-year follow-up period. The only plausible signals were for nonspecific adverse events (e.g., injection-site reactions, headache) on days 1-2 after vaccination, with attributable risks as low as 1 per 100,000 vaccinees. Considering the broad scope of the evaluation and the high statistical power, the findings of no specific serious adverse events should provide reassurance about this vaccine's safety.

Suitability of databases in the Asia-Pacific for collaborative monitoring of vaccine safety.

INTRODUCTION: Information regarding availability of electronic healthcare databases in the Asia-Pacific region is critical for planning vaccine safety assessments particularly, as COVID-19 vaccines are introduced. This study aimed to identify data sources in the region, potentially suitable for vaccine safety surveillance. This manuscript is endorsed by the International Society for Pharmacoepidemiology (ISPE). METHODS: Nineteen countries targeted for database reporting were identified using published country lists and review articles. Surveillance capacity was assessed using two surveys: a 9-item introductory survey and a 51-item full survey. Survey questions related to database characteristics, covariate and health outcome variables, vaccine exposure characteristics, access and governance, and dataset linkage capability. Other questions collated research/regulatory applications of the data and local publications detailing database use for research. RESULTS: Eleven databases containing vaccine-specific information were identified across 8 countries. Databases were largely national in coverage (8/11, 73%), encompassed all ages (9/11, 82%) with population size from 1.4 to 52 million persons. Vaccine exposure information varied particularly for standardized vaccine codes (5/11, 46%), brand (7/11, 64%) and manufacturer (5/11, 46%). Outcome data were integrated with vaccine data in 6 (55%) databases and available via linkage in 5 (46%) databases. Data approval processes varied, impacting on timeliness of data access. CONCLUSIONS: Variation in vaccine data availability, complexities in data access including, governance and data release approval procedures, together with requirement for data linkage for outcome information, all contribute to the challenges in building a distributed network for vaccine safety assessment in the Asia-Pacific and globally. Common data models (CDMs) may help expedite vaccine safety research across the region.

Pharmacodynamic analysis of hypertension caused by lenvatinib using real-world postmarketing surveillance data.

Lenvatinib is a tyrosine kinase inhibitor of the vascular endothelial growth factor receptor used against nonoperative thyroid cancer; however, hypertension is a major dose-limiting side effect. In this study, hypertension caused by lenvatinib was described through a novel population pharmacodynamic model using postmarketing surveillance data obtained in Japan. The model consists of two maximum effect model components based on the (1) concentration of lenvatinib in plasma and (2) cumulative area under the curve of lenvatinib. In addition, antihypertensive drug of either an angiotensin-converting enzyme inhibitor/angiotensin receptor blocker or calcium channel blocker accounted for by lowering effect on diastolic blood pressure. Based on virtual simulations, the combination of antihypertensive drug and dose adjustment of lenvatinib showed a reduction in the probability of grade greater than or equal to 3 hypertension. The present model provides useful guidance in managing hypertension during treatment with lenvatinib in the real-world setting.

Post-market utilization patterns of Alzheimer's disease treatments in South Korea: comparison with countries with universal health coverage.

PURPOSE: This study aimed to compare the utilization of Alzheimer's disease (AD) treatments, donepezil, galantamine, rivastigmine, and memantine, in Korea with Australia and other countries with universal health coverage. METHODS: Reimbursement criteria and the patent status of four AD treatments in Korea and Australia were reviewed. The monthly spending and utilization of the treatments were extracted from the national electronic database in Korea and Australia. The defined daily dose per 1000 elderly population per day (DDD/1000e/day) were calculated from July 2008 to June 2019. Annual cost trends of Norway and England were compared with Korea and Australia. RESULTS: With the highest share of the use of donepezil in both countries, the cost and utilization of AD treatments in Korea increased more rapidly and remained higher than Australia. The cost of AD treatments in Korea increased by 15.5% every year during the study period, while the spending of the same drugs in Australia decreased by 10.5% annually. The utilization in DDD/1000e/day of AD treatments in Korea increased by 18.3% annually compared with 1.4% in Australia. When compared with Norway and England, countries with similar universal health coverage (UHC) system and elderly polupation, the cost of AD treatments in Korea was still higher with the opposite trend from other countries. CONCLUSIONS: Despite the similar UHC systems, there were considerable differences in the post-market utilization of AD treatments in Korea from Australia and other countries. This results can be attributed to differences in re-assessment system, pricing and reimbursement policies, and prescribing culture. This study provides a baseline to explore more comprehensive cross-country studies on rational use of medicines.

Multivariable prediction model to estimate the probability of restenosis at proximal edge after 2nd-generation drug-eluting-stent implantation: development and internal validation using a quantitative coronary angiography from the post-marketing surveillance studies of everolimus-eluting stent in Japan.

Edge restenosis has still been reported after second-generation drug-eluting stent (DES) implantation. It was more likely attributable to post-procedural angiographic results than to the patient's background. The aim of this study was to develop and internally validate a prediction model for restenosis in proximal edge after 2nd-generation DES stent implantation using angiographic data. Data were obtained from several post-marketing surveillance (PMS) studies of the cobalt-chromium everolimus-eluting stent (CoCr-EES) and platinum-chromium everolimus-eluting stent (PtCr-EES), second-generation DES, in Japan. Angiographic analysis was conducted at baseline and after 8 or 12 months. We focused on the proximal edge of angiographic analysis. The main outcome was restenosis defined as ≥ 50% diameter stenosis at follow-up. The predictive performance of the prediction model based on multivariable logistic regression was assessed in terms of discrimination and calibration, which were internally validated by the bootstrap method. We also performed decision curve analysis to assess threshold of predicted probability of restenosis at which additional intervention was considered. Among 2053 lesions in 1860 patients, restenosis rates in proximal edge was 2.8%. The final model was constructed with % post-procedural diameter stenosis (DS) and post-procedural reference diameter (RD) as strong predictors for edge restenosis. Discrimination and calibration were satisfactory with optimism-corrected C-statistics 0.75. Predicted probability between 0.03 and 0.24 was preferable threshold for restenosis treatments. Our prediction model can be used to obtain valid prediction for restenosis in proximal edge, assisting to know complete stent coverage of lesion.