Comparison of Staged Angioplasty and Routine Single-stage Stenting (CAS) in the Treatment of Carotid Artery Stenosis

Data Quality

Before the initiation of the study sites, Cerebrovascular Disease Center of Beijing Tiantan Hospital will verify whether the investigators are qualified, inspect the study site's facilities, and inform the investigators of their responsibilities, as well as how to correctly and completely fill related documents. Through investigator training sessions, the investigators and their teams will understand the trial protocol, CRF and trial process, etc. Then start-up visit at each site will be initiated.

Trial monitors will pay start-up visit to the site, and pay the first inspection visit as soon as possible after the screening/randomization of the first patient. Afterwards, the monitors will pay inspection visits regularly until the trial visits end. They need to directly refer to source data to verify the accurate of CRF filling and conduct inspection/audits.

Data entered into the CRF shall be subject to examination of internal consistency (within known classification and scope), external inspection (data are consistent with the patient's other known data) and trend consistency (data changes do not exceed the allowable range. For example, excess weight loss prompts changes in dietary status). The contents of trial database include data collected by each site, abnormal data and delayed data in progress tracking.

File Data

Data entered into the CRF must be able to be verified by source data, or directly entered into the CRF. In this case, the data in the CRF can be regarded as source data, and used for subsequent data confirmation. Required data of all patients' continued treatment shall also be routinely recorded in their medical records. The site's investigators shall clearly make patients' medical records that the patient is participating in a clinical trial, and also require any other medical staff for treatment of the patient to contact the trial contact persons before treatment. To this end, the investigators shall provide their contact information. If a patient preserves his/her medical records on his/her own, the study site shall preserve a copy of all trial-related data. Unless related data or information is directly recorded in the CRF, all research data must be recorded in patients' medical records (or research documents), and entered into the CRF as early as possible. Source data to be verified and primary documents used for such purpose must be preserved. Patients' research documents and all source data shall be preserved until a notice for destruction is received from the sponsor.

Statistical Method and Sample Size

3.1 Statistical Analysis

This study is powered on the primary endpoint of hyperperfusion syndrome (HPS) rate at 30 days post CAS procedure. Superiority analysis will be performed using a two-sample t-test at a two-sided 5% alpha level. All secondary endpoints will be descriptively compared between the two treatment arms.

All analyses will be performed on the Per-Protocol (PP) population. The PP population will consist of patients who have received the treatment per randomization without major protocol deviations.

X2 test is adopted for statistical analysis of two groups of results, supplemented by compliance with the protocol and actual treatment analysis. The following confounding factors will be considered in the analysis to determine the effect of different preset confounding factors on prognosis:

• Age;
• Sex;
• Blood pressure;
• Degree, location, diameter and length of lesion stenosis;
• Cerebrovascular collateral compensatory conditions;
• Cerebral hemodynamics of the culprit side;
• Cerebrovascular reserve or cerebrovascular autoregulation status of the culprit side;
• Diameter and pressure of balloon for stage-1 balloon angioplasty;
• Carotid angioplasty experience of the operating site or doctor.

Appropriate methods will be employed to summarize and analyze patients' baseline characteristics, safety and other indicators during the trial (including treatment interruption/termination, lifestyle assessment, concomitant medication, etc.). Multi-variant method is employed to analyze the effect of demographics and other factors on primary results. All details of statistical analysis are shown in the separate 'Statistical Analysis Plan'(SAP).

3.2 Sample Size

The sample size calculation for the primary endpoint of HPS rate 30 days post CAS procedure is based on the following assumptions:

Two-tailed superiority test a = 0.05 Power = 90% Randomization ratio is 1 (staged angioplasty arm): 1 (routine single-stage CAS arm) The rate of HPS rate was expected to be 0.16 per subject during the procedure in the regular CAS arm for high hyperperfusion risk patients based on previous studies.

The rate of HPS rate was presume to be 0.02 per subject during the procedure in the staged CAS arm for high hyperperfusion risk patients depended on the study by Yoshimura, Kyung Mi Lee and us.

A total of 150 subjects were estimated to be required. Given the above assumptions, a sample size of 150 (75 Staged CAS: 75 regular CAS) will be required. To account for an estimated 5% dropout rate, approximately 158 patients will be enrolled. Of these 158 patients, 79 will be randomized to the Staged CAS arm and 79 will be randomized to the regular CAS arm.

The sample size calculation was performed using SAS software, version 9.3 (SAS Institute).

Collection of data during the procedure

• Nerve function evaluation
• NIHSS
• mRS
• Blood pressure
• TCD
• Laboratory examination
• CT, MR