This study was a randomized, parallel-group, single-blind clinical trial that tried to assess the efficacy and safety of VRVT for myopia control which was conducted in Shanghai, eastern China for 3 months. It was registered at Clinicaltrials.gov (NCT06250920) and followed the tenets of the Declaration of Helsinki and CONSORT guidelines. Poster advertisements were used to inform and recruit participants at the study site. Participants were randomly assigned to either the intervention group (receiving 20 min of VRVT per day) or the control group (only SVS without receiving VRVT). The participants were enrolled in December 2020, and follow-up was completed in August 2021. All children submitted a written informed consent form signed by themselves and their parents or guardians. All examinations at baseline and follow-up visits were performed by the same examiners using the same protocol and equipment throughout. Investigators and key personnel involved in the present study were trained before the study commencement. This study was approved by the Ethics Committee of Shanghai Tenth People’s Hospital (identifier, SHSY-IEC-4.1/20–260/01). All the datasets used throughout the study were identified before being transferred to the study investigators.

Eligible participants were children aged 8 to 13 years with myopia of cycloplegic spherical equivalent (SE) between − 0.50 diopters (D) and − 3.00 D, astigmatism less than − 1.00 D, anisometropia less than 1.50D, and best corrected visual acuity (BCVA) more than 0.0 logarithm (LogMAR) of the minimum angle of resolution in either eye. Participants were willing to participate in the study and accept random allocation in grouping. They were asked to return to the hospital at recruitment, 1 month, 3 months, and 6 months (3 months after the end of all interventions) to cooperate with data collection.

Children were excluded if they had ocular diseases, such as amblyopia, strabismus, binocular vision abnormalities, and other ocular abnormalities in either eye. Children who were using orthokeratology or other optical methods for myopia control, and with systemic diseases (e.g., endocrine, cardiac, respiratory diseases) and developmental anomalies were also excluded.

Eligible children were allocated randomly to either the intervention or the control group with the random number table method, after verifying participant eligibility and obtaining written informed consent. Due to the nature of the intervention, participants and their guardians were aware of the study allocation. The intervention group of the participants was unknown to the outcome assessor (e.g., optometrists, and statisticians).

All children were required to wear single-vision spectacles for participation throughout the study and updated their spectacles if needed. Children in the intervention group were instructed to receive VRVT, who were administered at home under the supervision of parents for 20 min per day with VRVT. Children in the control group lived without receiving VRVT. With the VR technology, twelve different scenarios were constructed in the solar-like full-spectrum illumination scenes, which have various training sessions and purposes. When the children’s binocular fixation point focuses on the set object like a butterfly, for a certain period of time, it will cause the object to move systematically (Supplement 1). Children were asked to follow the movement of the object with binocular fixation. The intervention adopted a virtual reality visual training system.

In the intervention group, children were asked to record a single English word that appeared randomly and feedback to investigators and their parents during the last training session every day for 3 months, and this method was used to ensure compliance with the intervention.

The outcome was the efficacy and safety of VRVT in myopia control. The primary outcome was changes in axial length (AL) at 3 months. Macular choroidal thickness (mCT) was regarded as a key secondary outcome. Children who underwent at least 1 session of VRVT intervention were analyzed for safety. A questionnaire on adverse events, including but not limited to VR vertigo and asthenopia, was collected from children at each follow-up and any unplanned visits if needed.

Professional optometrists performed all examinations, including uncorrected visual acuity (UCVA) and BCVA at a standard testing distance of 4 m, near visual acuity (NVA) at 40 cm, non-cycloplegic and cycloplegic auto-refraction (KR-8900, Topcon, Japan), non-contact intraocular pressure (IOP) measurement (CT-IP, Topcon, Japan). After 1 drop of 0.5% Alcaine (Alcon, Fort Worth, USA), the cycloplegic agent used in this study was 1% cyclopentolate hydrochloride (Alcon, Fort Worth, USA); one drop was applied every 5 min for a total of three drops. Before observing pupil response and diameter, participants were asked to close their eyes and rest for 30 min. When the pupillary response disappears or the pupil diameter is greater than 6 mm, auto-refraction can be performed. The spherical equivalent was calculated as spherical power plus half of the cylinder power.

The axial length (AL) and anterior chamber depth (ACD) were measured using laser interferometry (IOL-Master 700, Carl Zeiss Meditec AG, Germany) which were measured three times and averaged. The subfoveal choroidal thickness (the distance between outer choroid episcleral margin and retinal pigment epithelium Bruch’s complex) was performed by enhanced depth imaging optical coherence tomography (CIRRUS HD-OCT 5000, Carl Zeiss Meditec AG, Germany). It was measured by the built-in measurement software. The measured data will be accepted when the image signal intensity index is more than 7. Stereoacuity was assessed by using the Titmus test (Stereo Optical Co., USA). All study data were collected at recruitment, at 1 month, at 3 months, and 6 months.

Based on the preliminary studies, the change of AL was estimated to be 0.10 mm and 0.15 mm for the intervention and the control group respectively after 3 months. A sample size of 50 participants (25 per group) was selected to achieve 90% power at a significance level of 0.05. The final sample size of 60 participants (30 per group) was selected after factoring in a projected attrition rate of 20%.

Participants’ baseline demographic information and ocular parameters were summarized using descriptive statistics. Continuous variables were reported in terms of means, SDs, and 95% confidence interval (CI). Categorical variables were reported in frequencies and percentages. The independent-sample t-test was used to compare differences between the groups.

Intervention efficacy was calculated by dividing the between arm difference in values by the control arm value. Right eyes that met the enrollment criteria were used as the outcome data representing the participant. If the right eye did not meet inclusion criteria or if right eye data were missing, left eyes were used instead (n = 2). All adverse events were reported individually in detail.

All P-values were 2-sided and considered statistically significant when the values were 0.05. All statistical analyses were performed using SPSS Statistics software (version 25.0, IBM Corp., USA).