Safety and Efficacy of Autologous Adipose-Derived Regenerative Cells (ADRCs) Injection for Improving Diminished Ovarian Reserve

Trial design and setting This prospective, self-controlled clinical trial aims to evaluate the efficacy and safety of autologous ADRCs injection in improving DOR. A total of 25 patients diagnosed with DOR at the outpatient clinic of the Reproductive Medicine Center, Peking University People's Hospital, who meet the predefined inclusion and exclusion criteria, will be enrolled in the study.

Autologous adipose tissue will be harvested from patients via liposuction, and the automated autologous adipose tissue processing system Celution® system will be applied to process the adipose tissue for autologous ADRCs isolation and concentration. Subsequently, autologous ADRCs will be injected into the medulla of both ovaries (2-2.5 mL per ovary). Patients will undergo follow-up assessments at 1, 2, 3, 6, and 12 months to evaluate the safety and efficacy of the treatment.

Eligibility Criteria Potential subjects will undergo assessments within 4 weeks prior to the intervention, including medical history and physical examination, menstrual pattern assessment, and ovarian reserve testing (on days 2-5 of the menstrual cycle) involving transvaginal ultrasound for ovarian volume and AFC measurement, as well as serum hormonal assays to quantify AMH and FSH.

Inclusion criteria

1. Female patients aged 20-39 years.
2. Diagnosed with DOR, defined as AFC < 7 and/or AMH ≤ 1.1 ng/mL.
3. FSH ≤ 40 U/L.
4. Have signed informed consent for the clinical trial.
5. In good general health and able to tolerate local anesthesia, liposuction, oocyte retrieval, as well as pregnancy and childbirth.
6. Without relative contraindications to fertility such as hereditary tumor-related diseases.

Exclusion criteria

1. Presence of thyroid disorders.
2. Body Mass Index (BMI) < 18 kg/m².
3. Confirmed pregnancy (positive β-hCG with ultrasound confirmation of embryo implantation).
4. History of autoimmune diseases.
5. Family history of ovarian tumors.
6. Prior pelvic or systemic radiotherapy for cancer treatment.
7. Positive serology/virology results for human immunodeficiency virus (HIV), Hepatitis C, Hepatitis B, Syphilis, human papillomavirus (HPV), or abnormal ThinPrep cytology (TCT).
8. DOR caused by genetic disorders.
9. Reproductive system pathologies including polycystic ovary syndrome (PCOS), adenomyosis, uterine fibroids ≥ 5 cm, endometrial carcinoma, or uterine malformations.
10. Contraindications to liposuction.
11. Abnormal semen analysis of the male partner precluding natural conception or conception via assisted reproductive technology (ART).
12. Malignancy-related conditions such as active malignancy without complete remission or chemotherapy administration within the past 6 months.

Withdrawal criteria

1. Voluntary withdrawal of consent by the participant or loss to follow-up.
2. Protocol non-compliance involving significant deviations that may affect the validity of study results.
3. Ethical discontinuation, as determined by the investigator based on medical judgment indicating an unacceptable risk to the participant.
4. Occurrence of serious adverse events (SAEs) that preclude continued participation.
5. Engagement in confounding activities that compromise the integrity of study data.

Interventions

The patient undergoes lower abdominal liposuction (with an aspirate volume of approximately 100-360 mL) under general anesthesia administered. Immediately after the procedure at the operating room bedside, ADRCs are prepared using Celution® system (Cytori Therapeutics, Inc., Shanghai, China) and subjected to quality control as follows:

Firstly, under aseptic technique, the lipoaspirate is injected into the processing container. Following the device's prompts, the automatic cleaning procedure is initiated. The volume of composite enzyme indicated by the device is aspirated and added to the processing container. The automated system then proceeds to handle key steps including separation, centrifugation, and resuspension of the cellular fraction, ensuring standardized and consistent processing of ADRCs. Following 90-120 minutes of automated processing, the operator draws up a measured 5mL aliquot of ADRCs suspension. Of this volume, 0.5mL is retained for quality control testing, including assessments of cell count, viability, and surface marker expression; the remaining 4.5mL is used for injection into both of the patient's ovaries, with 2-2.5mL administered per ovary.

After the procedure, the patient remains supine for 30 minutes, with close monitoring for local reactions (e.g., abdominal pain, bleeding) or systemic reactions (e.g., allergic reactions). The patient is discharged if no abnormalities are observed.

Safety evaluation and AEs Monitoring To assess the safety profile of this trial, all adverse events including local or systemic reactions, will be closely tracked with detailed documentation of their onset, duration, severity, and frequency.

Risk mitigation strategies

The trial encompasses three key procedures: abdominal liposuction, automated preparation of ADRCs, and intraovarian ADRCs injection. Targeted strategies are implemented to mitigate risks across these steps:

For abdominal liposuction and intraovarian injection, preoperative measures include establishing intravenous access to enable rapid emergency response, while operators are required to master local anatomical details to avoid critical blood vessels and nerves. Intraoperatively, gentle manipulation is emphasized to reduce vascular injury and prevent accidental intravascular entry of adipose tissue or ADRCs; prior to injection, aspiration is performed to confirm the needle is not within a vessel, and a retrograde injection technique is adopted for ADRCs delivery.

In the automated ADRCs preparation process, standardized material preparation and pre-procedural equipment checks (conducted by designated staff) are mandatory. The automated system undergoes regular maintenance and calibration to ensure stable operation. Should equipment or consumable issues arise during preparation, the manufacturer's technical support team will be promptly consulted for troubleshooting.

Surveillance and management of potential AEs

Potential adverse events may arise from liposuction, injection, or ADRCs preparation, with corresponding management protocols in place:

For events related to liposuction or injection, fat embolism-though extremely rare-represents a critical concern, potentially presenting as chest tightness, visual impairment, or severe localized pain. Upon such presentation, the procedure will be halted immediately, and clinicians will administer symptomatic treatment.

Other potential reactions include bleeding, infection, abdominal pain, or allergic responses. Bleeding will be managed with hemostatic agents as needed, while infection prevention relies on strict adherence to aseptic techniques. Mild abdominal pain typically resolves with rest, but severe pain may require analgesics after definitive diagnosis. Allergic reactions, though rare, will be monitored closely, with protocols in place to address severe manifestations.

For AEs linked to ADRCs preparation, equipment or consumable malfunctions that could compromise product quality will be documented, and the manufacturer's support team will be engaged immediately to resolve issues, ensuring ADRCs safety before injection.

Continuous monitoring will be maintained throughout the procedure and post-procedural recovery to enable timely identification and management of any adverse events, prioritizing participant safety.

Outcome Measurements Primary composite endpoints

1. Cumulative clinical pregnancy rate within 12 months post-treatment: Conception achieved via natural pregnancy or assisted reproductive technology (ART), confirmed by ultrasound detection of fetal cardiac activity. Subgroup analyses will compare natural pregnancy versus ART-conceived pregnancy.
2. Changes in ovarian reserve markers among non-pregnant patients: Measured at 1-, 2-, 3-, and 6-month post-treatment. Parameters (FSH, AMH, AFC) will be assessed on days 2-5 of the menstrual cycle and analyzed using a repeated-measures mixed-effects model.

Secondary endpoints

1. 12-month changes in ovarian reserve markers among non-pregnant patients: FSH, AMH, and AFC (assessed on days 2-5 of menstruation) will be compared with baseline values and analyzed using paired t-tests.
2. Recovery of menstrual patterns among non-pregnant patients: Documented at 1-, 2-, 3-, 6-, and 12-month post-treatment, with binary outcomes ("recovered" vs. "not recovered") analyzed using Cochran's Q test.
3. Changes in ovarian volume among non-pregnant patients: Average bilateral ovarian volume (measured via ultrasound) will be assessed at 1-, 2-, 3-, 6-, and 12-month post-treatment and analyzed using a repeated-measures mixed-effects model.

Safety endpoint Incidence of treatment-emergent adverse events (TEAEs) related to ADRCs injection within 12 months post-treatment.

Statistical Analysis Methods The statistical analysis plan will be developed by the biostatistician and principal investigator based on the study protocol and finalized as a document prior to database lock. The primary endpoints are "Clinical Pregnancy" or "FSH, AMH, and AFC levels at various post-procedural time points". The full analysis set will be used for data analysis, which includes all enrolled subjects. For subjects who discontinue the study, data from their last follow-up visit will be carried forward as the final data point for analysis (Last Observation Carried Forward). Statistical analyses will be performed using SAS or R software.

Primary endpoint analysis

1. Clinical pregnancy rate: The pregnancy rate and its 95% confidence interval will be calculated using the single-group rate method (Clopper-Pearson method). Subgroup analyses will be performed, such as comparisons between natural pregnancy and assisted reproduction.
2. Changes in ovarian reserve markers (FSH, AMH, AFC): Changes from baseline at 1-, 2-, 3-, and 6-month post-procedure will be analyzed using a Mixed Model for Repeated Measures (MMRM). The model will include time and baseline value as fixed effects, and subject as a random effect. Data will be censored after pregnancy (i.e., only data from subjects who have not achieved pregnancy will be included in the analysis).

Secondary endpoint analysis

1. 12-Month change in ovarian reserve markers (FSH, AMH, AFC): Paired t-tests will be used to compare the differences between baseline and the 12-month time point in subjects who have not achieved pregnancy.
2. Menstrual status follow-up: Menstrual resumption will be described by recording the percentage of subjects who resume menstruation. Cochran's Q test (suitable for binary outcomes) will be used to compare the number of subjects categorized as "resumed" vs. "not resumed" over multiple time points (e.g., baseline, 1 month, 3 months).
3. Change in ovarian volume: Changes from baseline at each time point (average volume of bilateral ovaries measured by ultrasound (B-ultrasound)) will be analyzed using a Mixed Model for Repeated Measures (MMRM).

Safety Analysis

1. Adverse Event (AE) incidence: The incidence rate and its 95% CI will be calculated (Clopper-Pearson method). AEs will be described and categorized by severity.