Osteoarthritis (OA) is one of the most common joint disorders and the leading causes of work restrictions and decreased quality of life in the elderly (Dunlop et al., 2003). The cartilage degeneration is commonly considered to be the initial pathological defect in OA. However, current clinical interventions for OA, including knee arthroplasty, are all during the stage of managing symptoms rather than regenerative medicine (Coleman et al., 2010; Gademan et al., 2016; Kizaki et al., 2019). With the aging of the population, osteoarthritis patients are increasing year-by-year and effective cartilage regeneration approaches are urgently needed.

Recently, adult mesenchymal stem cells (MSCs) or adipose-derived stem cells (ADSCs) have emerged as promising candidates with great healing potential in regenerative medicine because of their capacity to differentiate into multiple tissue types and to self-renew (Richardson et al., 2016; Gu et al., 2017; Vural et al., 2017; Bonaventura et al., 2020). Nevertheless, before utilizing stem cells for cartilage regeneration by tissue engineering methods, in vitro amplification is necessary to obtain sufficient number of cells before transplantation. However, local microenvironment of culture media can affect stem cells differentiation and ex vivo manipulation may result in genetic changes that may affect functional and biological characteristics. Moreover, progenitor cells are able to regenerate damaged articular cartilage, but the lack of vascularity in cartilage region prevents the infiltration and survival of the implanted stem cells. Therefore, recent regenerative medicine switched to apply stromal vascular fraction (SVF) instead of MSCs or ADSCs, aiming to resolve these limitations in tissue engineering.

Stromal vascular fraction (SVF) isolated from adipose tissues enriched adipose stem cells (ADSCs) as well as other stromal cell types involved in tissue regeneration such as endothelial cells, pericytes, and fibroblasts (Charles-de-Sa et al., 2015; You et al., 2015). These various cell components in SVF may act synergistically with stem cells in facilitating and promoting tissue regeneration (Ramakrishnan and Boyd, 2018). More importantly, SVF cells can be readily obtained from liposuction procedure without the need for any cell propagation, which make it more convenient and feasible for clinical application than culture-expanded ADSCs (Pak et al., 2016a; Bora and Majumdar, 2017; Desando et al., 2019). The clinical applications of SVF therapy for osteoarthritis (OA) have attracted academic and clinical attention (Comella et al., 2017; Farre-Guasch et al., 2018; Tran et al., 2019; Bonaventura et al., 2020). On the ClinicalTrials.gov website, 34 clinical observations utilizing SVF have been registered, of which 7 are for the treatment of osteoarthritis. In the limited published clinical observations, visual analog score (VAS) for pain, functional rating index (FRI), range of motion (ROM) of the patients, and the Western Ontario and McMaster Universities Arthritis Index (WOMAC) questionnaire were assessed to evaluate the outcome of the treatment (Koh et al., 2014; Koh et al., 2015; Fodor and Paulseth, 2016). However, these measurement methods do not provide direct evidences of cartilage degeneration or regeneration.

Conventional Magnetic Resonance Imaging (MRI) can be performed to assess the extent of cartilage damage (Baysal et al., 2004), but cannot be analyzed quantitatively. In addition, most studies utilize two-dimensional MRI images to compare the changes in cartilage between different time periods of the same patient. However, there is no reliable criterion or protocol to ensure that the two images of the MRI section of the same patients from different time periods are actually anatomically identical. At present, there is a great need for a simple and non-invasive evaluation method to analyze the changes of joint cartilage qualitatively and quantitatively in clinical trials.

Herein we reported a randomized clinical phase I/II trial to evaluate the safety and therapeutic effect of autologous SVF cellular therapy on knee OA. In addition, a novel method to provide 3-dimension (3D) MRI imaging for human knee cartilages and bones was developed, enabling us to qualitatively and quantitatively evaluate cartilage regeneration in this study.

Study participants voluntarily provided written informed consent to participate in the study and signed the Health Insurance Portability and Accountability Act (HIPAA) authorization letter before participating in any research procedures. The study was reviewed and approved by the Medical Ethical Committee of Zhongnan Hospital of Wuhan University and was conducted in accordance with the Declaration of Helsinki, International Conference on Harmonization (ICH).

Since 2011, Pak reported for the first time a series of knee OA cases treated with autologous adipose SVF (Pak, 2011), the clinical applications of SVF therapy for OA have attracted academic and clinical attention. Several studies or clinical observations on the application of SVF therapy for knee OA were reported (Koh and Choi, 2012; Pak et al., 2013; Koh et al., 2015; Pak et al., 2016b; Fodor and Paulseth, 2016; Tran et al., 2019). The safety and effects of relieving pain and improving function were observed in most patients. However, The lack of a quantitative measurement to provide convincing evidence of cartilage regeneration makes it impossible to obtain a strong conclusion on the correlation between clinical improvement and cartilage regeneration (Orozco et al., 2013; Koh et al., 2014; Orozco et al., 2014; Koh et al., 2015; Vega et al., 2015; Fodor and Paulseth, 2016). In addition, in some studies the inclusion criteria of the study subjects are K-L III-IV, indicating the progression of arthritis has entered a stage that requires surgical intervention (Orozco et al., 2013; Pak et al., 2016b; Pers et al., 2016; Sampson et al., 2016).

In this study we established a novel protocol to provide 3D MRI imaging for human knee cartilages and bones. The establishment of both cartilage and bone 3D MRI models of the knee joint is of crucial importance to achieve the goal to qualitatively and quantitatively evaluate cartilage regeneration. Since the knee articular cartilage is a thin layer of cartilage attached to the articular surface of femur and tibia, the articular surface of femur and tibia can be used as a reference to identify a thin bowl-shaped layer of cartilage. In addition, to achieve quantitative comparative analysis of the same patient’s articular cartilage MRI imaging data between two different periods (refer to MRI imaging data I and II), it is most critical to spatially align the cartilage section of MRI imaging data I and II in the same 3D spatial domain. Due to the minimum change of joint osteogenesis over a short period of time, and its well-defined anatomic structure, we spatially align the MRI imaging data I and II based on the osteogenic section to ensure that the cartilage section was registered in the same 3D spatial domain (Figures 1j–n). This ensure the cartilage volume; area and thickness of two different time periods are comparable. In this manner, the cartilage volume, area, and thickness of the same patient between two different time periods can be quantitatively compared (Figure 3).

Cartilage degeneration is commonly considered to be the initial pathological defect in OA. In this study, we hypothesized that SVF therapy facilitates and promotes cartilage regeneration and thus will have profound effects to control and reverse the cartilage degeneration in the OA patients with cartilage defect grades I/II, the early and mid stage of OA. In this pilot study six OA patients with both knee cartilage defect I-II were enrolled. The two knees of each patient were randomly assigned to autologous SVF (10⁸ cells) treatment group or non-treatment control group, thus this study minimized the potential impact of the differences of age, sex, BMI, life style, and medications between the treatment group and the control group on the interpretation of the study’s outcome. We estimated in this pilot study six patients were able to provide data for large scale clinical trails.

In this study, the adipose tissue harvest and SVF cells processing protocols were followed within the guidelines of the International Federation for Adipose Therapeutics (IFATS) and Science and the International Society for Cellular Therapy (ISCT) (Bourin et al., 2013). SVF products met releasing criteria, including cell viability test, endotoxin test, and bacteriologic test. In this study, we did not register any adverse events, confirming recent literature which reported that the SVF therapy for knee OA is a safe procedure (Pak et al., 2013; Koh et al., 2015; Fodor and Paulseth, 2016).

Similar to previous pilot studies, the WOMAC score decreased from a preoperative mean of 85.33 ± 22.19 to a postoperative mean of 43.00 ± 19.93 at 12 weeks and 37.33 ± 19.76 at 24 weeks, which represents decreasing pain and improving movement. Lysholm Knee Scale, another recommended method for measuring knee function, increased from 42.50 ± 15.76 at the baseline to 64.83 ± 13.75 at 12 weeks, and 66.10 ± 16.09 at 24 weeks. No statistically significant differences were detected at most time points in both the WOMAC score and Lysholm score which may be related to the limited sample size. The patient serving as self-control also affected the pain and function evaluation of SVF treated knee.

In this study, we qualitatively evaluated knee cartilage defects 3D MRI imaging. As illustrated in Figure 4 detailling that most defects of femur and/or tibia cartilage of the SVF treated group demonstrated improvement, some showed significant reduction at 12 and 24 weeks in comparison with 0 week. In contrast, most the defects of untreated group showed deterioration and some new defects were noted.

The quantitative evaluation of 3D MRI imaging of knee cartilage demonstrated that there was significant increase of cartilage value both in defect cartilage area and whole cartilage area of treated group (Figure 5). The cross-sectional area and thickness of the femoral and tibial cartilage at different positions were also significantly increased. This study shows that the defect cartilage achieved around 1.67 times improvement 12 weeks after SVF injection (from 112.96 ± 56.85 to 301.42 ± 298.05). Even in the cartilage non-defect area, the thickness and cross-sectional area were also improved to some extent. No obvious cartilage repair was observed from week 12 to week 24, which may indicate that the repair of cartilage by SVF therapy mainly occurred during the 12 weeks post treatment. In contrast, there was no significant change of cartilage value, thickness, and area within the untreated group (p > 0.5). Since only 6 patients were recruited in our clinical phase I/II study, a large-scale clinical phase III trial is required to further confirm the safety and efficacy of SVF therapy in treating chondral defects of knee osteoarthritis.

Our clinical phase I/II study indicated that SVF therapy is a safe clinical procedure for human knee OA. This study provided convincing evidence that SVF therapy reduced cartilage defects; significantly increased cartilage value, cross-sectional area and thickness, indicating SVF therapy facilitate cartilage regeneration in cartilage defect I-II knee OA patients 12–24 weeks post treatment. In this study, we have established a novel protocol to provide 3D MRI imaging for human knee cartilages and bones enabling us to qualitatively and quantitatively evaluate cartilage regeneration. The ZhongnanH articular cartilage defect index may provide a new tool to standardize diagnoses of cartilage defects of knee OA. This pilot study provided novel protocol and data for future large-scale phase III trial to further confirm the efficacy promises of SVF therapy.

From a public health point of view, if we are able to make early definite diagnoses for all knee OA patients, and provide safe and efficient therapy to facilitate cartilage regeneration, we will be able to control or reverse cartilage degeneration and completely change the epidemiology of OA worldwide. We hope our study will make a significant contribution to achieve this goal.