Adipose tissue derived stem cells: in vitro and in vivo analysis of a standard and three commercially available cell-assisted lipotransfer techniques

Fat grafting is a widely used technique that has been developed by plastic reconstructive and aesthetic surgeons [1, 2]. Adipose tissue has been used for many years as autologous filler in breast augmentation, hemifacial atrophy, facial rejuvenation and rhinoplasty [3]. However, the potential regenerative effect of fat grafting has been hypothesized and, to some extent, experimented only in recent years, as in the treatment of tissues damaged by radiotherapy [4]. The employment of adipose tissue, both as autologous filler and as a source of adipose stem cells (ASCs), has introduced the concept of regenerative therapy into plastic reconstructive surgery, thus leading to the publication of a large body of literature that supports the regenerative properties of the fat grafting techniques. ASCs are indeed mesenchymal stem cells characterized by the ability to differentiate, at least in vitro, into multiple cell types of all three germ layers, including adipocytes, smooth muscle cells, endothelial cells, neurons and hepatocytes [5–11].

Despite the reported promising results, however, lipofilling is burdened by variable and unpredictable graft-take rates in the postoperative period, with reported absorption rates varying from 20 to 90% [12].

The variability in fat graft survival has been attributed to several causes, such as the local angiogenesis, the donor site, sampling and processing techniques, and the type of receiving area and its characteristics. Summarizing, three main theories have been reported that try to explain the evolution of the grafted adipose tissue: Neuhof in 1923 advanced the host-replacement theory [13]; Peer in 1956 outlined the importance of adipose cell survival [14]; and Yoshimura and colleagues in 2009 demonstrated the primary role played by ASCs on the rate of fat engraftment [15]. According to this latter theory, after the lipofilling procedure, almost all of the injected cells die within a few months and are replaced by differentiation of the mesenchymal stem cells contained in the stromal vascular fraction (SVF) [16], the portion of the adipose tissue particularly enriched in ASCs [11].

For this reason, it has been proposed that the enrichment of adipose tissue with ASCs could contribute to a de novo formation of adipocytes and blood vessels, thus promoting long-term volume retention of the autologous fat graft. Consequently, a new approach named cell-assisted lipotransfer (CAL) has been introduced [17]. CAL consists of freshly isolating SVF cells from one-half of the aspirated fat and recombining the isolated cells with the other half, thus using, for reconstructive purposes, autologous primitive cells in combination with lipoinjection. This process allows converting relatively ASC-poor aspirated fat to ASC-rich fat [17].

Consistently, in the last years, numerous devices intended for clinical use have been developed and commercialized. These devices are aimed at increasing the concentration of primitive cells in the fat graft, by automating and standardizing the procedures for extracting, washing and concentrating the SVF from the autologous adipose tissue, thus ensuring safety and reproducibility of the procedures [3, 15, 18–23].

Although some of these clinical devices, often relying on different principles, have been described in a number of scientific reports specifically focused on the effectiveness and safety of the new proposed technologies, very few papers have directly compared these different procedures in terms of SVF and ASC enrichment as well as clinical results [18, 24]. In fact, if the long-term volume retention of the autologous fat graft is guaranteed by the presence of ASCs, it has to be demonstrated that these cells are more abundant in the enriched-adipose tissues and that they are able to persistently proliferate and differentiate into adipocytes and vessels.

In this study we analyzed four different lipotransfer techniques used for breast reconstruction: the gold-standard technique used in our department, which is a modified Coleman’s procedure, and three different devices available on the market – the Cytori Celution System (Cytori Ltd., Deeside, UK), the Lipokit Medikhan System (Medikan International Inc, Pusan, Korea) and the Fastem Corios (CORIOS Soc. Coop, San Giuliano Milanese (MI), Italy). Specifically, we focused our attention on the ASCs isolated from the different products, studying stem cell yield, proliferation and differentiation ability. This paper has: optimized an in vitro protocol to compare, for each patient, standard fat grafts and stem cell-enriched fat grafts, in terms of SVF cells and ASCs; demonstrated the effectiveness of the enzymatic stem cell enrichment procedures in enriching fat grafts in primitive cells; and presented preliminary clinical data showing that the use of adipose tissue enriched in stem cells seems to correlate with a better persistence of the fat graft at 6 and 12 months.

The independent ethics committee of the Azienda Ospedaliero – Universitaria of Udine approved the research (Parere N. 103/2011/Sper). Written informed consent was obtained from patients and all clinical investigations have been conducted according to the principles expressed in the Declaration of Helsinki.

Adipose tissue has been used for a century as autologous filler. After the first attempts, explored by Neuber and Czerny at the end of the 19th century [1, 2, 34], lipofilling techniques have undergone a huge development in the 1990s thanks to Coleman’s work, rapidly gaining widespread diffusion [35, 36].

The success of this technique is strictly related to the properties of the adipose tissue, an easily accessible source that can be collected by procedures that are relatively simple, safe and repeatable with minimal morbidity. Moreover, since the adipose tissue is available in the majority of the patients, it allows autologous approaches devoid of immunogenic drawbacks. The main disadvantage of lipofilling remains the unpredictable fat graft’s resorption rate that may vary between 20 and 90% [12, 37].

Since the demonstration that adipose tissue is an abundant and accessible source of multipotent stem cells [11], the concept of stem cell-enhanced fat grafting has been formulated [17]. Briefly, the hypothesis is that the supplementation of fat grafts with cells obtained from the SVF, a cell population known to be enriched in ASCs [32], would improve graft stability, taking advantage of the capacity of ASC to reduce apoptosis, contribute to vessel formation and maintain a self-renewing reservoir of cells able to give rise to all the mature cell types constituting the adipose tissue [18, 38, 39]. In this regard, the first clinical studies of the Yoshimura group – showing the ability of the so-called CAL to improve the results of the lipostructure technique, leading to a more predictable and stable result [17, 39, 40] – were further confirmed by other groups both at clinical level [23, 41, 42] and preclinical level [38, 43–45]. Led by the accumulation of data documenting safety and efficacy of CAL approaches, different automated, semiautomated and manual devices to be used in the operating room at the point of care have been developed to isolate with safe and standardized protocols the SVF and to produce SVF-enhanced fat grafts [18, 21, 39, 46–50]. However, it is difficult for the clinician to choose among the available devices, since very few comparative studies have been performed that have taken into account systematically both biological and clinical criteria. In fact, it would be advisable to have standards based on reproducible parameters and endpoints that could allow comparing different techniques [32, 51].

In this paper, we compared a standard lipotransfer technique, based on Coleman’s procedure, with three commercially available cell lipotransfer techniques. Two of the devices, Cytori [18, 22] and Lipokit [18, 49], are closed systems allowing an enzymatic separation of the SVF cells, while the third, Fastem, is a closed system of filtering bags that mechanically isolates the SVF.

The comparison took into consideration both biological and clinical criteria. Regarding the first evaluation, we compared, for each patient, standard lipoaspirates with the respective stem cell-enriched ones in terms of yield, immunophenotype and colony-forming capacity of the SVF cells as well as immunophenotype, clonogenicity and multipotency of the culture-expanded ASCs [32]. Regarding the clinical criteria, we compared, in terms of subcutaneous thickness gain as detected by ultrasound imaging, the group of patients treated with stem cell-enriched lipoaspirates with a group of patients treated with the standard procedure.

Considering the cell yield, a higher number of cells could be isolated from lipoaspirates enriched by both of the enzymatic methods. Regarding the characterization of the isolated cells, Cytori and, especially, Fastem significantly enriched the lipoaspirate product in hematopoietic cells, while the CD34⁺CD31^-CD45^- putative stromal cell-containing population [32] was higher in both Lipokit and Cytori enriched samples. A recent report by Yoshimura’s group trying to define the cellular events guiding the graft take in a murine model showed that CD34⁺ cells were proliferating in the first 2 to 4 weeks after grafting [52], while a xenograft transplantation study by Rubin’s group suggested that the concentration of CD34⁺ progenitor cells within the SVF could be used to predict the human fat graft retention [45].

Accordingly, Cytori and Lipokit enriched lipoaspirates were also characterized by an increased number of colony-forming unit cells. The colony-forming unit assay, commonly used to assess the frequency of progenitor cells [32], supported an increased number of primitive cells in Cytori and Lipokit enriched products. The trend was evident also for the Fastem procedure, although it did not reach statistical significance.

We further analyzed the stem cell-enriched lipoaspirates isolating, culturing and characterizing ASCs. Many works have been conducted showing that stem cell enrichment devices, such as Cytori, are able to isolate ASCs, thus allowing the surgeon to perform a CAL [18, 21, 22]. Additionally, recent work by Dominici’s group has demonstrated, in a rabbit subcutaneous adipose tissue autologous regeneration model, the capacity of ASCs to reduce cell apoptosis favoring a long-term graft performance [38]. More importantly, a randomized placebo-controlled trial has demonstrated that enrichment of autologous fat grafts with ex vivo expanded ASCs significantly improved graft survival [37].

In our study, ASCs could be isolated both from non-enriched and stem cell-enriched samples. However, differences could be detected in ASC phenotype and function at the third passage in culture. Specifically, a larger fraction of ASC obtained from Lipokit and Cytori enriched lipoaspirates expressed stem cell markers such as Nanog, Sox-2 and c-Kit, although differences reached statistical significance only for Cytori. Similarly, ASCs from the stem cell-enriched lipoaspirates tended to grow faster. Regarding clonogenicity, Lipokit and Cytori derived ASCs were characterized by a double number of cells able to grow when seeded as single cells. Similarly, ASCs obtained from the samples enriched by Lipokit and Cytori were characterized, with respect to those obtained from the non-enriched lipoaspirates, by a significantly increased ability to differentiate into adipocytes (Lipokit and Cytori), endothelial cells (Cytori) and muscle cells (Cytori). Conversely, the mechanic Fastem enrichment procedure seemed not to significantly enrich the samples in clonogenic and multipotent ASCs.

Altogether, the cytological criteria analyzed in this paper seem to indicate a clear ability of the enzymatic-based method to enrich the samples in both vascular stromal cells and ASCs. In a recent study, Aronowitz and Ellenhorn compared four different commercial cell separation systems – Cytori, Lipokit, PNC’s Multi Station (PNC. International, Gyeonggi- do, Republic of Korea) and CHA Biotech Cha-Station (CHA Biotech, Kangnamgu, Republic of Korea) – evaluating the cell yield, frequency of CD34⁺CD31^-CD45^- and endothelial cells as well as colony-forming units of the isolated cells, showing a clear superiority for the Cytori system [18]. However, the residual enzyme levels observed with Lipokit was 57-fold higher than that observed with Cytori, and this can explain why in our work, in which we used not the SVF but the lipoaspirate enriched in the isolated cells, we did not observe such a clear difference between Cytori and Lipokit. In fact, addition of the isolated cells to the lipoaspirates could have neutralized the residual enzymatic activity, possibly reducing the negative effect of the enzymes on the isolated cells. This suggests the importance of the standardization of the sample to be used to define the properties of the donated fat graft. In Aronowitz and Ellenhorn’s work, moreover, the authors did not study ASCs and a clinical evaluation is missing.

Regarding the clinical results, in this pilot study we used an ultrasound evaluation of the subcutaneous thickness of the breast area corresponding to the SMQ and MMQ. The clinical evaluation of the results is a key point to establish the entity of the graft take and, ultimately, the effectiveness of the technique. In this regard, the magnetic resonance imaging and three-dimensional surface imaging are considered ideal examinations to quantify the entity of the graft takes. In fact, both techniques are non-invasive tools characterized by high accuracy and reproducibility in evaluating the effective volume persistence [53]. However, these methods are cost-effective, and, in the case of the three-dimensional surface imaging, no information on the breast tissue is available. To ensure the possibility to have easy, fast and low-cost measurements, we have employed an ultrasound device [54, 55]. The ultrasonographic analysis gives the possibility to extend the analysis to several patients and even to increase the frequency of the examinations, thus improving the evaluation of the fat graft persistence/resorption kinetic. In this paper we have shown that, 6 months after fat grafting, patients treated with stem cell-enriched lipoaspirates showed, with respect to those treated with the standard procedure, a significant superior gain in the subcutaneous thickness of both MMQ and SMQ; moreover, the significant superior thickness loss between 6 and 12 months in the MMQ of patients treated with standard lipoaspirates confirmed a better stability over time of fat grafts obtained with stem cell-enriched procedures.

The millimeter size of the thickness analyzed has made ultrasound imaging a proper, easy, and low-cost technique to be used for the measurements. Moreover, to overcome the subjective nature of ultrasound measurements, a dedicated team has been selected to control the patients and a standardization of the data collection procedure through a nine-quadrant breast pattern has been implemented.

Our results are in line with all reports indicating the clinical advantage of using the CAL procedure over a standard lipotransfer [17, 23, 39–42, 44].

Recently, Peltoniemi and colleagues raised doubts regarding the advantage of using stem cell enrichment in cosmetic fat transplantation to the breast [56]. Specifically, for breast augmentation they used either lipoaspirates obtained by water-assisted lipotransfer or water-assisted lipotransfer lipoaspirate enriched with stromal stem cells obtained using the Cytori system. The water-assisted lipotransfer method is based on the use of the Bodyjet System (Body-Jet; Human Med, Eclipse Ltd., Dallas, Texas), which allows a gentle liposuction through the aid of a water jet pulse dissecting the adipocytes from the subcutaneous tissue; its effectiveness, in terms of both fat graft survival rate and overall clinical outcome, has been reported by different groups [57, 58]. However, in Peltoniemi and colleagues’ work, no differences in the ASC phenotype was seen between the two lipoaspirates, thus suggesting that a more specific study on ASC presence and function in the two lipoaspirates would add crucial information.

Considering instead the three different devices, despite the fact that, accordingly with in vitro data, Cytori-treated patients displayed a trend to a better clinical outcome, no significant differences between the different CAL procedures could be detected, perhaps because of the small number of analyzed patients.

In fact, the major weakness points of this study are represented by the sample size of enrolled patients and by the short-term follow-up (12 months). A larger case study and a longer follow-up period are needed to confirm the results, and a double-blind ultrasound measurement study is needed to assess the accuracy and reproducibility of our standardized protocol. Moreover, a cost analysis on a large population would clarify whether CAL represents a real advantage. In fact, lipostructure techniques are widespread and usually patients must be subjected to repeated interventions to achieve the optimum result. The use of stem cells should increase the long-term graft performance, reducing the number of the surgical procedures required to reach the desired clinical outcome. The cost analysis should therefore take into account the costs related to the CAL procedures and the savings due to the reduction in the number of surgeries required.

In this paper, applying our in vitro protocol to isolate, quantify and characterize SVF cells and ASCs from fat grafts, we have demonstrated a significant superiority of the enzymatic stem cell enrichment procedures over the standard procedure. Accordingly, our in vivo protocol, based on the ultrasonographic assessment of the subcutaneous thickness of the breast area, confirmed the clinical advantage of CAL techniques in reducing the graft resorption at 1 year, especially in the central part of the breast.