Regenerative Medicine With Endometrial Regenerative Cells for Critical Ischemia: Limb Salvage From the Cradle of life?

Peripheral arterial disease (PAD) encompasses a spectrum of diseases involving arterial occlusive disease, aneurysmal disease and vasculitis. Of these, the most prevalent and insidious of conditions is arterial occlusive disease, which can clinically manifest as, in the mildest form, intermittent claudication to the most severe form, tissue necrosis. Critical limb ischemia (CLI) is manifested as rest pain, ulceration or gangrene and ultimately, if untreated, progresses to limb amputation. In the USA, there are between 100,000 and 150,000 amputations annually and over 80% of these are due to vascular insufficiency. Over the last two decades, there have been dramatic leaps in the ability to revascularize critically ischemic limbs with surgical and endovascular techniques, with limb salvage rates of 87% using dorsalis pedis artery bypass [1]. However, the limitation is not surgical technique but rather the availability of a bypass conduit that can keep patent at low flow rates. Currently, the only reliable conduit is autogenous vein and when this is not available, the ability to salvage limbs is dramatically reduced. Endovascular therapy has become a fundamental part of our treatment armamentarium and had made dramatic progress due to advances in technology, yet the durability of the treatment is often limited and often cannot be successfully performed in all patients. Despite decades of research, an alternative conduit to the saphenous vein does not exist and alternative means of revascularization remain a pressing need. It is within this context that regenerative medicine with stem cell therapy may offer new hope.

To date, no noninterventional therapy has made even a minimal impact on limb salvage. The most promising, and yet the most disappointing, candidate for revascularization had been ‘angiogenic therapy‘, sometimes described as a ‘biologic bypass equivalent‘. A potent angiogenic factor noted in many ischemic conditions, including cardiac ischemia, stroke and CLI, is vascular endothelial growth factor (VEGF) [2,3]; this particular cytokine has been extensively studied for its ability to induce angiogenesis, perhaps arteriogenesis, and also in its role as a vasculoprotective factor in restensosis. However, despite many promising animal studies, VEGF administration as a sole therapeutic agent for inducing arteriogenesis has failed in clinical studies performed thus far. There are a myriad of reasons but fundamentally there is the fact that the development of mature new blood vessels that can provide adequate nutritive flow requires more than a cytokine that stimulates endothelial cell proliferation and migration.

The lack of scaffolding in the development of an artery, whether it be pericytes, smooth muscle cells or perhaps adventitial cells, limits the functional capacity of a VEGF-induced blood vessel. In a double-blinded, placebo-controlled, randomized trial, VEGF plasmid injection intramuscularly resulted in no reduction in limb amputation rates, the primary end point of the study [4]. In another study, VEGF gene transfer increased ‘vascularity‘ as gauged by digital subtraction angiography but not in clinical end points. The limitation of these therapies is that we lack a good understanding of the milieu necessary for blood vessel formation, and thus the rather inelegant injection of protein or the particular gene plasmid can be expected to provide less than stunning results. It isn‘t that it is the wrong therapy; rather, it is administered in an inchoate context.

Regenerative medicine in the form of cellular therapy for CLI is based on the rationale that delivery of endothelial progenitor cells into areas of ischemia may result, not only in differentiation of the cells into endothelium and thus contributing to angiogenesis, but also in the production of various growth factors capable of stimulating angiogenesis through paracrine interactions with adjacent cells. Rather than providing the signaling cytokines alone with the goal of recruiting and stimulating the desired cell population, cell-based therapy directly aims to deliver the putative cell type into the theater, as it were, and thus accelerate the development of the cell population that would mature into new blood vessels.

The principal reason that this therapy has a better chance of success compared with angiogenesis is that it is not restricted by our limited understanding of the milieu necessary for the development of new blood vessels. This is not to say that it does not require it; rather, the therapy itself will foster a better understanding of what is necessary to grow new blood vessels and what impedes this development.

To date, the majority of research utilizing stem cell therapy has focused on bone marrow-derived stem cells (BMSCs) or endothelial progenitor cells (EPCs) and has formed the platform upon which future iterations of cell-based therapy can be considered. Postnatal revascularization occurs not only by the migration and proliferation of resident endothelial cells, but also by circulating progenitor cells from the bone marrow (BM) that then differentiate in the area of ischemia. The characterization of these EPCs remains problematic owing to the fact that certain monocyte-derived cells express endothelial cell markers does not translate to these cells having functional characteristics of endothelial cells. In addition, the actual volume of cells recoverable from peripheral blood that can be characterized as EPCs is very small.

In 2002, the first report on the use of BM cells for therapeutic angiogenesis for patients with limb ischemia was published by the Therapeutic Angiogenesis using Cell Transplantation (TACT) study investigators [5]. There was a pilot study in 25 patients with unilateral limb ischemia, who received intramuscular BM-mononuclear cell transplantation (BM-MNC; 1.6 × 10⁹ cells) in the ischemic leg and saline in the other leg as a control. In this study, there was a significant improvement in the ankle–brachial index, transcutaneous pO₂ and rest pain. In the following study, in 22 patients with bilateral limb ischemia, the ischemic limb was treated with BM-MNC and in the contralateral limb, peripheral blood-derived MNC were injected. There was once again a significant improvement in the BM-MNC limb and to a lesser degree in the peripheral blood-MNC limb, leading the authors to hypothesize that the BM-MNC fraction had higher concentrations of EPC compared with peripheral blood.

There are limitations to BMSC therapy; specifically, the extraction process is invasive and quite painful and the BM from elderly patients may not have the regenerative potential necessary for therapy [6], and in particular, patients with diabetes, PAD and other common comorbidities may have a markedly reduced angiogenic potential compared with age-matched controls [7]. Thus, an ‘off-the-shelf‘ solution as proposed by Murphy et al. may offer the most promise [8].

In one of the first seminal observations in 2004, Taylor demonstrated that in women undergoing BM transplantation, the endometrium was populated by cells derived from donor BM [9]. Given the high rate of angiogenesis associated with proliferation of the uterine lining, it is known that a high proportion of proangiogenic cells and perhaps EPCs may be found there. Murphy et al. have reported a unique population of cells derived from the menstrual blood, which after cloning they termed endometrial regenerative cells (ERCs) [10]. Meng et al. published data highlighting the fact that the ERCs represent a unique population of cells in contrast to endometrial stromal cells based on:

• Different rate of proliferation;

• Ability to differentiate into a wide number of tissues; stromal cells have only been shown to differentiate into bone, fat and cartilage;

• ERCs lack expression of STRO-1, whereas other stromal-derived cells expressed this marker.

Endometrial regenerative cells and BM-derived mesenchymal stem cells both express CD90 and CD105 and do not express CD45 and CD34. Murphy et al. reported an experiment using a chronic limb ischemia model of femoral artery ligation to assay angiogenic activity of ERCs implanted distal to the site of ligation [8]. In a total of 16 immunocompetent Balb/c mice, (eight in each group), the femoral artery and its branches were ligated and the peroneus muscle ligated to induce a neurotropic type ulcer. At day 14, the ERC-injected group showed complete limb salvage whereas in the control group, all eight mice developed necrosis. This striking result was highlighted by the astounding fact that human ERCs were used in these immunocompetent xenogeneic recipients.

In these particular studies, ERCs were injected at progressive intervals rather than in one injection; the optimal timing of the delivery of ERCs is an issue that remains to be investigated. There is also an unresolved issue regarding the immunomodulatory effects of ERC, as in BM-derived mesenchymal cells, active immunomodulatory properties of the cellular injectate has been demonstrated. Is it really the case that allogenic ERCs can be implanted without an immune response from the recipient? Detailed studies have to be performed to document the safety of this therapy.

Endometrial regenerative cells as therapy sound very promising based on early data but a lot remains to be investigated and resolved. Besides the obvious issues of dose and timing, a more fundamental set of questions remain on the horizon. Part of the reason that VEGF therapy has disappointed is that VEGF only affects a single, albeit important, component of the architecture of the blood vessel. The other equally critical components such as the scaffolding for the endothelial cells, the directionality of the blood vessel development, the gradient of the chemotactic stimulus, the milieu and the matrix on which these vessels will ideally develop are all questions that remain to be systematically investigated. The attractive aspect of cell-based therapy is that these cells stimulate proangiogenic cytokines such as VEGF, angiopoeitin and PDGF-BB, and as such represent the armamentarium that can be deployed effectively depending on the milieu in which the cells are introduced, without the clinician having to tailor every minute aspect of the therapeutic arena. What the specific stimulus is that drives the differentiation and how it can be optimized remains to be discovered.

In vascular surgery, an important precept is that to heal tissue loss, an adequate perfusion pressure must be present in the foot and no matter how rich the collaterals, if that perfusion pressure, obviously affected by the flow, is not reached, then healing becomes unlikely and leads to limb loss. In the study cited above, the ERC fraction was injected immediately after the ligation of the femoral artery and its branches [8]. It is clear that the ERCs successfully and impressively augmented the angiogenic response immediately after ischemia was induced. However, this is not quite the same as a chronically ischemic limb that then develops toe gangrene or tissue ulceration. In this sense, this experiment more closely resembles acute limb ischemia with a wound than chronic limb ischemia with subsequent tissue loss. Whether this same impressive result would be obtained if ischemia had been induced, and at some time after ischemia had been induced the ulcer had been created rather than at the time of the onset of ischemia, is an interesting experiment that needs to be performed. The kinetics of the angiogenic response in the setting of a chronically underperfused limb with a wound may be quite different. A patient that has had an ankle–brachial index of 0.3 that develops ulceration or toe gangrene may be better off with a bypass and one that has the same ulcer with an ankle-brachial index of 0.6 may be ideal for ERC injections.

The distance between the ligated artery and the downstream wound in the mouse is on the order of 1–2 cm, at most. In the human, arterial insufficiency results in foot ulcers or toe gangrene and the distance between the normal patent axial artery and the poorly perfused tissue is considerable. The average diabetic patient with tissue loss almost always has tibial occlusive disease and a patent femoral and popliteal artery. As such, the average distance between the knee (popliteal artery) and the ischemic foot ranges from 18–21 inches. Are the kinetics of injecting ERCS or any other cellular therapy with this kind of distance to be spanned favorable for tissue repair or for tissue loss? The rate of tissue necrosis may be more rapid than the ability to develop new arterioles.

These are known unknowns and as someone once said, there are also unknown unknowns. As we move forward in the field of cell-based regenerative medicine, it will no doubt take a long time to discover the answers but if we can change the unknown unknowns to known unknowns, then we can study, design, dissect and experiment to find the answers to those unknowns.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.