Controversies in regenerative medicine: Should intervertebral disc degeneration be treated with mesenchymal stem cells?

Low back pain (LBP) can significantly reduce the quality of life of patients, and has a considerable economic and social impact worldwide. It is commonly associated with disc degeneration, even though many people with degenerate discs are asymptomatic. Degenerate disc disease (DDD), is thus a common term for intervertebral disc (IVD) degeneration associated with LBP. Degeneration is thought to lead to LBP because of nerve ingrowth into the degenerate disc, inflammation, or because degradation of extracellular matrix (ECM) alters spinal biomechanics inappropriately. Thus, while the objectives of some interventions for LBP are to control pain intensity, other interventions aim to deal with the consequences of disc degeneration through stabilizing the disc surgically, by inserting artificial discs or by repairing the disc biologically and preventing progressive IVD degeneration. Despite tremendous research efforts, treatment of LBP through the use of regenerative interventions aiming to repair the IVD is still controversial. The use of mesenchymal stem cells for IVD regeneration in a patient‐based case will be discussed by an ensemble of clinicians and researchers.

Taking account of the diverse results of preclinical and clinical trials as well as the different experiences of the authors on LBP treatment, this article is structured as "Yes" and "No" chapters, arguing in favor or against the use of mesenchymal stem cells (MSCs), respectively. In order to give a high priority to the translation of research to clinical application, this article will focus on the evaluation of MSCs. Noteworthy, other cell types like induced pluripotent stem cells (iPSC) are on the rise in preclinical trials but will not be subject of this manuscript.

| THE CASE
This is a T2-weighted magnetic resonance imaging (MRI) sagittal reconstruction ( Figure 1) of a 22-year-old female with persistent LBP for 8 months despite conservative treatment (physical and medical).
The MRI reveals IVD degeneration at the L4/5 level Pfirrmann grade III without endplate changes (MODIC). The annulus fibrosus is intact with preserved disc height of at least 75%. Segmental instability and isthmus pathology were excluded. The patient is normal weight with a BMI of 28 kg/m 2 .

| The importance of selecting a suitable MSC subpopulation for IVD regeneration
MSCs can be isolated from bone marrow or adipose tissue and are recognized as a feasible donor cell candidate for regenerative therapies because of their accessibility and proliferation characteristics. 4,5 The classical approach to identify MSCs is via a set of surface markers, for example, CD90+, CD73+, CD105+, CD14−, CD34−, CD45−, CD79−, or CD19 and HLA-DR-, according to the International Society for Cellular Therapy (in addition to plastic adherence and differentiation capability). 6 However, this approach yields a relatively heterogeneous population with progenitor cell characteristics. It recently became increasingly clear that preselection of particularly suitable MSC populations can further enhance the clinical outcome. CD271-MSCs, for example, have a higher potential for nucleus pulposus (NP)like differentiation than their CD271+ counterparts and may hence be primarily used in the future. 7 In cartilage research, a shift towards subpopulations has already taken place. Perez-Silos et al summarize the recent findings and highlight that CD73 + and CD73 + CD39 + cell subpopulation evidenced high chondrogenic potency. 8 These results may also be of relevance for IVD research. Overall, some of these developed techniques and approaches may also substantially help to promote clinical success for DDD treatment.

| Successful techniques to boost MSC survival in the harsh IVD microenvironment
There is no denial: The IVD represents a very harsh microenvironment even in its healthy state, due to low oxygen levels, high osmolarity, nutritional deficits and high mechanical loading. These conditions are further aggravated by acidity and inflammation during degeneration. [9][10][11] In fact, the detrimental effects of this harsh IVD microenvironment on MSC survival and functionality were highlighted a decade ago 12,13 and confirmed by later studies. 14 (a) Treatment of MSCs with specific induction medium is the most classical and well-established approach. Recent research highlighted that the use of a certain growth factor combination (eg, TGF-β1 plus GDF-5) is distinctively effective in promoting an NP phenotype. 17 (b) Alternative methods include adenoviral/lentiviral gene delivery, for example, modulation of Wnt11 to promote expression of SOX-9, aggrecan, and collagen type-2. 18 In the coming year, the major break-through in MSC-based regeneration of the IVD may arise from a technique that has been hailed as FIGURE 1 T2-weighted magnetic resonance imaging (MRI) sagittal reconstruction of a 22-year-old female with persistent LBP for 8 months despite conservative treatment and IVD degeneration at the L4/5 level Pfirrmann grade III without endplate changes the discovery of the century: CRISPR/Cas9 genome editing. CRISPR/-Cas9 could for example be used to protect MSCs from the inflammatory IVD microenvironment by repressing the expression of cytokine receptors. 19,20 Other possible applications of MSC-targeted genome editing with excellent potential for IVD regeneration may entail inhibition of cell senescence (eg, by inhibition of p16 by CRISPRi) as well as activation of extracellular matrix protein expression (eg, induced synthesis of aggrecan by CRISPRa). 21 The development of novel techniques, ranging from sophisticated differentiation methods to genome editing, can lead to a massive improvement in MSC functionality and survival within the harsh microenvironment of the IVD, hence obliterating many of the concerns provided by the MSC opponents. MSCs are able to promote disc regeneration in dogs with experimentally induced IVD degeneration. 32 Importantly, similar results were observed in sheep, not only after nucleotomy, 33 but also after annular injury 34,35 and annular incision, 36 highlighting the therapeutic potential of MSCs in large animal models and hence possibly also in humans.

| Which patients have been treated so far?
Six cohort studies investigated the feasibility and outcome after MSC transplantation in patients with DDD and chronic LBP (Table 1). [37][38][39][40][41][42] The mean age range in the different studies was 35 to 52 years. On radiological examination all patients revealed focal IVD degeneration with involvement of one or two segments of the lumbar spine. The Pfirrmann grading is reported for most of the studies and ranged from grade II to IV. After conservative management (physical and medical) of 3 to 6 months had failed to improve symptoms, patients were considered for MSC transplantation. Baseline pain and disability indexes revealed considerable lumbar pain with resulting disability in all patients. Only a limited number of patients also reported leg pain. Discography was performed in most patients to ascertain the symptomatic disc.

| MSC Transplantation is a safe procedure
Severe adverse events (SAEs) have not been reported in any of the six cohort studies. 38

| NO
The idea of using MSCs to treat back pain appears very attractive.
The treatment is simple, and minimally invasive, comprising of a single injection of a cell suspension into the disc. Tests on animals have been very encouraging with noticeable repair of degenerate nucleus pulposus within months after injection. Clinical tests too appear to relieve pain, and in some cases, show improvement in Pfirrmann grading of disc degeneration.
The underlying hypothesis for the use of MSCs, is that degeneration of the intervertebral disc is responsible for low back pain.

| Will the patient benefit from MSC injection?
As shown in Figure 2, with MRI imaging and use of supportive clinical tests as the only means of diagnosis, choosing the right patient is difficult but essential. As MSC therapy would only help those patients whose pain arises from disc degeneration, the origins of the symptoms like disturbance in pain mechanism, pain sensitization or non-specific FIGURE 2 Clinical pathway to support the indication for MSC injection. Low back pain must be evaluated and correlated by imaging and supporting clinical tests. The treating physicians must rule out non-specific pain sources and ensure a sufficient nutritional supply before considering MSC injection. Each treatment should be followed by outcome measurements that are attributable to the intervention psychological factors must be ruled out. Moreover, while there is some association between MRI gradings of disc degeneration and low back pain, many people with degenerate discs are asymptomatic. 3 There is no consensus about the best diagnostic method and classification system to identify ideal candidates for regenerative IVD treatment. 22,29,43 Hence whether the current case ( Figure 1) might potentially benefit, cannot readily be determined.

| Is there clinical evidence of the efficacy of MSC therapies?
For routine use of MSC therapies, the outcome must be attributable to the intervention, rather than non-specific placebo effects, which are known to also be effective. 44 The placebo effect is very strong for the treatment by means of injection into the disc. 45 Hence, the evidence from randomized controlled trials (RCTs) using appropriate controls is essential before such treatments are routinely adopted. Eight clinical trials using MSCs to treat low back pain have been set up. 46 None have reported any results so far.
Of the relatively sparse number of clinical studies using MSCs injection into the IVD to treat DDD, only one study finds that precursor cells, like MSCs are of no benefit in the treatment of IVD, 47 whereas several other studies suggest MSC injection is a rational treatment option. [37][38][39][40][41][42] However, none of these studies have appropriate controls. Therefore, the placebo effect cannot be ruled out. In

| Can MSCs remain viable post implantation?
For MSCs to be able to effectively perform their proposed roles of stimulating matrix production and inhibiting inflammation, they have to remain alive in the environment found in degenerated discs. They thus require a sufficient supply of nutrients, particularly glucose as, like native disc cells, MSCs are primarily glycolytic and die rapidly if glucose (but not oxygen) is removed. 48 Efficient removal of lactic acid, the main metabolic product of glycolysis, is also essential as the acidic pH found in degenerated discs is detrimental to cellular activity and survival of both MSCs and disc cells. 12 The disc is large and avascular and nutrients, necessary for survival of cells in the disc, are supplied by blood vessels at its margins.
Concentrations of nutrients throughout the disc depend on the balance between supply of nutrients and cellular demand, so nutrient concentrations across the nutrient fall with distance from the blood supply. Hence, discs can only support a limited number of cells before nutrient concentrations in the disc center become too low to maintain cell viability. The cell density supported thus varies inversely with disc height 46 with the cell density in rabbit discs for instance, being 30 to 40 times greater than that in adult human lumbar discs (cell density around 2-3 × 10 6 cells/mL in a normal human disc).
Nutrient supply is compromised in degenerated discs 49

| Can surviving MSCs in the disc function adequately?
The constituent glycosaminoglycans (GAGs) of aggrecan are responsible for maintaining disc hydration and disc height. 51

| Why do tests in animals look so promising?
Animal experiments are usually carried out on young (even immature) animals. Their discs are much smaller than human discs and hence have a much higher cell density. 46 Moreover, animal models of degeneration are usually acute, so the microenvironment is less hostile, with no reports of impairment of the nutrient pathway. In some species, the presence of notochordal cells (NC) in the NP is known to protect the IVD from the development of DDD, preserving structural integrity and biomechanical properties. 55 In addition clinical studies suggesting MSC injections as a possible treatment for LBP, all lack high patient numbers and long-term results.
As not all of them have controls, they are not able to rule out that the positive results reported could arise from a placebo effect. Besides, the level of evidence of all existing studies is low, as there are no results from multicenter, prospective, randomized trials. Ultimately, identification of the right patient for MSC treatment will remain a challenging task.
Hence the treatment of the illustrated case through injection of MSCs cannot thus be recommended. MSC implantation is very unlikely to effect a clinically relevant repair and even if it could, it is unclear whether the patient would benefit from repair of her disc.

CONFLICTS OF INTERESTS
The authors have no conflicts of interest to report.