Advancing cell therapies for intervertebral disc regeneration from the lab to the clinic: Recommendations of the ORS spine section

Intervertebral disc degeneration is strongly associated with chronic low back pain, a leading cause of disability worldwide. Current back pain treatment approaches (both surgical and conservative) are limited to addressing symptoms, not necessarily the root cause. Not surprisingly therefore, long‐term efficacy of most approaches is poor. Cell‐based disc regeneration strategies have shown promise in preclinical studies, and represent a relatively low‐risk, low‐cost, and durable therapeutic approach suitable for a potentially large patient population, thus making them attractive from both clinical and commercial standpoints. Despite such promise, no such therapies have been broadly adopted clinically. In this perspective we highlight primary obstacles and provide recommendations to help accelerate successful clinical translation of cell‐based disc regeneration therapies. The key areas addressed include: (a) Optimizing cell sources and delivery techniques; (b) Minimizing potential risks to patients; (c) Selecting physiologically and clinically relevant efficacy metrics; (d) Maximizing commercial potential; and (e) Recognizing the importance of multidisciplinary collaborations and engaging with clinicians from inception through to clinical trials.

a progressive, cell-mediated cascade of molecular, structural and biomechanical changes, which is strongly implicated as a cause of "discogenic" back pain. 4 The incidence of disc degeneration is linked to aging, trauma, genetics, lifestyle and the presence of comorbidities. [5][6][7][8][9] The intervertebral discs are the pliant, fibrocartilaginous joints required for transferring compressive loads and supporting complex mobility of the spine. 10 Each disc is comprised of a central, proteoglycan-rich, gelatinous nucleus pulposus (NP), a peripheral, fibrocartilaginous annulus fibrosus (AF), and superiorly and inferior, hyaline cartilage end plates that interface with the vertebral bodies. 10 The earliest degenerative changes typically manifest in the NP, where reductions in proteoglycan content and hydration compromise biomechanical function, leading to progressive degeneration of the entire intervertebral joint. 11,12 Self-repair and regeneration of the NP is limited by a low cell density and a limited nutrient supply. 13,14 Conventional treatments for disc degeneration are focused solely on alleviation of symptoms and often have limited long-term efficacy. [15][16][17] This is exacerbated by the lack of an accepted clinical standard for discogenic pain, where clinicians are often unable to identify a specific nociceptive cause. 18,19 Because of this, the rationale for choosing surgery for these patients is controversial. Spinal fusion does not restore natural biomechanical function and has been shown to induce adjacent segment pathology (ASP) caused by increased mechanical stress in adjacent segments. 20 More specifically, in a study comparing 725 lumbar fusion cases to 725 randomly selected controls with chronic low back pain diagnoses, focusing on the outcome return to work (RTW) revealed only 26% of patients had RTW 2 years after fusion surgery, while 67% of nonsurgical controls had RTW within 2 years from the date of injury. 21  There is therefore a tremendous unmet need for new treatment options for patients with disc degeneration and associated lower back pain. To address this need, there has been significant recent interest in developing injectable cell-based therapies for the treatment of disc degeneration with the specific aim to stimulate tissue repair. 22 Such therapies represent a potentially low risk and low cost long term solution for a very large patient population, making them attractive from both clinical and commercial standpoints.
The pathway to bringing new cell therapies to patients involves extensive in vitro testing, pre-clinical validation in small and large animal models, and human clinical trials to demonstrate safety, efficacy, and commercial viability. Successful translation involves multidisciplinary engagement and interactions between clinicians, scientists, engineers and industry. Clinicians are particularly important as they can provide access to primary human cells, advise on patient selection criteria and outcome metrics, assist with the design and conduct of clinical trials, support risk/value discussions with regulatory and reimbursement agencies, and advocate the treatment to patients and health systems.
Even after demonstrating success in animal studies, cell-based therapies are at risk of stalling in the "valley of death", which comprises multiple significant barriers that include scale-up, manufacturing, regulatory, business and market obstacles. To effectively move cell therapies out of the lab and into the clinic, the following must be

| CURRENT COMMERCIAL APPROACHES FOR CELL-BASED DISC REPAIR AND REGENERATION
A small number of companies have successfully entered the clinic to evaluate the efficacy of cell-based disc regeneration treatments in humans. These treatments have employed a variety of cell types including those obtained from both homologous (from the disc) and non-homologous (eg, from knee cartilage, adipose tissue) sources, and have utilized both autologous and allogeneic approaches. Autologous Disc Cell Transplantation (ADCT) by Co.Don AG (Berlin, Germany) has been in clinical use for the treatment of degenerate discs for a number of years, 23 with recent investigations demonstrating that reimplantation of culture-expanded NP cells can retard degenerative changes in patients with herniated discs 24 ; however, there are limitations associated with ADCT delivery and efficacy, including cell leakage following injection into the disc, 25  Australia) also showed promising results using immunoselected adipose-derived mesenchymal stem cells (MSCs) injected into the discs of patients, including finding that the treatment resulted in pain relief and improved water content in treated discs compared to controls. 28,29 When injected into degenerate ovine discs, these cells were shown to elicit regenerative effects. 30 : EVEN THE BEST  THERAPY WON'T SUCCEED IF APPLIED TO  THE WRONG DISC For effective clinical translation it will be important to stratify patient cohorts, for example using sophisticated magnetic resonance imaging techniques [33][34][35][36][37] and quantification of circulating biomarkers 38 to select suitable candidates for treatment. The opportunity for cellbased therapy may be most appropriate at the early stage of the degenerative cascade, prior to the onset of advanced disc degeneration coupled with endplate calcification and extensive annular degeneration. 39 As a first step, patient selection for cell-based therapies should therefore be limited to single-level, moderate severity disc degeneration, which is reflected by Pfirrmann Grade III or IV on MRI.

| PATIENT SELECTION
Patients presenting with a Pfirrmann grade of II or V may represent a patient population either too mild or too advanced in the disease process to show measurable improvement. 40 Currently, patient selection for intradiscal therapies includes chronic back pain (>4 VAS) and disability (>40 ODI) that is refractory to conservative care (longer than 6 months  41 Recently, a group of international experts concluded that there is no accepted clinical standard for discogenic pain. 18,19,42,43 Equally important, treatment outcomes are currently judged by subjective and qualitative patient reports, which can be influenced by significant placebo effects. 44 Therefore, to support clinical application and evaluation of intradiscal cellular therapies, there is a critical need for new techniques that localize pain generators and quantify therapeutic effects on treated levels. New, advanced imaging tools for this purpose are currently in development and have been summarized in several recent reviews. 45,46 Examples include MRI techniques such as T1ρ and T2 relaxation times, and chemical exchange saturation transfer (CEST), which, unlike more widely used subjective grading disc grading schemes, are quantitative surrogates for disc composition.

| The challenges of the degenerate disc microenvironment
The rationale for and benefits of delivery of cells to the NP are 2-fold: firstly, transplanted cells may stimulate endogenous NP cells to produce neo-matrix through paracrine effects; and secondly transplanted cells may adopt an NP cell-like phenotype and directly reconstitute native tissue. 47,48 A central challenge, however, that limits the potential of cell-based regeneration is the harsh local cellular microenvironment within the degenerate disc, which is characterized by low oxygen and nutrient supply, increased acidity, altered osmolarity, as well as elevated levels of pro-inflammatory cytokines. [49][50][51][52][53][54] Therefore, there is a clear need to identify robust cell populations to enhance the likelihood of survival post-injection, and characterize how such cells will function in the typical degenerate microenvironment to determine if they can contribute effectively to functional disc repair. The precise cell number for intradiscal delivery needs to be carefully considered, and to-date the exact number of cells required for functional regeneration is unclear. This will most likely depend on the cell-type being employed as well as the metabolic activity and percentage survival post intradiscal delivery. A recent review of clinical trials by Sakai and Schol reported cell numbers ranging from 1 to 40 million cells being used in investigations. 55 It is postulated that the delivery of large numbers of cells may exacerbate the degenerative microenvironment due to competing nutrient demands, which may in turn result in ineffective clinical outcomes. 56 Given the varying stages of degeneration and microenvironments that exist in vivo, it will be important to tailor or design treatments for individuals. One way to achieve this could be to condition stem cells prior to implantation by acclimatizing them to patient specific in vivo microenvironmental milieu, which may include inflammatory cytokines, acidity, oxygen or nutrient deprivation. [57][58][59] These treatments must be assessed using appropriate in vitro and ex vivo culture conditions which mimic those of the degenerate niche to assess the likely behavior and survival of transplanted cells prior to progressing to in vivo testing. 49

| Primary cells, stem cells or gene therapy?
From a surgical and practical perspective, given the unique structure and location of the disc, the difficulty in obtaining autologous primary NP tissue and cells for therapeutic use from either herniated discs or adjacent intact levels is clear. Delivery of primary NP cells to the disc appears safe and has shown some potential for efficacy in early clinical trials. 60 Additional limitations of primary cells include the diminished tissue forming capacity of culture expanded NP cells derived from degenerated tissue 26 and the altered catabolic phenotype of such cells, together with the paucity of NP cells that can be isolated.
Furthermore, the isolation of tissue from adjacent healthy disc levels may increase the risk of initiating degeneration at the harvest site. 61 This has motivated many researchers into identifying and characterizing alternative cell sources for disc regeneration, which is a key step towards translating therapies into clinical practice. 62 Primary, differentiated cells from other skeletal sites with reduced risk for donor site morbidity, such as articular and nasal cartilage, have been investigated in vitro and in animal models for NP regeneration. [63][64][65] While these cell types do exhibit some phenotypic differences to native NP cells, their relative ease of isolation, differentiated state and higher propensity to deposit extracellular matrix may make them attractive alternatives in the future and warrant further consideration and exploration.
The likelihood that these alternative cell types can adopt a true discogenic phenotype is remote, and whether the extracellular matrix constituents they deposit are suitable to provide biomechanical functionality and longevity to restore disc function remains to be fully established.
Stem cell therapies have received considerable attention due to their versatility, translatability and potential for long term native tissue regeneration. Some successful patient outcomes have been demonstrated in clinical studies using this approach. 28,29,55,[66][67][68] The bone marrow has been the prime site of MSC isolation for disc regeneration applications, inclusive of in vitro, preclinical and clinical studies. 61,69 Progenitor cells derived from other tissue sources such as adipose tissue have also been shown to possess significant potential for differentiation and tissue forming capabilities. 70,71 Adipose derived stem cells (ADSCs) may provide a better alternative and candidate for cell therapy and disc regeneration, due to their abundance and ease of isolation, and may also have a lower inherent capacity than bone marrow derived MSCs to undergo endochondral ossification. 72 Harvesting of autologous ADSCs can be performed in outpatient clinics with typical yields of up to 25 000 adherent ADSCs per gram of tissue. 73 One of the potential challenges in utilizing stem cells is that they are undifferentiated. The precise mechanism of action for regeneration is unknown; whether injected stem cells have the capacity to differentiate towards a discogenic phenotype or stimulate resident native cells has not been clearly demonstrated in preclinical studies. Priming stem cells using growth factors or other environmental cues such as altered oxygen or glucose availability to direct or maintain a specific phenotype may enhance their regenerative potential, 58,59,74-76 but may also add complexity to the regulatory approval process.
Notochordal cell-based approaches for disc regeneration have received considerable interest in the past several years. Notochordal cells are the progenitors of adult NP cells, and their loss in humans during postnatal growth is thought to contribute to the onset of degeneration later in life. 10 As notochordal cells themselves cannot be readily obtained for direct clinical application, most studies have focused on leveraging the therapeutic potential of notochordal cellsecreted factors. 77 Gene therapy approaches may hold significant promise for disc regeneration, for example through silencing of catabolic or activating anabolic pathways. Gene therapy has advantages over direct delivery of proteins or small molecules, among them the possibility of sustained efficacy through long-term, regulated, endogenous synthesis of growth factors or anti-inflammatory factors. 78 However, the need to establish a safety profile and possible off target effects will most likely lead to a longer and more complex regulatory process before realizing a commercially available product and clinical translation.

| Effective cell delivery and retention
With respect to delivery of cell-based therapies, the primary mode of choice has been injection through the AF. A key translational advantage of this approach is that, with image-guided assistance, it can be performed percutaneously. However, there are associated challenges that have the potential to compromise both safety and efficacy, including the potential for needles to induce permanent AF damage, 79 cell leakage from the delivery site and diminished cell viability due to shear forces as cells are forced through small diameter needles. [80][81][82] For example, injecting a cell suspension into the lumbar discs of rabbits resulted in a 90% loss of the injected cells within the first 30 minutes. 83 Biomaterials as delivery vehicles, and in particular hydrogels, have been utilized to overcome retention issues and provide additional support for cell survival and phenotype retention. [84][85][86][87][88][89] Injectable biomaterial systems utilizing alginates, collagens, hyaluronic acid, fibrin and a variety of other substances have all shown promise for improving cell delivery for disc repair [90][91][92] ; however, for clinical translation, the biomaterial itself will require regulatory approval.
Using biomaterials already approved and in clinical use for other indications may accelerate the approval process. That being said, because the disc is avascular, carrier degradation products may accumulate to levels beyond those found in other target tissues, justifying discspecific toxicity testing. Selection of needle size for trans-annular delivery should be a careful trade-off between the minimum size necessary to prevent shear-induced cell death and allow injection of viscous hydrogels, and the maximum size necessary to prevent permanent AF damage and/or cell leakage. In a recent study using a preclinical large frame goat model, a 22G needle was found not to induce discernable degenerative changes on MRI or histology after 12 weeks. 93 In addition, the volume to be delivered needs to be optimized, as excessive volumes may result in over pressurization, increasing the likelihood of cell leakage upon needle withdrawal. Further, the spatial distribution of delivered cell populations may also impact cell differentiation or activity and subsequent regeneration outcomes, as injected materials will tend to preferentially concentrate within preexisting fissures that are common in degenerated discs. For hydrogeldelivered cell populations, migration of cells through the NP and integration of the injected material with native tissue should be considered. An alternative delivery route that has been investigated, and which does not cause disruption to the AF, is through the pedicles/ endplates (transpedicular approach); 94,95 however, whether such an approach will have detrimental effects to the integrity of the endplate in the long term is yet to be established.

| MINIMIZING POTENTIAL RISKS TO PATIENTS
Demonstrating safety is critical for disc therapies, as the consequences of an adverse event are potentially severe. The disc is one of the most highly loaded tissues in the body, which creates the significant potential that injected biomaterials will migrate in response to long term, cyclic loading. Material that is expelled from the disc space may impinge on nerves and result in neurological complications or even paralysis. Therefore, a fundamental safety requirement of a disc therapy, particularly one that involves a structural component such as a hydrogel or scaffold, is that it structurally integrates, and remains completely and permanently contained within the disc space upon delivery. Retention should be verified in the presence of physiological loading; ex vivo, this can be accomplished through extensive cyclic and multiaxial loading. Long term in vivo retention can be assessed via incorporation of radiopaque contrast agents or nanoparticles at the time of delivery, facilitating non-invasive and three dimensional longitudinal assessments. 85 Ensuring that a therapy does not impart local or systemic toxicity is also a critical safety consideration, and in vivo biocompatibility studies are generally a requirement of the FDA (food and drug administration) and other regulatory bodies for new biological therapies. Toxicity could potentially result directly from the implanted material, or over time from its degradation by-products. For example, biomaterials that break down into acidic by-products in the confined space of the disc could contribute to eliciting a catabolic, local, cellular response. 96  Emerging non-integrating techniques for cellular reprogramming, for example using synthetic mRNA or chemical methods, may mitigate these risks without compromising induction efficiency. 98,99 Adult, multipotent stem cells, while not exhibiting the same tumorigenic risk as pluripotent stem cells, may also exhibit genetic instability through chromosomal aberrations that manifest progressively with increasing numbers of expansion passages. 97 Minimizing the length of culture and degree of cellular manipulation prior to implantation minimizes the propensity for these in vitro genetic alterations to occur. Assessing the heterogeneity of stem cell populations in culture is also important to reduce tumorigenic and immunogenic risk.
Gene therapy, which employs cellular reprogramming, poses unique safety challenges that may limit or complicate translatability. 100 These risks, which include the potential for insertional mutagenesis and cancer, and antigenicity with associated adverse immunological responses, result in additional regulatory hurdles, which may be harder to overcome for therapies targeting non-lifethreatening conditions such as disc degeneration. Risks associated with gene therapy may be mitigated through careful selection of gene delivery systems (eg, viral vs non-viral), which minimize risk without comprising efficacy. 101 The method and frequency of administration will also impact a therapy's safety profile. Therapies that can be delivered using minimally invasive procedures, such as image-guided percutaneous injection, and do not require general anesthesia, will pose less risk to for painful disc degeneration is long term (>6 months) alleviation of symptoms. Current treatment strategies, whether they be conservative or surgical, may be effective in providing short term relief from pain, but their long-term efficacy is more problematic, in part because they do not seek to restore disc structure or mechanical function.
Given the nature of disc degeneration and its cascading effect on adjacent tissues and overall spine mechanical properties, it is often difficult to specify the fundamental treatment target. Disruption of normal disc biomechanical properties is associated with accelerated degeneration of other structures such as facet joints, and osteophyte formation due to adaptive bony changes or calcification of ligaments. 102,103 Back pain patients commonly have associated leg pain and symptoms which are frequently secondary to nerve compression, due to stenosis caused by loss of disc height and/or bulging or herniation. Consequently, stabilization (if not restoration) of disc structure is a key goal.
Back pain can also arise from irritation of peripheral disc tissues.
For example, pro-inflammatory crosstalk between the disc and vertebra is thought to underlie bone marrow edema evidenced as Modic changes (MC) on clinical MRI. 104  For cell therapies to be clinically adopted, they need to demonstrate benefits beyond current treatment options. The rationale for choosing between surgical and nonsurgical care for chronic low back pain patients is not well-defined. While the benefit of surgery for "mechanical" back pain (such as instability and sciatica) is supported by significant outcome literature, the appropriate intervention for disc-related pain is less clear. Fusion surgery may alleviate pain by restoring intervertebral height and decompressing nerves, but immobilizes the intervertebral joint. 15,17 Metal/polymer-based total disc arthroplasties preserve a limited degree of joint mobility, but are subject to wear and potential failure requiring revision surgery, and have not seen widespread adoption. 108 Reported success of surgical care for back pain ranges from 41% to 57%, 109 and with 5% to 16% early complication and reoperation rates. 110 Discectomy may alleviate symptoms by removing herniated or bulging disc material and decompressing nerves, but the damaged disc is not repaired and may continue to degenerate, and many patients continue to experience pain and require ongoing treatment. 111 Non-surgical approaches such as steroid injections and physical therapy are temporary treatments that seek to mask symptoms that might otherwise resolve over time.
Recurrent treatment for unresolving symptoms is more problematic as permanent nerve damage can occur. Non-surgical, like surgical, treatment does not alter the long term progression of degeneration. Restoration of disc biomechanical function is therefore key to effective long-term alleviation of painful symptoms.
The primary attraction of biological, cell-based disc regeneration therapies is that they have the potential to alleviate symptoms and stabilize disc structure as well as biomechanical function by reconsti-  and biomechanical-to therapeutic cell survival and function. With respect to the biochemical environment, as the disc has little or no direct blood supply, the cells must survive and function with access to very limited oxygen and nutrition. As discs degenerate, even in the early stages, this biochemical environment is further characterized by increasing local catabolic cytokine expression 112 and acidity. 113 Therapeutic cell types such as MSCs are more sensitive to microenvironmental stressors such as oxygen, nutrient deprivation and inflammatory stimuli compared to endogenous cells. [114][115][116] For in vitro cell culture models, mimicking this in vivo disc microenvironment can be accomplished by culturing cells in low oxygen, and reduced glucose and serum, to simulate nutrient deprivation. [115][116][117] The degenerate environment can be further simulated by including catabolic mediators such as inflammatory cytokines and by increasing acidity. [118][119][120] The physical environment can be mimicked through culturing in soft 3D scaffolds such as hydrogels. Mechanical stress can be simulated using bioreactors that apply dynamic compressive loads and/or hydrostatic pressure. [121][122][123][124] Organ culture systems that employ living, degenerate discs from human donors may be the ultimate platform to effectively mimic the biological complexity of the in vivo cellular microenvironment, particularly when combined with a bioreactor that applies physiological loading. 125 Ensuring accurate recapitulation of the human disc cellular microenvironment is equally important when moving to in vivo systems.
One of the greatest challenges facing in vivo models of disc therapeutics is the need to replicate the size, and in particular the height of the human disc, necessary to accurately recapitulate physiological nutritional demands. Large animals obviously more closely approximate the nutritional environment compared to rodents, however, even in livestock, lumbar disc size is still significantly smaller than in humans (eg, lumbar disc height in sheep is~4 mm compared to~11 mm in humans). 126 Therapeutic cell types such as MSCs may exhibit different phenotypic characteristics as a function of donor, age, species and anatomical site, and these characteristics may be closely linked to regenerative potential. [127][128][129][130] While the development phase may necessitate the use of non-human cells that are accessible and compatible with in vitro and in vivo models, the results obtained with FIGURE 1 Requisite steps for demonstrating the efficacy of cell-based disc regeneration therapies. Model systems should be applied iteratively, balancing experimental control, cost and throughput in the early stages with biological complexity and clinical relevance at more advanced stages, in order to maximize the chances of success in the clinic these cells may not accurately reflect the performance of human cells that will be used clinically. It is therefore important to validate findings using cells from human donors, wherever possible controlling for variables such as age, sex, co-morbidities and anatomical site. Human cells may be obtained both from commercial sources, as well as from clinical collaborators undertaking spinal procedures. The latter of these sources is particularly attractive where it permits recruitment of donors who would also be prospective recipients of disc therapeutics.
For example, a surgeon performing a procedure such as a spinal fusion may be able to provide not only surgical waste tissue as a source of endogenous disc cells, but also bone marrow from the adjacent vertebrae without subjecting the patient to any additional surgical procedures or risks. As already discussed, organ culture using degenerate human discs obtained from cadavers represents an avenue for evaluating therapeutic efficacy not only using human cells, but also in a physiologically accurate and degenerative microenvironment. The importance of cell source also extends to the preclinical development phase. For example, a therapy that demonstrates efficacy in vitro using cells of a particular age and species, may not demonstrate similar efficacy in a large animal model of a different age and species. In vitro studies should therefore seek to validate a therapeutic approach using cells from the full range of sources that will be required at each phase of development.

| Pain and disability outcome measures
As alleviation of pain is the primary goal of disc therapies, including pain as an outcome measure in laboratory experiments is desirable.
Selection of physiologically-relevant pain outcome measures in the preclinical setting is complicated, as in the clinical setting pain is subjective and its direct source may be ill-defined. 131,132 Pain outcome measures can be classified in different ways. They may include surrogates such as nerve ingrowth and regression, 133,134 local and systemic levels of secreted neurogenic and inflammatory factors, 135,136 as well as imaging to demonstrate nerve compression. 45 Direct outcome measures of pain may include assessments of overall behavior and mobility, and pain sensitivity assays. 137 Small animal models are particularly attractive with respect to assessing pain, as they are compatible with most if not all of these outcome measure types. For example, if a therapy is being evaluated in a mouse or rat, it is possible and practical to assess mobility, pain sensitivity, and serum and radiological biomarkers in vivo, and tissue level expression of neurogenic markers post mortem. Using in vitro models, pain outcomes are limited to surrogates such as expression of neurogenic factors and nerve ingrowth. Large animal models are in theory compatible with many of the same outcome measures as small animal models, but practical considerations make implementation of functional and pain sensitivity assays more challenging.

| Structural and biomechanical outcome measures
Including appropriate outcome measures to assess restoration of disc structure and mechanical function is equally important, as these are the key indicators of the likelihood for a therapy to provide long term alleviation of symptoms. Like pain, the ability to assess these parameters directly or indirectly is dependent on the type of model system.
With respect to structure, the most clinically relevant outcome measure is stabilization or restoration of disc height as demonstrated through imaging (MRI, plain radiographs or computed tomography), and is possible for in vivo models as well as in vitro whole-disc organ culture models. 93,138,139 Importantly, these non-invasive imaging metrics can be applied longitudinally and thus used to confirm both acute regenerative effects and sustained, long term efficacy. With respect to biomechanical properties, in vitro model systems that incorporate three-dimensional constructs or whole-disc organ culture facilitate direct assessment of mechanical function. 89,140 For in vivo models, mechanical properties can be assessed ex vivo on intact spinal motion segments or isolated tissue samples. 85 In the case of motion segments, multi-axis loads can be applied to similar physiological deformations, including those likely to lead to failure. 141  Other factors should be considered and continuously reviewed as the concept evolves, including complexity, cost of components and preparation time, in addition to the ease and method of administration.
Early and frequent discussions with regulators (eg, FDA) are key to defining the studies required to translate such products into the clinic, and to ensure that manufacturing approaches will meet the standards required for human testing.
One predictor of the potential commercial viability of a new therapy is the size of the prospective target patient population. Further, the overall manufacturing costs associated with producing the product, in contrast to the reimbursement/pricing potential of the treatment, must be favorable. In considering the clinical motivation for developing a specific biological disc therapy, the overall socioeconomic burden is often cited, as is the current lack of effective treatment options. In establishing and justifying the clinical need for a specific, new therapeutic approach, it is necessary to move beyond general patient populations (ie, "degenerative disc disease") and stratify subgroups of patients that will be suitable candidates for that therapeutic approach based on the specific disease features experienced. For example, injectable disc therapies may be tailored to target a patient subgroup that exhibits a specific set of symptoms or a narrow window of degeneration severity that includes an intact annulus and relatively healthy endplates. The need for a therapy targeting symptomatic low back pain in patients with mild to moderate severity disc degeneration (eg, minimally invasive cell therapy) could be justified by citing the number of patients in this subgroup that are not candidates for surgery and are unresponsive to conservative treatment approaches. Alternatively, the need for a new therapy targeting patients with severe disc degeneration (eg, tissue engineered total disc replacements) could be justified by citing the number of patients that exhibit poor outcomes from current surgical approaches such as fusion or total disc arthroplasty.
The second important consideration to ensure future commercial viability is early protection of relevant intellectual property. At most academic institutions, protection of intellectual property begins by submitting an invention disclosure to the relevant university office, which will then assess the concept or technology for commercialization potential. Such disclosures are a requirement for any potentially patentable invention, and institutions may stipulate that this should be done in advance of any public disclosure (eg, publication of manuscripts or conference presentations). Without such protection before public disclosure, the ownership is lost, and no patent may be filed, significantly decreasing the chances of the product ever being translated to the clinic. Following disclosure, where an invention is identified as having commercialization potential, the institution should be equipped to guide the investigator through the intellectual property protection process, including filing of patents. Trade secrets represent an alternative approach to patenting an idea, though not one necessarily supported by academic institutions. When a patent is published and becomes public, smart competitors have the opportunity to circumvent it by implementing minor alterations or improvements to the overall concept. In the case of trade secrets, key properties of the product that are critical to its function never enter the public domain.
The relative cost and complexity of a therapeutic strategy, with respect to production, preparation and administration will also impact commercial viability. A complex and costly in vitro approach may be necessary during early stages of development; however, investigators should consider how this can be reduced as the development process advances towards preclinical and clinical translation. This process of refinement might include minimizing (or eliminating) the number of growth factors, cell types or biomaterial components, substituting growth factors with cheaper compounds, and reducing the time needed for in vitro manipulation prior to administration, which may include cell expansion or conditioning.
For first or second generation therapeutic products, investigators should also be cognizant and take steps at an early stage to avoid developing "complex" approaches, which will most likely face significant regulatory hurdles and barriers. For example, this might include considering using autologous vs allogeneic cells, or using appropriately conditioned and sourced adult stem cells vs pluripotent cells that require viral reprogramming. Therapies that require in vitro manipulation using animal serum face tough regulatory hurdles and alternative culture conditions should be considered. If the approach includes gene therapy, then gene delivery systems with the best safety profile should be selected. The number of regulatory hurdles could be further reduced by incorporating or substituting compounds that have an established safety record and are already FDA approved for other indications.
The final important consideration with respect to ensuring commercial viability is to undertake a comprehensive survey of the current competitive landscape. This requires identifying competing therapeutics, including those in clinical use and those undergoing clinical or preclinical trials, that target a similar cohort of patients, and reviewing where those competing products are with respect to the development pipeline. If a competing product has not already come to market, it should be estimated when that is likely to happen. Even if the number of competing products is relatively small, factors such as relative risk, cost and potential efficacy compared to the therapy being proposed should be assessed.

| THE IMPORTANCE OF MULTIDISCIPLINARY COLLABORATION AND ENGAGING WITH CLINICIANS
Closely linked to commercial viability and prospective market uptake is the willingness of clinicians to embrace the newly developed therapeutic approach. There may be resistance within a particular market sector to switch away from traditional treatment approaches, for example, for philosophical, financial or logistical reasons. To ensure a new disc therapy will successfully and effectively translate from the lab to the clinic, basic scientists must therefore work to ensure that the new product will ultimately be enthusiastically embraced by those clinicians required to administer it. To achieve this, it is essential to engage clinicians early and continuously from inception through to translation, and these clinical collaborators may well already be passionate about identifying novel treatment strategies for their low back pain patients. Not only will the participation of clinicians (as the end users) aid market acceptance, it will also add significant value to the overall scientific endeavor. In the early stages, clinicians will be able to draw on their extensive, firsthand experience to help identify prospec- The goal now must be to more effectively translate laboratory findings into the clinic to achieve improved patient outcomes. To this end, the primary intention of this perspective from members of the Orthopedic Research Society Spine Section is to provide a consensus with respect to the central challenges that have limited effective clinical translation of cell-based therapies. Motivating this perspective was a uniform and sincere belief among Spine Section members that cell-based therapies, if effectively designed and implemented in conjunction with appropriate diagnostic tools, patient selection criteria, and with the support of industry partners have the potential for substantial clinical impact.
Recommendations for researchers to effectively address these challenges are provided, falling broadly under the key themes of "Safety, Efficacy, Commercial Viability and Engaging Clinicians" (summarized in Table 1). The authors urge researchers to consider and leverage these recommendations to enhance the translational relevance and clinical potential of their research endeavors and activities.