Chapter Seven - From Bench to Beside to Cure Spinal Cord Injury: Lost in Translation?

https://doi.org/10.1016/B978-0-12-407178-0.00008-9Get rights and content

Abstract

Despite intense efforts to overcome the inhospitable milieu for axonal regeneration within the damaged spinal cord an evidence-based repair strategy promoting relevant functional improvement is still not available for spinal cord injured individuals. Nevertheless, several preclinical axonal regenerative strategies were developed all the way to phase I/II clinical trials, which have recently been terminated. The aim of this chapter is to critically review translated preclinical treatment strategies with respect to their conformity with previously published guidelines and requirements for preclinical studies leading to clinical trials in human subjects with spinal cord injury (SCI). Cell-based strategies (macrophage and embryonic stem cell grafting) and the administration of C3 transferase inhibitors and anti-Nogo-A antibodies were investigated. Overall, these four approaches comply with preclinical quality standards to varying degree. For future preclinical analyses, several additive components such as defined criteria for robustness of observed effects, a clear confirmation of underlying structural mechanisms, and the implementation of appropriate preclinical rehab approaches should be considered in order to increase the quality and consequently the likelihood of respective therapeutic strategies to succeed in human individuals suffering from SCI.

Introduction

Despite preclinical findings—the first promising study was published more than 30 years ago (Richardson, McGuinness, & Aguayo, 1980)—clearly demonstrating the capacity of the injured mammalian spinal cord to regenerate, not a single experimental treatment translated into the clinical setting. A recent review pointed this out in a rather provocative title: “Central nervous system (CNS) regeneration does not occur” (Illis, 2012). Illis states “the long history of research in regeneration in the CNS centered on the site of the lesion has proved to be sterile. After a century of such research the focus should move away from the site of the lesion to the intact CNS where there is real promise of improvement of function.” Of course, it is considered easier to achieve structural rearrangements rostral and caudal to the SCI site in order to compensate for lost function after irreversible transection of long descending and ascending fiber tracts. However, such a strategy will be helpful only in incomplete spinal cord injury (SCI), where a relevant proportion of spared axons still cross the lesion site. However, the vast majority of severely injured sensory–motor complete patients will not gain clinically meaningful functional benefits from such neural rearrangements beyond the lesion site. For these patients, separate from efforts to compensate lost function with neuroprosthetic devices (Rupp & Gerner, 2007), the sole neurobiological option is to reconnect the injured spinal cord appropriately.

Although results in terms of therapeutic efficacy are discouraging, we can learn tremendously from existing preclinical and clinical research in order to implement this knowledge into new intelligent approaches. In this review, four different bench to bedside treatment strategies will be critically appraised (Table 7.1): (1) cell transplantation approaches to modulate inflammatory responses after SCI (stimulated macrophages), (2) replacement of oligodendroglia in order to remyelinate axons in the injured spinal cord (human embryonic stem cell, hESC-derived oligodendroglial precursor cells). Further, approaches investigating molecular compounds like: (3) inhibition of the Rho pathway with BA-210 and (4) neutralization of Nogo-A with specific antibodies, which aim to interfere with axon regrowth. These four strategies will be reflected in the context of reported preclinical evidence that triggered further development into clinical trials. The detailed analysis of respective clinical trials will not be within the scope of this review.

One hypothesis explaining the poor regenerative capacity of the injured CNS is based on the differential immune response in the injured peripheral versus the CNS. Macrophages in the peripheral nervous system are considered as key players to remove debris after injury and secrete appropriate growth factors to pave the road for successful axon regrowth. This aspect is missing in the immune-privileged CNS, thus contributing to poor intrinsic tissue repair. The group around Michal Schwartz demonstrated that the transfer of macrophages incubated with pieces of peripheral nerve prior to transplantation promotes limited repair of injured optical nerve (Lazarov-Spiegler, Solomon, Zeev-Brann, et al., 1996).

This approach was transferred to the completely transected rat spinal cord (T8/9 level) (Rapalino et al., 1998). Animals received homologous blood-derived macrophage grafts (after prior incubation with peripheral nerve) at and caudal to the lesion site. Animals were assessed up to 19 weeks post injury. Subgroups were generated based on the arbitrary assumption that BBB scores (locomotor rating scale for open field testing in rats from 0 to 21; Basso, Beattie, & Bresnahan, 1995) up to 5 were not considered as recovery but rather local reflex activity. Animals receiving macrophage grafts with a BBB score > 5 reached an average score of 7.1, whereas animals from the same group with a BBB score ≤ 5 averaged at 3.4. Control animals receiving (1) medium, (2) aFGF, or (3) fibrin glue injection had average BBB scores of 1.1. Motor-evoked potential (MEP) responses recovered only in 2/12 treated animals, not in control animals. Supraspinal input rather than mechanisms of the local spinal circuitry was considered to be responsible for the observed locomotor recovery since retransection abolished the observed MEP recovery. Precise structural correlates (e.g., target reinnervation, remyelination), which could have accounted for the observed functional recovery, were not shown. The fate of grafted macrophages in terms of survival and migration behavior was not investigated. One follow-up study supported these findings demonstrating functional recovery in rat contusion SCI with grafted macrophages, which were stimulated with dermis instead of peripheral nerve (Bomstein et al., 2003). The phenotype of macrophages after coincubation with dermis was analyzed (expression of CD86, CD54, MHC class II, CD80; secretion of soluble factors), however, not compared with previously described preparations of peripheral nerve-stimulated macrophages. Macrophages were grafted between 4 and 9 days post contusion into the “caudal border of the lesion” (Bomstein et al., 2003). No information was provided as to how the caudal border was identified or defined prior to grafting.

The preclinical evidence provided the grounds for a phase I open-label nonrandomized study. The Israel-based single center study investigated eight AIS grade A patients recruited between July 2000 and February 2002 from Israel, Poland, the Netherlands, and the United States. Patients received injections of 4 million peripheral blood-derived CD14 cells into four sites at the caudal myelopathy border. Details regarding the identification of the cell injection site were not provided. Three out of eight patients converted to AIS B (motor complete, sensory incomplete) already at 1 month post injury. After 12 months, these three patients reached AIS C grade (some motor function below the neurological level). As severe adverse events, two cases of pulmonary embolism and one case of osteomyelitis were described. The described conversion rate to less severe AIS grades (37.5%) went beyond naturally occurring AIS conversion rates in AIS A patients, which range between 20% and 30% (Fawcett et al., 2007, Spiess et al., 2009).

Based on the promising conversion rate, a multicenter (five sites in the United States, one site in Israel) randomized controlled phase II study followed, which failed to show a significant difference between macrophage grafted and nonoperated patients (Lammertse, Jones, Charlifue, et al., 2012). Compared to the phase I trial (Knoller, Auerbach, Fulga, et al., 2005), the experimental design was modified in several aspects. Patients received 1.5 Mio cells into six sites-in the phase I study 4 million into four sites. In almost half of the operation procedures, cell injection at the caudal lesion border was controlled with intraoperative spinal ultrasound. After enrollment of 12 patients, the cell concentration for transplantation was increased, since analysis of the intended cell concentration within syringes was only 66% of the previously estimated concentration. From 1816 prescreened patients, only 26 patients received dermis-stimulated macrophage grafts within 14 days post injury. Seventeen patients with standard of care treatment served as controls. All patients recruited from October 2003 to March 2006 suffered a cervical/thoracic SCI (AIS A). The clinical trial was prematurely halted for financial reasons. Only 10 patients reaching the 6-month follow-up visit were used for primary outcome analysis, which surprisingly showed a conversion rate in control patients of 58.8% versus 26.9% in macrophage grafted patients. This result was not statistically significant but suggested a trend favoring the control group. As severe adverse events of spinal instability and bacterial meningitis were reported.

Some aspects specific for this approach have to be addressed. A characterization of grafted cells preclinically (cell migration, survival) was completely missing. Details of the grafting procedure were not evaluated systematically in the preclinical setting (location of injected cells, visualization of damaged cord prior to grafting). The interpretation of behavioral data (BBB locomotor score), which represented a key element in assessing the therapeutic potential of this strategy, needs to be discussed.

The loss of oligodendrocytes at the lesion as well as rostral and caudal to the lesion site, resulting in demyelination and loss of appropriate nerve conduction, significantly contributes to functional deficits after SCI (Crowe et al., 1997, Gledhill et al., 1973, Hulsebosch, 2002). Clinical assessment and postmortem morphological studies in SCI individuals indicate that a preserved rim of spared axons frequently exists even in motor complete SCI (Kakulas, 1999, Sherwood et al., 1992). The majority of these spared axons was found to be demyelinated (Kakulas, 1999). The adult spinal cord exhibits only a limited intrinsic capacity for oligodendroglial replacement and remyelination. In particular, the terminal differentiation of oligodendroglial precursor cells into mature myelinating oligodendrocytes is insufficient (Levine & Reynolds, 1999). Therefore, regenerative strategies aiming to replace myelinating oligodendrocytes represent one potential means to promote recovery even in functionally complete SCI patients.

The group around Hans Keirstead differentiated hESCs of the H7 hESC line into oligodendrocyte progenitor cells (OPC) of high purity (Keirstead et al., 2005). Two lakhs and fifty thousand enriched OPC were grafted 7 days after a moderate or severe thoracic contusion SCI 4 mm caudal and rostral to the “lesion epicenter.” The “lesion epicenter” was not further specified. The grafted OPC differentiated into mature adenomatous polyposis coli (APC) expressing oligodendrocytes. The total number of axons remyelinated by oligodendrocytes and Schwann cells increased by 136%. Grafted OPC accounted for 55% of axon remyelination. OPC grafting elicited locomotor improvement (around 17 on the BBB locomotor scale = “frequent to consistent weight supported plantar steps and frequent FL–HL coordination” compared to around 13 in medium injected control animals = “consistent plantar stepping and consistent FL–HL coordination during gait”; four-parameter kinematic analysis). OPC grafting into the chronically injured rat spinal cord 10 months after injury failed to promote remyelination and functional recovery (Keirstead et al., 2005). Of note, tissue sparing remained unchanged in treated and control animals. In a follow-up study, identical hESC-derived OPC grafted 7 days after a rat cervical midline contusion injury promoted gray and white matter sparing (neuroprotection) paralleled by recovery of locomotor function compared to the nontransplanted group (Sharp et al., 2010). Surprisingly, in this study, remyelination was not increased after OPC grafting.

In early 2009, the U.S. Food and Drug Administration approved the first clinical trial with hESC-derived OPC for transplantation after acute SCI. Geron Corp. initiated a phase I multicenter trial in patients with complete thoracic SCI (AIS grade A). Two million hESC-derived OPC were injected into the spinal cord lesion site within 14 days after injury accompanied by immunosuppression with tacrolimus. The trial's primary end point was safety (adverse events related to the injected stem cells, the injection procedure, or the immunosuppressive treatment). Neurological function was assessed as a secondary end point (Mayor, 2010). In November 2011, the company announced to discontinue the clinical trial for financial reasons after five patients underwent the cell transplantation procedure. Apparently establishing proper cell purification protocols, problems with cyst formation after transplantation and slow patient recruitment caused an unforeseeable cost explosion. It took around 12 years from the bench all the way to the clinical trial. Developmental costs for this stem cell-based therapy to the point of first patient enrollment in 2010 were estimated $170 million (http://content.usatoday.com/communities/ondeadline/post/2010/10/us-doctors-begin-first-clinical-trial-of-embryonic-stem-cell-on-humans/1#.T8U_OJgjPCY). According to the company, the therapy did not promote any improvement but was well tolerated with no serious adverse events (Kaiser, 2011).

The prime goal of OPC grafting to replace oligodendroglia and thus promote remyelination and functional recovery is based on the evidence that even in the most severe cases of SCI (AIS grade A) a spared rim of demyelinated axons remains (Kakulas, 1999). However, it has yet to be demonstrated that spared axon remyelination alone is sufficient to promote relevant functional recovery. Conflicting preclinical data exist for remyelinating capacities of hESC-derived OPC in different lesion models (Keirstead et al., 2005, Sharp et al., 2010). The approach was not recapitulated by an independent laboratory before translation into a clinical trial.

Almost 25 years ago, membrane proteins of CNS myelin (NI-250) were identified to inhibit axon growth in vitro (Schwab & Caroni, 1988). In the following years, specific antibodies developed to neutralize respective myelin-associated molecules (IN-1 antibody) promoted axon regrowth in small animal models after corticospinal tract (CST) lesions at brainstem and spinal cord level (Bregman et al., 1995, Caroni and Schwab, 1988, Schnell and Schwab, 1990, Thallmair et al., 1998). After purifying and cloning of the main component of NI-250, Nogo-A (Chen et al., 2000, Grandpre et al., 2000), new antibodies (11C7 and 7B12) were generated.

Further, studies identified the inhibition mediating receptor of soluble Nogo-66 (Ngr) (Fournier et al., 2001). Albeit not mediating the inhibitory effects of membrane-bound amino Nogo, this receptor was identified to mediate inhibitory effects of two other myelin-associated proteins (Mag, Omgp) (Domeniconi et al., 2002, Liu et al., 2002, Wang et al., 2002). Consequently, it was hypothesized that there is redundancy of inhibitory molecules converging to the same receptor complex (Ngr-p75NTR) (Filbin, 2003, Wang et al., 2002, Wong et al., 2002, Yamashita et al., 2002). In line with this redundancy, several rodent gene knockout strategies exhibited heterogeneous (and even strain dependent) results with respect to axonal growth inhibition (Cafferty et al., 2007, Dimou et al., 2006, Kim et al., 2003, Kim et al., 2004, Simonen et al., 2003, Zheng et al., 2005, Zheng et al., 2003).

Numerous related preclinical studies investigated the effects of inhibiting Nogo-A- or Ngr-mediated signaling with specific antibodies or peptide administration (intracerebral/intrathecal/intraperitoneal/subcutaneous) to elicit structural and functional recovery after SCI (Bregman et al., 1995, Fouad et al., 2004, Freund et al., 2006, Grandpre et al., 2002, Li et al., 2004, Li and Strittmatter, 2003, Liebscher et al., 2005, Merkler et al., 2001, Schnell and Schwab, 1990). In terms of structural repair, sprouting of CST axons at supraspinal and spinal level (above and below the lesion) was the most frequent correlate. A confirmation of target neuron reinnervation by sprouting axons has not been reported. The most frequent reported functional outcome parameter was BBB locomotor assessment. Across the board, locomotor improvement effect size was moderate (on average BBB locomotor score improvement between two and four points) achieved within 4–8 weeks post injury (Table 7.1). Only one study (intrathecal application of the Ngr antagonist NEP1-40) reported a delayed treatment effect (initiation of treatment 7 days post injury) (Li & Strittmatter, 2003). None of the preclinical studies in this context investigated regenerative effects in a rodent contusion model. The treatment with the Ngr antagonist NEP1-40 was reassessed by an independent laboratory using the identical animal model (mouse dorsal hemisection) and identical mode of application (Steward et al., 2008), which failed to replicate previous findings of structural and functional recovery (Li & Strittmatter, 2003). The antibody-mediated neutralization of Nogo-A was not reevaluated by an independent laboratory (Zorner & Schwab, 2010),

Based on the extensive preclinical investigations, an open-label multicenter clinical phase I study of anti-Nogo-A treatment was initiated (http://clinicaltrials.gov/ct2/show/NCT00406016?term=ATI355&rank=1). At this point, 51 ASIA-complete para-/tetraplegic patients received intrathecal infusions/boli with the anti-Nogo-A antibody ATI355 within 28 days of SCI between 2006 and 2011 (Abel, Baron, & Casha, 2011). No study drug related deaths were reported. In the group receiving continuous drug infusion, five severe adverse events (infection, mechanical-device complications) were reported. Overall, the study medication was considered to be safe. Compared to the continuous infusion regimen, repeated intrathecal bolus injections improved safety. Data regarding treatment efficacy have not been disclosed.

Overall, the extensive preclinical evaluation of therapeutic approaches interfering with Nogo-A/Ngr-associated axon growth inhibition demonstrated limited structural and functional recovery without providing a clear structural mechanism such as proper target reinnervation. The delay in treatment initiation (up to 28 days) in the clinical trial was not reflected in any of the Nogo-A related preclinical studies.

The Rho-associated kinase (ROCK) signaling pathway has been identified as an important mediator of axon growth inhibitory signaling. The analysis of the downstream intracellular signaling cascade after myelin-associated Ngr–p75NTR-complex activation identified Rho activation as the most upstream event (Yamashita et al., 2002). Efforts to overcome growth inhibitory effects of this pathway focused on the inactivation of RhoA by C3 transferase and the inhibition of ROCK by Fasudil, Y-27632, and p21Cip1/WAF1 (Tonges, Koch, Bahr, et al., 2011). Numerous studies investigated these compounds in a variety of rat/mouse SCI models (Chan et al., 2005, Dergham et al., 2002, Fournier et al., 2001, Nishio et al., 2006, Sung et al., 2003, Tanaka et al., 2004; Table 7.2). After immediate or delayed, local or systemic application, these compounds yielded different morphological and functional outcomes. Four out of thirteen studies reported tissue sparing effects, whereas 2/13 experiments failed to show this effect. In 1/13 studies, scar reduction effects were observed with one (C3 transferase) but not with the other compound (Y-27632). The majority (6/13) reported axonal sprouting as a potential regenerative mechanism, whereas 2/13 failed to identify axon sprouting as a relevant mechanism. Regrowth of transected axons (termed axon regeneration) was seen in 2/13 and not seen in 6/13 experiments interfering with the ROCK pathway. Finally, 9/13 studies reported beneficial functional effects, while 4/13 studies could not confirm functional effects. Taken together, results from different studies show a remarkable variety of morphological effects culminating in controversial effects on behavioral outcome (in most instances locomotion). In particular, the investigation of C3 transferase inhibitory effects in a mouse thoracic hemisection model, which yielded morphological (CST regrowth up to 12 mm) and functional recovery (improvement of 8 BBB locomotor score points within 24 h after injury compared to controls), fueled further efforts toward clinical translation. Subsequently, the investigation of a dura-permeable formulation of C3 transferase (BA-210, Cethrin; local release from a fibrin sealant) demonstrated efficacy-locomotor function improvement from an estimated 8.2 to 9.5 on the BBB scale-in a rat SCI contusion model (Lord-Fontaine et al., 2008). Interestingly, the main morphological effect reported in this study was tissue sparing, whereas experiments with C3 transferase and Y-27632 from the same group (Dergham et al., 2002) identified axon regeneration as the prime mechanism of functional recovery.

These findings laid the corner stone for a subsequent clinical phase I/II a trial, which demonstrated safety of the invasive procedure (Fehlings, Theodore, Harrop, et al., 2011). Within 52.6 h (range 7.83–146.1 h) after injury, BA-210 was applied epidurally and released from a fibrin sealant in a dose range from 0.3 to 9 mg. Forty-eight patients, selected from nine clinical sites in the United States and Canada, were enrolled between February 2005 and November 2007, 35 patients completed the study. Of these, 23 had thoracic and 12 cervical SCI, all AIS A. Adverse events were within the typical range of SCI patients, severe adverse events were not reported. Efficacy data were inconclusive due to the low sample size.

In the relevant rat contusion model (Lord-Fontaine et al., 2008), the Cethrin/fibrin compound was tested to be effective only in the immediate drug application paradigm. In a spinal cord hemisection mouse model, a treatment delay up to 24 h post injury yielded improved locomotor outcome data. In contrast, patients in the respective clinical trial received the treatment with a mean treatment delay of 52.6 h after injury (Fehlings et al., 2011).

Based on this review of four “bed to bedside” treatment strategies, the following question arises: How can we improve preclinical research in order to increase the likelihood that translation into the clinic will become successful? Several reviews (Anderson et al., 2005, Blesch and Tuszynski, 2009) and a recently published survey among the SCI research community (Kwon, Hillyer, & Tetzlaff, 2010) identified prerequisites which preclinical research should fulfill before respective therapeutic strategies can be applied to the clinical setting.

Overall, both reviews and the survey come to similar conclusions. Efficacy of a pharmacological and in particular a cell-based intervention should be assessed not only in rodent models of SCI but also in large animal models (e.g., cats, dogs, rabbits, sheep). According to the survey, primate experiments are not considered as crucial. Contusion SCI is the most relevant animal model to mimic human SCI pathology and should therefore be employed in preclinical studies prior to translation. Moreover, a given therapy should be investigated in different injury models as well as different injury severities to document the robustness of a given approach. In the survey, the majority of respondents felt that morphological, biochemical, or neurophysiologic outcome parameters are not sufficient to replace behavioral tests, since they do not represent functionally meaningful outcome parameters. According to the survey, locomotor assessment-in particular, the frequently employed semiquantitative analysis (BBB locomotor rating scale)-is still considered to represent “clinically meaningful efficacy” best (Kwon et al., 2010). In terms of treatment timing, a therapy with the intention to be applied within 12 h after SCI injury in humans should be investigated with a minimum treatment delay of 6 h in the preclinical setting. If the treatment delay in humans is longer (e.g., 5 days), corresponding delays need to be investigated preclinically (Anderson et al., 2005, Kwon et al., 2010). Finally, there is strong consensus that evidence of efficacy from a single laboratory is not sufficient. Respective experiments need to be replicated by other independent laboratories.

How did the approaches discussed here perform in this respect (Table 7.3)? None of the studies reported a large animal model (pig, sheep, cat, dog, rabbit), with the exception of anti-Nogo-A treatment, where primate experiments were conducted (Fouad et al., 2004, Freund et al., 2006). Despite thorough investigations of the anti-Nogo-A treatment over more than a decade, a contusion SCI model was never used. All other compounds and cell-based therapies were analyzed for treatment efficacy in rodent contusion SCI. A clear dose response curve is missing in all studies reviewed here. Efficacy in different injury models has been reported in all four treatments paradigms, whereas efficacy in different injury severities has not been assessed systematically in any of the paradigms analyzed. Behavioral improvement has been reported consistently—in almost all instances investigated with the BBB locomotor rating scale. Looking specifically at BBB locomotor assessment, the magnitude of improvement was moderate at best, typically varying between 1 and 4 BBB score points on a scale from 0 to 21 (Table 7.1). Overall, treatment timing in preclinical experiments corresponded roughly to the time points of the drug/cell administration in the respective clinical trials except for the anti-Nogo-A therapy. There, specific neutralizing antibodies were exclusively delivered immediately after the injury—in contrast to the clinical trial, where the compound was delivered within 28 days post injury. ROCK pathway inhibiting studies included treatment delays up to 5 days, which correlated well with the clinical drug (BA-210; Cethrin) administration (average treatment interval 52.6 h after the injury date). Only Ngr and ROCK pathway inhibition approaches have been replicated by independent laboratories-a criterion, which is valued quite high in predicting robustness and efficacy of a given treatment strategy. In case of Ngr inhibition, a follow-up study (Steward et al., 2008), which represents one of the first published studies (based on the NIH funded initiative “facilities of research excellence (FORE) in SCI replication studies”) with the intention to reassess promising therapeutic approaches in SCI by independent investigators, failed to replicate previous restorative effects (Li & Strittmatter, 2003). None of the exact compounds, which were investigated in respective clinical trials (anti-Nogo-A antibody, BA-210), were analyzed by independent laboratories.

Beyond published requirements for preclinical experiments, additional aspects should be considered. The key for a successful clinical translation is to demonstrate a clear and precise structural mechanism, which explains the observed functional improvement. For neuroprotection as underlying mechanism, this has been accomplished by the demonstration of a significant white/gray matter tissue sparing. For axon sprouting/regeneration as the proposed structural mechanism, not only sprouting/regeneration for a limited distance needs to be demonstrated but also the reinnervation of appropriate target neurons. Recently, published studies provide excellent examples for specific treatment strategies to promote reinnervation of target neurons (Alto et al., 2009, Bonner et al., 2011). Moreover, suitable neurophysiologic assessment tools allow for a specific correlation of observed structural changes with function (Bonner et al., 2011), which is often completely lacking in rough locomotor assessment measures such as the BBB scale.

Besides vague structural mechanisms, the demonstration of a robust structural and functional effect is almost always missing. Typically, effect sizes in terms of functional recovery range in the magnitude of 2–4 BBB points, which obviously is not sufficient for clinical translation. For example, a BBB score of 3 is defined as “extensive movements of two joints,” whereas a BBB score of 7 represents “extensive movements of all three joints of the hindlimb” (Basso et al., 1995). This magnitude of BBB improvement cannot be considered as being clinically meaningful, leaving aside the fact that an untrained observer is unable to detect such subtle differences. Therefore, it is suggested to define a threshold to be reached of at least nine on the BBB locomotor rating scale (“…weight support in stance only…or occasional, frequent, consistent weight supported dorsal stepping…”). Further, average improvements of at least 6 BBB points compared to an appropriate control should be achieved to increase probabilities that the observed improvements predict a clinically meaningful difference.

The “robustness” factor also applies to the extent of structural repair. Looking at the lesioned CST, the smaller subcomponents (ventral and dorsolateral and lateral), which account for 2.5% (roughly 400 axons) of the complete CST in rats, allow to compensate for the transected main dorsal component (roughly 17,000 axons) to restore forelimb reaching function (Weidner, Ner, Salimi, et al., 2001). Therefore, at least 2.5% of a transected tract system should regenerate to predict meaningful functional improvement in a planned clinical trial. A recent study, which is one of very few studies fulfilling the “robustness” criterion, demonstrated massive axon outgrowth into the completely transected adult rat spinal cord originating from embryonic day 14 rat spinal cord (Lu, Wang, Graham, et al., 2012). Ideally, the observed axon regeneration should lead to substantial target reinnervation. In a case where axon regeneration is determined as the relevant mechanism, proper myelination of these axons represents another hurdle toward restored nerve conduction and ultimately functional recovery. As an example, the unmyelinated state of regenerated axons prohibited restoration of nerve conduction despite the fact that target reinnervation took place (Alto et al., 2009).

An issue relevant for cell-based therapies, which is not at all addressed at the preclinical state, is the precise transplantation site. Realistically, transplantation interventions will most likely be performed days to weeks after injury. Translated to rodents, transplantation delayed for several days makes the precise identification of the lesion area impossible by just looking at the cord surface through a microscope. Therefore, the variation of the target site within the injured spinal cord is significant with the consequence that grafts are found either in the midst of the cystic lesion at the host–graft border or even within the mostly uninjured host tissue. Noninvasive high-resolution imaging techniques need to be introduced in small animals to allow proper identification and targeting of the defined graft site. Visualization methods such as high-resolution ultrasound or high-field MRI (Weber, Vroemen, Behr, et al., 2006) are available but not trivial to apply in the preclinical setting. In cell-based clinical trials, ultrasound-guided cell transplantation is becoming a standard procedure to delineate the cystic lesion from adjacent spinal cord (Lammertse et al., 2012).

In order not to miss a higher magnitude of structural and in particular functional effects, repair strategies need to be combined with appropriate rehabilitation interventions already at the preclinical level. This has been demonstrated in case of chondroitinase treatment, where the combination with task-specific rehabilitation (forelimb reaching training) after cervical SCI significantly augmented behavioral effects observed with the regenerative treatment alone (Garcia-Alias, Barkhuysen, Buckle, et al., 2009). Another rationale to introduce rehabilitation already at the preclinical level is to determine the appropriate timing of regenerative therapies in relation to the administration of physical therapy activities, which represent standard care in SCI individuals. Indeed, the parallel combined treatment of Nogo-A neutralization and locomotor treadmill training in SCI rats yielded impaired functional outcome (Maier, Ichiyama, Courtine, et al., 2009) suggesting that the timing needs to be adjusted in order to potentiate positive effects from both approaches.

Almost completely neglected are nonmotor functional deficits such as bowel and bladder dysfunction impairing quality of life more than any other SCI sequel particularly in chronic stages (Anderson, 2004). Future preclinical studies need to integrate respective parameters (analysis of regrowth of relevant axon pathways; analysis of relevant functional parameters) to address patients’ unmet medical needs.

Section snippets

Conclusion

It is impressive to see to which extent preclinical research has been translated into clinical studies. Fifteen years ago, this was considered a goal of the distant future. In order to realize properly conducted clinical trials, clinical networks have been established which allow to perform multicenter studies in a highly standardized fashion. As an example, within the European Multicenter Study about Spinal Cord Injury (EMSCI) network, almost 20 SCI centers across Europe continuously train

References (76)

  • P. Lu et al.

    Long-distance growth and connectivity of neural stem cells after severe spinal cord injury

    Cell

    (2012)
  • Y. Nishio et al.

    Delayed treatment with Rho-kinase inhibitor does not enhance axonal regeneration or functional recovery after spinal cord injury in rats

    Experimental Neurology

    (2006)
  • A. Pinzon et al.

    A re-assessment of erythropoietin as a neuroprotective agent following rat spinal cord compression or contusion injury

    Experimental Neurology

    (2008)
  • A.M. Sherwood et al.

    Evidence of subclinical brain influence in clinically complete spinal cord injury: Discomplete SCI

    Journal of the Neurological Sciences

    (1992)
  • M. Simonen et al.

    Systemic deletion of the myelin-associated outgrowth inhibitor Nogo-A improves regenerative and plastic responses after spinal cord injury

    Neuron

    (2003)
  • O. Steward et al.

    A re-assessment of the consequences of delayed transplantation of olfactory lamina propria following complete spinal cord transection in rats

    Experimental Neurology

    (2006)
  • O. Steward et al.

    A re-assessment of the effects of a Nogo-66 receptor antagonist on regenerative growth of axons and locomotor recovery after spinal cord injury in mice

    Experimental Neurology

    (2008)
  • J.K. Sung et al.

    A possible role of RhoA/Rho-kinase in experimental spinal cord injury in rat

    Brain Research

    (2003)
  • H. Tanaka et al.

    Cytoplasmic p21(Cip1/WAF1) enhances axonal regeneration and functional recovery after spinal cord injury in rats

    Neuroscience

    (2004)
  • B. Zheng et al.

    Lack of enhanced spinal regeneration in Nogo-deficient mice

    Neuron

    (2003)
  • R. Abel et al.

    Therapeutic anti-Nogo-A antibodies in acute spinal cord injury: Safety and pharmacokinetic data from an ongoing first-in-human trial

  • L.T. Alto et al.

    Chemotropic guidance facilitates axonal regeneration and synapse formation after spinal cord injury

    Nature Neuroscience

    (2009)
  • K.D. Anderson

    Targeting recovery: Priorities of the spinal cord-injured population

    Journal of Neurotrauma

    (2004)
  • D.K. Anderson et al.

    Recommended guidelines for studies of human subjects with spinal cord injury

    Spinal Cord

    (2005)
  • D.M. Basso et al.

    A sensitive and reliable locomotor rating scale for open field testing in rats

    Journal of Neurotrauma

    (1995)
  • J.F. Bonner et al.

    Grafted neural progenitors integrate and restore synaptic connectivity across the injured spinal cord

    The Journal of Neuroscience: The Official Journal of the Society for Neuroscience

    (2011)
  • B.S. Bregman et al.

    Recovery from spinal cord injury mediated by antibodies to neurite growth inhibitors

    Nature

    (1995)
  • M.S. Chen et al.

    Nogo-A is a myelin-associated neurite outgrowth inhibitor and an antigen for monoclonal antibody IN-1

    Nature

    (2000)
  • M.J. Crowe et al.

    Apoptosis and delayed degeneration after spinal cord injury in rats and monkeys

    Nature Medicine

    (1997)
  • A. Curt et al.

    Providing the clinical basis for new interventional therapies: Refined diagnosis and assessment of recovery after spinal cord injury

    Spinal Cord

    (2004)
  • P. Dergham et al.

    Rho signaling pathway targeted to promote spinal cord repair

    The Journal of Neuroscience: The Official Journal of the Society for Neuroscience

    (2002)
  • L. Dimou et al.

    Nogo-A-deficient mice reveal strain-dependent differences in axonal regeneration

    The Journal of Neuroscience: The Official Journal of the Society for Neuroscience

    (2006)
  • J.W. Fawcett et al.

    Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: Spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials

    Spinal Cord

    (2007)
  • M.G. Fehlings et al.

    A phase I/IIa clinical trial of a recombinant Rho protein antagonist in acute spinal cord injury

    Journal of Neurotrauma

    (2011)
  • M.T. Filbin

    Myelin-associated inhibitors of axonal regeneration in the adult mammalian CNS

    Nature Reviews. Neuroscience

    (2003)
  • K. Fouad et al.

    Regenerating corticospinal fibers in the Marmoset (Callitrix jacchus) after spinal cord lesion and treatment with the anti-Nogo-A antibody IN-1

    The European Journal of Neuroscience

    (2004)
  • A.E. Fournier et al.

    Identification of a receptor mediating Nogo-66 inhibition of axonal regeneration

    Nature

    (2001)
  • P. Freund et al.

    Nogo-A-specific antibody treatment enhances sprouting and functional recovery after cervical lesion in adult primates

    Nature Medicine

    (2006)
  • Cited by (0)

    View full text