InVivo Therapeutics: A Reality Distortion Field

| About: InVivo Therapeutics (NVIV)

The term Reality Distortion Field (RDF) was coined by Bud Tribble at Apple Computer to describe how by distorting sense of proportion and scales of difficulties a skillful manager can turn a concept into reality. Leaders of companies with such products often have an ability to convince themselves and others to believe almost anything with a mix of charm, charisma, bravado, hyperbole and persistence. However, only the most exceptional business leaders (e.g., Steve Jobs) have succeeded in coaxing product developers to deliver breakthrough products based on RDF.

Enter the world of InVivo Therapeutics (NASDAQ:NVIV) whose stock has soared more than 300% through 2013 boasting at one point a market cap of over $400 million. At the core of investor enthusiasm lies a "Biocompatible Polymer Scaffolding Device," set to enter clinical development for the treatment of spinal cord injury (SCI). A review of the company's investor presentation highlights the supposed extraordinary attributes of this device and its potential end-market. These include videos demonstrating the remarkable recovery of an African Green Monkey following an induced spinal cord injury; a description of the limited clinical development program the company is set to purse under the FDA's Humanitarian Device Exemption (HDE) regulatory path; and a projection of a unit price to exceed $60K.

In this article I shall delineate why current investor enthusiasm in relation to NVIV is unwarranted. While RDF might be a legitimate internal management tool, NVIV's selective and unsubstantiated promotional claims are misleading at best. A careful review of these claims makes it evident that the current NVIV share price reflects overly inflated expectations. Realistically, NVIV should trade at least 80% lower than its current level.

In the first part of this article, I review: (I) the nature of spinal cord injury, (II) the current management of SCI, and (III) emerging therapeutic approaches to SCI. In the second part, I make the following observations about NVIV and its products:

IV. Existing data do not demonstrate any clinical utility. Contrary to the company's assertions, detailed non-human primate data, presented to date, fails to demonstrate clinical utility for the biomaterial scaffold developed by NVIV. Strikingly, management consistently excludes such data from its investor presentations.

V. HDE approval unlikely to generate meaningful revenue. FDA approval of NVIV's scaffold product through the HDE regulatory path is unlikely to allow for any meaningful product sales. Management's assertion of per unit pricing of $60K or more is inconsistent with current HDE guidelines.

VI. Full approval would be complex and costly. Demonstration of the scaffold's clinical utility to a level that would enable significant reimbursement and sales would require lengthy, complex and costly registrational studies. Those would require a significant commitment that is well beyond the company's current financial resources. Importantly, even if the company were to undertake such a path, to date, there is no compelling evidence to suggest a potential positive outcome in such studies.

I. Spinal Cord Injury

Damage to the spinal cord, often as a result of motor vehicle accidents, falls, violence and sport injuries, has severe clinical consequence. The initial trauma to the spinal cord typically produces pathologic flexion, rotation, extension and/or compression of the tissue, blood vessel damage, dislocation of bones, rupture of inter-vertebral discs, and injury to ligaments. This initial damage can cause immediate death of neurons and transection of axons at the site of injury. The ensuing sequence of pathological events is referred to as secondary injury beginning immediately as a result of trauma and continuing for several weeks leading to an expanded region of tissue destruction. These progressive mechanisms of cord injury are complex and incompletely understood. Possible mechanisms include ischemia, hypoxia, inflammation, edema, excitotoxicity, disturbances of ion homeostasis, and apoptosis. It sometimes clinically manifests itself by continued neurologic deterioration within the initial hours post injury.

II. Management of Spinal Cord Injury

Before weighing the potential value of Invivo's technology, it is worthwhile to understand the complexities of traumatic spinal cord injury (TSCI) management, and the history of attempted treatment options including: therapeutic approaches, cellular transportation and tissue engineering.

Medical Support

On presentation, patients with TSCI require intensive medical care and continuous monitoring of vital signs, cardiac rhythm, arterial oxygenation, and neurologic signs in the intensive care unit for the first 7-14 days after injury. In observational studies, the standardized admission of patients with spinal injuries to an ICU has been associated with reduced mortality and morbidity in addition to improved neurologic recovery.

There is consistent evidence that avoiding hypotension and maintaining aggressive blood pressure targets improves neurologic recovery and reduces mortality. Further, patients should be started on occupational and physical therapy as soon as possible to facilitate potential functional recovery.

Decompression and stabilization

The preclinical literature provides biological evidence for the need to decompress the spinal cord subsequent to injury. There are, however, no current standards regarding the role, timing, and method of vertebrate decompression in acute spinal cord injury. Options include closed reduction using traction, and open surgical procedures.

Closed reduction is a treatment option for cervical spine fracture with subluxation but not for thoracic and lumbar fractures. It may obviate surgery and promote neurologic improvement in some cases. Indications for cervical spine surgery include significant cord compression with neurologic deficits, especially those that are progressive, that are not amenable or do not respond to closed reduction, or an unstable vertebral fracture or dislocation. Retrospective studies provide conflicting reports as to the effect of early surgery on neurologic recovery. However, a prospective trial (The Surgical Timing in Acute Spinal Cord Injury Study - STASCIS) found that early surgery was associated with better neurologic recovery at 6 months as defined by a 2-grade improvement in the American Spinal Injury Association impairment scale.

III. Therapeutic Approaches to TSCI

Pharmacologic agents

The synthetic glucocorticoid methylprednisone is the only treatment that has been suggested in clinical trials to improve outcomes in patients with acute, non-penetrating TSCI. It has indeed become standard treatment in many centers. Evidence of its potential efficacy in improving motor recovery, however, is limited, relying on post-hoc subgroup analysis in one study - The National Acute Spinal Cord Injury Study (NASCIS II).

In 2013, based upon the available evidence, the American Association of Neurological Surgeons and Congress of Neurological Surgeons stated that the use of glucocorticoids in acute spinal cord injury is not recommended. Instead, it is viewed as a treatment option rather than a standard.

Cellular transplantation

Stem cell transplantation and autologous non-stem cells of various cellular subtypes (e.g., bone marrow-derived stem cells, olfactory ensheathing cells or Schwann cells) have been studied in preclinical models. Such studies, employing cellular transplantation, either alone or in combination with other therapies, have been associated with enhanced neurobehavioral recovery, with no single cellular subtype showing superiority. Early clinical studies, while exhibiting relative safety, have yet to establish efficacy for the transplantation of any cellular line.

Tissue engineering and scaffold technology

The current focus of tissue engineering research is to develop strategies to overcome the unfavorable nature of the scarring, enhance intrinsic axon regeneration ability, increase the number of damaged ascending and descending axons that regenerate, guide the regenerating axons back to their appropriate target regions and encourage the remyelination of these regenerative axons.

Implantation of substrates such as cellular grafts can potentially promote axonal regrowth. Such growth, however, is typically highly random and does not extend past the graft site to re-enter host tissue. Artificial tissue scaffolds are conceptually designed to provide mechanical support for axon regrowth and potentially to serve as a local delivery system of therapeutics and/or stem cells that might facilitate repair.

A variety of biomaterials, both synthetic and natural, have been modified to fabricate tissue scaffolds. Invivo's core product is a synthetic porous biopolymer consisting of polylactic-co-glycolic acid (PLGA) and polylysine. PLGA is considered to be the gold standard in tissue engineering due to its biodegradability and bioresorbability in vivo. It is approved by the FDA for applications such as surgical sutures, drug delivery and tissue engineering. The incorporation of polylysine into a variety of scaffolds was observed to promote neuronal cell adhesion and neural network formation.

Of note, the use of artificial scaffolds for the repair of SCI is still in its infancy and what constitutes an optimal scaffold remains to be defined. Parameters such as morphology, architecture, mechanical properties, surface properties, degradation rate, composition of biological components, and the variation of these factors with time could have substantial effect on potential outcomes.

IV. InVivo's biomaterial scaffold fails to demonstrate clinical utility in non-human primate studies

Needless to say that therapeutic development is an extraordinarily risky and expensive business proposition. No such risks, however, seem to be conveyed by InVivo's management in its public presentations or in the CEO's media interviews. Instead, a consistent message is delivered to investors suggesting that InVivo's technology can effectively address the consequences of spinal cord injury whether acute or even chronic. To bolster such arguments, impressive video images, of monkeys recovering from surgically-induced spinal cord injury, are presented as the most compelling evidence.

As presented below, close examination of actual data from InVivo's studies (which are never highlighted by the CEO to investors), suggest that the preclinical model employed by the company is highly variable and does not lend itself to the bold assertions made by management. The following are my specific observations, expanded below:

  • All study monkeys, controls included, exhibit a return toward baseline neuromotor scores following post-operative paralysis. This observation counters notions and perceptions that recovery was achieved with the aid of InVivo's scaffold alone.
  • Data across the primate studies demonstrate wide variability of post-paralysis recovery. Specifically, the same treatment across the two studies reported by NVIV results in as much as a 150% difference in neuromotor scores (15 versus 6 on a 20-point scale). Such variability brings into question the value of this preclinical model in evaluating a treatment effect and suggests a great degree of randomness.
  • The suspicious absence of standard deviation bars in the second primate study (n=16) compared to their inclusion in the first study (n=4), suggests potential recovery variability that the company elects not to highlight.
  • We observe an inconsistent supposed treatment effect across studies. Specifically, the scaffold alone group exhibits much better recovery rates compared to the scaffold + human neural stem cells. This contrasts with the data from the first trial where the opposite was true.

Early studies, conducted by InVivo's scientific founder, using PLGA+polylysine scaffold alone, or seeded with murine neural stem cells, demonstrated potential improvements in an adult rat hemisection (at the T9-T10 level) model of SCI as compared to controls (see Figure 1).

Figure 1. Functional recovery (Rodent study)

(Click to enlarge)

Source: Teng YD et al., 2002

Rodents, however, exhibit high rates of spontaneous recovery from induced spinal cord injury, even following profound lesions. The use of non-human primates, therefore, was seen as potentially offering greater relevance in facilitating translation of preclinical findings to humans. To date, InVivo has conducted and disclosed data from two non-human primate studies for the preclinical evaluation of its scaffold technology (referred to as the 2008 and 2010 studies). Data from an additional ongoing trial (the 2011 study) is yet to be disclosed, although management has disclosed in its filings that initial results are consistent with prior data seen in rodents and monkeys

All monkeys, control included, exhibited a return toward baseline neuromotor scores

The initial 2008 study conducted by InVivo served the purpose of potentially establishing a spinal cord injury model in the African green monkey. There were a total of only four monkey subjects in this study, making any genuine data analysis, let alone conclusions, weak at best (see Figure 2). The scaffold seeded with human neural stem cells (n=2 except for day 1 and 44) did appear to perform best on rate of recovery exhibiting a separation post day 24. No difference is observed between the control (n=1) and scaffold alone (n=1) to day 30. Post day 36 there is indeed an abrupt separation.

The difference among subjects, if any, might have been in the recovery rate of subjects but not in their ultimate recovery. As discussed by the investigators in the published manuscript, "All subjects exhibited a return toward baseline neuromotor scores in the ipsilateral hind limb," following complete post-operative paralysis as a consequence of the creation of a hemisection in the thoracic spine at the T9-T10 level. The authors indeed conclude that the observed degree of recovery and neural plasticity is consistent with reported findings in other primate SCI models.

Figure 2. Functional recovery (study 2008)

(Click to enlarge)

Source: Pritchard CD et al., 2010

Second primate study highlights variability of recovery rates and implies great degree of randomness.

The second study included 16 African green monkeys undergoing a surgical hemisection. Strikingly, despite the greater number of subjects for data analysis, no error bars were included in the data presented in the company's SEC filing (see figure 3).

Upon review of the data, the relevance of this preclinical monkey model is called into question in relation to exploring potential treatment effects. As highlighted below, we observe wide variability in rates of recovery and inconsistent potential treatment effects limiting any meaningful conclusions from such trials.

The following specific observations can be made:

  • As expected, consistent recovery is observed in all subjects.
  • By day 44 (the day to which data is presented in the 2008 study - see dashed line in figure 3) the scaffold seeded with human neural stem cells group exhibited a mean neuromotor score of around 6. This contrasts with a score as high as 15 in the prior study by day 44 in the same treatment monkeys (see figure 2). A full 9 point difference (out of a total score of 20).
  • In contrast to the prior study, the scaffold alone group seemed to perform better than the group implanted with the scaffold seeded with human neural stem cells.

Figure 3: Functional recovery (study 2010)

(Click to enlarge)

Source: Company Form 10-K for the fiscal year ended December 31, 2012 p. 13

V. FDA approval via the HDE regulatory path unlikely to allow for any meaningful product sales

InVivo management consistently conveys bold assertions regarding the supposed effectiveness of its scaffold device. Yet, ironically, it has elected to pursue a regulatory approval path that avoids the hurdles of demonstrating such efficacy in robust clinical studies. As delineated below, while the company's device may indeed be approved by FDA, such approval is likely to come with critical restrictions that are unlikely to allow for any meaningful product revenues.

HDE regulatory path

The Safe Medical Devices Act of 1990 authorized the Humanitarian Device Exemption (HDE) to encourage the development and introduction of complex device technologies to meet the needs of small populations. Eligibility for an HDE requires the manufacturer to request a designation from the Office of Orphan Products Development (OOPD) as a Humanitarian Use Device (HUD). A HUD is specifically defined as a "medical device intended to benefit patients in the treatment or diagnosis of a disease or condition that affects is manifested in fewer than 4,000 individuals in the United States per year." The HDE path allows companies to avoid the expense and practical challenges of conducting FDA acceptable trials to demonstrate safety and effectiveness. [21 CFR 814.102(a)(5)]

An HDE application is similar to the traditional device premarket approval application (PMA), except that it need not include evidence of effectiveness. It does, however, need to contain sufficient information for FDA to determine that the device does not pose an unreasonable or significant risk of illness or injury and that the probable benefit to health outweighs the risk of injury or illness from its use.

Restrictions associated with products approved through the HDE path

FDA's Center for Devices and Radiological Health (CDRH) has put in place a number of restrictions for devices approved through the HDE path that limit price and allow for only limited cost recovery. InVivo management has included in its public presentations a $60K or more per unit price that appears inconsistent with current guidelines as presented below:

  • Companies are allowed to ship up to 4,000 devices per year (or a higher number if the data show that patients need more than one device within a year).
  • Sponsors must report data to the CDRH on a periodic basis to support the continued appropriateness of the HUD designation. If evidence indicates that the population criterion is no longer met, an HDE could be withdrawn by the agency.
  • Sponsors are limited to recovering the cost of research and development, manufacturing and distribution. They are not permitted to make a profit on the sale of the HUD, if the device is sold for more than $250. Profits are allowed for products intended for pediatric populations up to a specific annual limit. Further, the sponsor must provide supporting financial documentation to FDA about the price it proposes to charge.

There are a number of additional practical complications that can limit the actual use of HDE devices in the market place. While a sponsor can charge for the device, an institution might not be able to purchase it if adequate third-party reimbursement is not available to cover the institution's cost to purchase. Secondly, the use of a HUD requires the approval of an institutional review board (IRB) at the institution where it is to be used. Securing such approvals can be a difficult and a costly proposition for a sponsor to undertake.

VI. Demonstration of potential clinical utility is well beyond NVIV's current financial resources

In contrast to InVivo's current regulatory strategy, a traditional registrational path focused on the demonstration of clinical safety and efficacy would involve a complex, lengthy process of multiple studies.

Traumatic spinal cord injury remains a highly challenging indication for therapeutic development from both biological and study design standpoints. In an attempt to counter the complex theoretical cascade of pathological events (referred to as secondary injury and described above) subsequent to the initial trauma, neuroprotective and neuroregenerative agents have been and still are evaluated in clinical trials. To date, five pharmacologic therapies have been studied in TSCI Phase III trials, all of which, despite early promise, failed to demonstrate a conclusive benefit (see Table 1). To compound on those issues, challenges of demonstrating clinical benefit in TSCI studies relate to differences in patient baseline characteristics, their injuries and hence highly variable spontaneous recovery rates.

Pursuing such a complex development program could therefore easily extend for a period of at least 5 years, involve as many as 400 patients at a cost of more than $40 million. Only such robust demonstration of product safety and efficacy could allow for potential meaningful product revenues to sustain NVIV's current valuation. The scope of such a path, however, is well beyond the company's current financial resources and we see no mention of such plans at this point in time.

Table 1. Agents evaluated in Phase III trials of acute TSCI

(Click to enlarge)

Source: Wilson JR et al., 2013

In Summary

As delineated above, there is little evidence that management's reality distortion field has come anywhere close to inspiring the creation of products that support NVIV's market cap. Bold assertions in relation to its scaffold's efficacy and market potential cannot be substantiated based on existing data and the company's proposed regulatory path respectively. Further, management minimizes the actual challenges in bringing important new therapeutic alternatives to spinal cord injury patients.

Disclosure: I am short NVIV. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.