Biologic Considerations for Clinical Study Design: Fusion

Biologic Considerations for Clinical Study Design: Fusion

S. Raymond Golish, MD, PhD, MBA, FAAOS

Dr. Golish or an immediate family member serves as a paid consultant to or is an employee of Bio2Tech, Centinel Spine, Icotec, Intrinsic Therapeutics, Kuros Biosciences, Paradigm Spine, Simplify Medical, SpineBiopharma, and Wright Medical Technology, Inc.; serves as an unpaid consultant to Cytonics; has stock or stock options held in Cytonics and Cytonics, Inc.; and serves as a board member, owner, officer, or committee member of AAOS Biomedical Engineering Committee, ASTM, and North American Spine Society.


Two major themes emerge naturally from the topic of biologic considerations for clinical study design. The first theme is the distinction between design considerations for products that are being studied currently and those in the future versus the historical considerations from previously completed trials. Especially within regulatory affairs, the decision making that results in success is sufficiently complex that the history and rationale for why previous approaches were undertaken must be clearly understood.. Even when the approved study design for a new product departs from that of prior products, a new approach must be clearly substantiated and weighed against the potential pitfalls. To enable progress to occur, prior designs provide an essential stress test to any new approach.

With respect to past versus present, it is important to ask: Are transforaminal lumbar interbody fusion (TLIF) trials the new normal? In the past, anterior lumbar interbody fusion (ALIF) and posterolateral fusion (PLF) trials have been preferred. Currently, there are good reasons why TLIF trials are commonly performed, with multiple contemporary TLIF trials running under IDE/IND (investigational device exemption/investigational new drug). However, their design considerations are unique and must be carefully addressed because there are fewer historical guideposts compared with prior ALIF and PLF trials. The history of ALIF and PLF trials still guide decision making for a TLIF (or any other) trial.1

The second major theme is more complex and accentuated by the first: the three-way trade-off that exists among clinical, commercial, and regulatory considerations in product development.2,3,4,5 It may seem obvious that all three of these are in play in a mature product development program, but they interact in nuanced ways that must be scrutinized closely to avoid circular reasoning. For example, the following, simple clinical question is considered: Is iliac crest bone graft (ICBG) donor site morbidity severe enough to warrant complex orthobiologic products as alternatives to existing local autograft, allograft, and synthetics? A clinical answer may be yes for some clinicians, whereas it might be debatable for others; that is, a qualified yes versus perhaps a strong maybe.

But the ramifications of that clinical answer should be examined. From a commercial perspective, the answer cannot be a qualified yes or a strong maybe, it must be: ICBG is absolutely unacceptable. Otherwise, there would be no rationale for a major investment in science to develop our orthobiologic product. Further, the commercial answer must also be: In addition, local autograft is either insufficient in quantity or subpar for fusion quality, and allograft and synthetic bone graft extenders are subpar for fusion quality. Otherwise, a new product will only be able to achieve commodified pricing similar to existing extenders and will not be worth the investment. Without strongly affirmative answers to these commercial questions, one cannot proceed with a risky and expensive program.

Given the regulatory implications of the aforementioned commercial reasoning, a standard experimental design would be a two-arm noninferiority trial of the experimental product compared with a fusion control. ICBG is an acknowledged gold standard for fusion, but if ICBG is rejected for donor morbidity, it cannot (or should not) be enrolled as a control arm. Therefore, if local autograft, allograft, or synthetics are chosen as the control, the commercial belief that these low-cost products are inadequate and therefore not a gold standard for a control will be contradicted. In this case, the clinician should not have a noninferiority design. If commodified products are a poorer control, perhaps the clinician should conduct a superiority study instead of a noninferiority study. But then superiority would have to be shown to an active control, a challenging task for many reasons without precedent in historical trials.

From this simple example, the complexity and interrelationship of these trade-offs become clear.2 Figure 1 encapsulates some of these considerations, and the others are elucidated in the sections that follow.


To date, the most successful orthobiologics for fusion have been regulated as drug-device combinations jointly by the Center for Devices and Radiological Health (CDRH) and the Center for Drug Evaluation and Research. CDRH has acted as the lead center, with the most experience with spinal devices of all types, and the approval process was that for a class III medical device seeking premarket approval through CDRH on conducting IDE trials. Many products have been presented to CDRH’s Orthopaedic and Rehabilitation Devices Panel. Sponsors who interact with the Center for Drug Evaluation and Research or Center for Biologics Evaluation and Research for spinal drugs and biologics face similar regulatory considerations, so the text applies to all biologic technologies for fusion.6

The regulatory, clinical, and political dimensions of orthobiologics for fusion have evolved over time.2 Figure 2 presents a timeline for all products that are orthobiologics for bony fusion, regardless of spinal indication; the nonspinal products are presented to paint a clear historical picture to avoid implying off-label use. In all cases, the drugs have consisted of recombinant/engineered proteins and peptides and the devices were carriers for drug delivery and release. In some cases, additional implants (which would normally be regulated as class II devices cleared through 510(k)) were included in the product, such as threaded cages in the case of Infuse. In some cases, additional implants were indicated on the label but were not formally a part of the product, such as osteosynthesis hardware for Augment.

In the early 2000s, the orthobiologic segment began inauspiciously with two humanitarian use exemptions for osteogenic protein-1 (OP-1) because the scientific evidence supported only partial regulatory success, which resulted in a commercial failure; a subsequent attempt to revitalize OP-1 for spinal indications also failed at the
panel level in 2009. Despite the commercial failure of OP-1, this era was defined by the approval of Infuse for ALIF with a threaded cage in 2002 (and tibial osteosynthesis in 2004). A decade of unprecedented commercial success for Infuse resulted from explosive growth in clinical use, most of it off-label for numerous spinal indications.

The first signs of a backlash against Infuse came in 2010 with a contentious meeting of the orthopaedic devices panel for Amplify, the planned successor to Infuse. This resulted in a nonapprovable letter in 2011 that ultimately led to the demise of the program. Ironically, Amplify could be seen as an attempt to right much of the off-label use of Infuse, by producing a product designed for posterolateral fusion. Although some of the problems identified at the panel level were unique to Amplify, the popular response may have been partially a hangover from the exuberant adoption of Infuse and the perception of excess use associated with it. As shown in Figure 2, subsequent controversy and attempts to ameliorate it followed, including a special issue of The Spine Journal, a report from the Senate Finance Committee, the naming of the YODA centers, and their publications in the Annals of Internal Medicine.

At the time of this writing, Infuse might now be seen as a still successful commercial program that has been rehabilitated clinically, having withstood controversy, with its indications and its risk-benefit profile clarified over time. But the history of this product shaped the environment for contemporaneous and future products. During the height of the Infuse controversy, Biomimetic brought Augment to the FDA. After another contentious panel discussion, they received a nonapprovable letter. The company was acquired and the product ultimately received approval for hindfoot and ankle fusion. But the additional years of study and investment highlighted how risk and complexity of orthobiologics. Subsequently, iFactor was approved for a spinal indication, but for anterior cervical diskectomy and fusion, which is perceived as clinically less challenging and commercially less rewarding. Because of this, it is important to focus on lumbar fusion, the most demanding application.

Since then, approvals for orthobiologics have included label expansions for Infuse for use with more modern cages. As of the time of this writing, no major new product or indication has achieved approval in this space. It is known in the industry that many sponsors are studying new products and indications; however, making this statement precise is difficult. The presence of IDE and investigational new drug trials is not a routine matter of public record, and their disclosure requires a Freedom of Information Act filing. Although IDE and IND trials appear in, that database contains many trials of unclear provenance and activity. Nevertheless, the design considerations for these new products and indications, especially for TLIF, are discussed in detail in the next sections.


To best understand the contemporary design considerations for biologics for spinal fusion, it is essential to understand the regulatory paradigms and the history of fusion trials and products to date.2 Of note, this includes orthobiologic products for fusion, but also motion preservation products with fusion arms as a control. The goal is to seek a consensus or guideposts for typical trial designs. Although future products may follow different regulatory
and commercial paradigms, the successful regulation and commercialization of such products is sufficiently complex and risky that one must clearly understand past and present products, even if planned products differ. What emerges from the study of successful and failed products is something close to a consensus on trial design for these prior studies. Even with some variation, many trial design features have become standard.7

Table 1 presents new lumbar spinal devices since 2000 that were either class III devices that received premarket approval or went to the Orthopaedic Devices Panel regardless of subsequent approval; devices that were nonapprovable but did not go to panel or are still undergoing IDE trials are not publicly reportable. Table 2 presents those trials from Table 1 with fusion arms, including both orthobiologic products for fusion and motion preservation products with a fusion control arm.1

Tables 3, 4 and 5 provide a synopsis of the major design elements of lumbar fusion trials, highlighting the similarities while accentuating the points of variation. All trials are two-arm randomized noninferiority trials to an active surgical control with 2-year final follow-up of a composite primary end point (Table 3). All were followed up at multiple interim timepoints prior to 2 years and subsequently out to 5 years as a condition of approval for a postapproval study. Most were asymmetrically randomized with the most common ratio being 2:1 experimental to control. All were analyzed with a noninferiority margin of 10%, although some were designed and approved with a less-stringent margin but were analyzed with 10% at the FDA’s direction.1

Only gold members can continue reading. Log In or Register to continue

Oct 25, 2023 | Posted by in ORTHOPEDIC | Comments Off on Biologic Considerations for Clinical Study Design: Fusion

Full access? Get Clinical Tree

Get Clinical Tree app for offline access