Biologic Considerations for Clinical Study Design: Cartilage Repair

James L. Cook, DVM, PhD

David P. Trofa, MD

Clayton W. Nuelle, MD, FAAOS

Clark T. Hung, PhD

Dr. Cook or an immediate family member has received royalties from Arthrex, Inc. and Musculoskeletal Transplant Foundation; serves as a paid consultant to or is an employee of Arthrex, Inc. and Trupanion; has received research or institutional support from Arthrex, Inc., Collagen Matrix Inc., DePuy, a Johnson & Johnson Company, Musculoskeletal Transplant Foundation, National Institutes of Health (NIAMS & NICHD), Purina, Regenosine, SITES Medical, and U.S. Department of Defense; and serves as a board member, owner, officer, or committee member of the Midwest Transplant Network and the Musculoskeletal Transplant Foundation. Dr. Nuelle or an immediate family member is a member of a speakers’ bureau or has made paid presentations on behalf of Arthrex, Inc. and Vericel, Inc.; serves as a paid consultant to or is an employee of Guidepoint Consulting; has received nonincome support (such as equipment or services), commercially derived honoraria, or other non-research-related funding (such as paid travel) from AO Foundation; and serves as a board member, owner, officer, or committee member of the American Academy of Orthopaedic Surgeons, the American Orthopaedic Society for Sports Medicine, and the Arthroscopy Association of North America. Dr. Hung or an immediate family member has received royalties from Allosource and Musculoskeletal Transplant Foundation and has received research or institutional support from Musculoskeletal Transplant Foundation, National Institutes of Health, Orthopedic Science & Research Foundation and Department of Defense. Neither Dr. Trofa nor any immediate family member has received anything of value from or has stock or stock options held in a commercial company or institution related directly or indirectly to the subject of this chapter.

INTRODUCTION

One of the greatest challenges faced by surgeons specializing in complex knee disorders is determining the optimal approach to manage articular cartilage lesions. This challenge is complicated by the fact that there is a paucity of high-level evidence available in the literature to help guide patient-specific and lesion-specific decision making. This is problematic given that an estimated 200,000 to 300,000 procedures are performed annually in the United States for symptomatic chondral and osteochondral defects.¹ Biologic options available to manage these lesions include arthroscopic débridement, abrasion arthroplasty, marrow stimulation procedures that may be augmented with various orthobiologic treatment strategies, osteochondral autograft transfer (OAT), osteochondral allograft (OCA) transplantation, and cell-based and matrix-based techniques. Although extensive preclinical and single-cohort outcomes studies support relative indications and potential benefits for each of these treatment strategies, robust head-to-head comparisons are unfortunately rare, usually underpowered, and statistically fragile.² Robust clinical trials to assess current and future options to manage articular cartilage lesions are needed to address this deficiency.

BRIEF HISTORY OF CLINICAL STUDY DESIGN IN ARTICULAR CARTILAGE REPAIR

Clinical trials are categorized into one of four basic stages:³

Feasibility—first-in-human application, primarily focused on safety
Clinical research—safety data are further developed and efficacy assessments are included for well-defined patient groups under close supervision of investigator/inventor
Validation—efficacy evaluations expanded to include multiple clinical sites
Acceptance—postmarket surveillance to monitor long-term outcomes and adverse events

Accordingly, clinical trial study design will reflect the intended purpose. With respect to defining the level of rigor of a clinical trial, level of evidence for clinical application is considered to graduate from case series to single-cohort studies, case-control or comparison cohort studies, and finally randomized controlled trials (RCTs), systematic reviews, and meta-analyses.⁴,⁵ However, it is important to also consider other key aspects of the experimental design such as study population, sample size, outcome measures, and duration when determining strength and generalizability of the data. In addition, the level of evidence for systematic reviews and meta-analyses can only be considered to be as robust as the studies included in the analyses.

Despite their importance, RCTs represent 3% of orthopaedic literature.⁶,⁷ With respect to recent level I or II clinical studies and their chosen comparator groups, 6 studies have compared OATs with autologous chondrocyte implantation (ACI),⁸,⁹,¹⁰,¹¹,¹²,¹³ 1 has compared OATs with matrix-induced ACI,¹⁴ 8 have compared OAT with microfracture,¹³,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹,²⁰,²¹ and 12 have compared ACI with microfracture.²² A recently published article by Saltzman et al²³ evaluated the methodology used among level I and II studies and reported a significant heterogeneity in methodology including design, follow-up, and outcome measurements. The heterogeneity in such investigations is well demonstrated in Table 1.

TABLE 1 PICO Table for Cited Studies With Functional Outcomes

Population	Intervention	Comparisons	Outcome	Studies
Patients (n = 100) with symptomatic cartilage defects in the knee (mean size, 4.7 cm²)	Randomized to ACI or OAT-mosaicplasty (OAT-M)	Modified Cincinnati and Stanmore scores; 1-year second-look arthroscopy; 10-year treatment survivorship	Improved functional scores (88% versus 69%), postoperative arthroscopic evaluations, and failure rates (17% versus 55%) in ACI versus OAT-M	Bentley et al 2003⁸, Bentley et al 2012⁹
Patients (n = 55) with isolated femoral condyle defects (2.5 to 7.5 cm²)	Randomized to MACI (agarose-alginate scaffold) or OAT-M	IKDC score; 2 year histology (O’Driscoll score); adverse events	Significantly improved IKDC scores (81.5 versus 73.7), improved histologic characteristics, and lower adverse events in the mosaicplasty group versus MACI group	Clave et al 2016¹⁰
Patients (n = 23) with full-thickness chondral defects in the knee (mean, 1.9 cm²)	Randomized to ACI or OAT-M	IKDC and Lysholm Knee Scale	No differences in average Lysholm or IKDC scales	Dozin et al 2005¹¹
Patients (n = 40) with isolated femoral condyle defects (mean, 3.75 cm²)	Randomized to ACI or OAT	2 year Lysholm, Tegner activity scale, and Meyers rating scores; histology	No difference in Tegner and Meyers scores. Improved Lysholm score after OATs versus ACI; histology demonstrated fibrocartilage repair tissue for ACI versus maintained hyaline cartilage for OAT	Horas et al 2003¹²
Patients (n = 70) with isolated cartilage defects in the knee (mean, 1.0 to 4.0 cm²).	Microfracture, OAT, ACI cohorts	Minimum of 3-year follow-up; Lysholm, Tegner, and HSS scores; MRI using modified Outerbridge scale; 12 to 18 months second-look arthroscopy with ICRS grading system	No significant differences in functional scores, Outerbridge scale using MRI, or arthroscopy findings	Lim et al 2012¹³
Athletes (n = 60) with symptomatic articular cartilage lesions in the knee	Randomized to OAT or microfracture	3-year HSS and ICRS scores, 10-year ICRS, and Tegner scores; return to sports and failure rates; 1-year histologic biopsy; MRI	OAT demonstrated significantly better functional outcomes scores and return to sports rates and activity levels 3 and 10 years postoperatively. Histologic examination demonstrated normal hyaline cartilage in all patients undergoing OAT versus 57% of microfracture samples demonstrating fibrocartilage. There was a higher failure rate at 10 years in the microfracture (38%) versus OAT (14%) cohort. Improved chondral surface filling was found on MRI in the OAT group compared with microfracture	Gudas et al 2005¹⁵, Gudas et al 2006¹⁶, Gudas et al 2012¹⁸
Young (aged 12 to 18 months) patients (n = 50) with OCD of the femoral condyle (mean, 3.2 cm²)	Randomized to OAT or microfracture	Follow-up of 3 to 6 years, ICRS scores, failure rates, 18-month MRI	Good to excellent outcomes in 83% OAT versus 63% microfracture; significantly higher failure rate for microfracture (41% versus zero); MRI good to excellent for 91% OAT versus 56% microfracture	Gudas et al 2009¹⁷
Patients (n = 102) with ACL rupture and articular cartilage damage of the medial femoral condyle	Randomized to OAT, microfracture, or débridement with ACLR and matched control (ACLR with intact cartilage)	3-year follow-up for IKDC, Tegner, and clinical assessments	IKDC: Control > OAT > microfracture or débridement; lower Tegner scores identified for the microfracture and débridement cohorts versus control and OAT group	Gudas et al 2013¹⁹
Patients (n = 40) with 1 or 2 symptomatic focal full-thickness cartilage defects (2 to 6 cm²) on the femoral condyles or trochlea	Randomized to OAT-M or microfracture	15-year follow-up for Lysholm score	Lysholm score significantly better for OAT-M at 1, 5, 10, and 15 years	Solheim et al 2018²⁰
Patients (n = 25) with full-thickness chondral lesion of the distal femur (2.0 to 6.0 cm²)	Randomized to OAT-M or microfracture	Follow-up of 5 to 11 years for Lysholm score, KOOS, and isokinetic muscle strength	No significant differences in measured outcomes	Ulstein et al 2014²¹
ACI = autologous chondrocyte implantation, ACL = anterior cruciate ligament, ACLR = anterior cruciate ligament reconstruction, HSS = Hospital for Special Surgery, ICRS = International Cartilage Repair Society, IKDC = International Knee Documentation Committee, KOOS = Knee Injury and Osteoarthritis Outcome Score, MACI = matrix-induced ACI, OAT = osteochondral autograft transfer, OCD = osteochondritis dissecans

DESIGN CONSIDERATIONS AND CHALLENGES FOR RCTs

FDA Regulatory Considerations (Device, Biologic, or HCT/P Designation)

In the United States, cartilage repair and restoration therapies can be used clinically as medical devices; biologics; drugs; or human cells, tissues, or cellular or tissue-based products (HCT/P). The specific classification in combination with the determination regarding validity greatly affects the regulatory pathway for bringing the product to market. Any product “containing or consisting of human cells or tissues that are intended for implantation, transplantation, infusion, or transfer into a human recipient”²⁴ that meets HCT/P criteria, including minimal manipulation and homologous use, is not subject to premarket FDA approval or clearance. Any cartilage repair or restoration device that does not meet FDA HCT/P criteria is subject to premarket FDA clearance, approval, or granting:

Clearance: This is applicable to devices that qualify for and complete the 510(k) pathway, such that the FDA reviews and clears them for clinical use.
Approval: Class III medical devices must be submitted for premarket approval or Humanitarian Device Exemption and then complete the rigorous review and approval process for clinical use.
Granted: Class I and II devices may be eligible for the de novo pathway if they are considered novel such that they are not listed in the standard FDA classifications. The device must also be considered low or moderate risk to be granted for clinical use by the FDA.

Cartilage repair or restoration strategies that are considered biologics or drugs, or combination products, must go through the Center for Biologics Evaluation and Research development and approval process. This process may involve 510(k), premarket approval, Investigational New Drug, Biologics License Application, Expanded Access, or New Drug Application pathways (https://www.fda.gov/vaccines-blood-biologics/development-approval-process-cber).

Completing RCTs required by the FDA to establish safety and efficacy for novel and promising approaches to cartilage repair and restoration has been a challenge based on major barriers including costs, enrollment feasibility to meet needed sample size, and required controls.²⁵

Inclusion/Exclusion Criteria

Patient-related criteria for cartilage repair interventions that often result in RCT screening failures and prolonged enrollment timelines include patient age, nature of the lesion, comorbidities, activity level, and willingness to be randomized.²⁶ Examples of typical criteria for patient inclusion and exclusion for cartilage repair studies gleaned from a sampling of studies on ClinicalTrials.gov are discussed in Tables 2 and 3.

Control Arm Selection for RCTs

The control arm is intended to include patients who are representative of the attributes of the experimental group and account for a placebo effect and to reduce selection bias where patient enrollment is skewed to individuals who are thought to respond more positively to the intervention. If clinical equipoise exists, a genuine uncertainty about the best treatment, and with preclinical data justifying human experimentation, patients can ethically be randomly assigned to an intervention to reduce the effects
that could influence the outcome. The results observed in an RCT are more likely to be the consequence of the intervention.⁷

TABLE 2 Sample Inclusion Criteria

Symptomatic focal or multifocal defects, often specified as contained defects

Symptomatic lesion grading International Cartilage Repair Society 3 or 4 (for osteochondral product, otherwise exclusion criteria)

Symptomatic total treatable area (eg, 1 to 7 cm²)

Symptomatic, single or multiple, full-thickness cartilage defects of the knee with or without bone involvement

Knee pain and/or nonsurgical treatment (anti-inflammatory drugs) for a defined period (2 to 4 months)

Knee Injury and Osteoarthritis Outcome Score pain score at baseline is not less than 30 and not more than 65

Age range: Typically 18 to 55 years

Must be willing and able to undergo MRI

Must be physically and mentally willing and able to comply with postoperative rehabilitation protocol and scheduled clinical and diagnostic imaging visits

Willingness to incorporate normal standard of care or add on (nonstandard of care)

Documented informed consent

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