A multilayer biomaterial for osteochondral regeneration shows superiority vs microfractures for the treatment of osteochondral lesions in a multicentre randomized trial at 2 years

Chondral and osteochondral lesions are debilitating conditions that, if not properly treated, may lead to the development of osteoarthritis, with a high impact on patients and society, both in terms of healthcare and workforce loss [14, 15]. The greater emphasis on physical activity in all age groups is responsible for the growing incidence of these lesions, and concomitantly, patient expectations about function recovery have risen as well. Thus, in the attempt to fulfil patients’ expectations and successfully treat this pathology, several techniques have been developed over the years [26].

The available surgical options range from simple strategies such as microfracture (MF) or subchondral drilling to the more ambitious regenerative approaches. MF is a bone marrow stimulation (BMS) technique aimed at recruiting bone marrow cells by creating a communication between cartilage lesions and subchondral bone, thus allowing stem cells to migrate to the fibrin clot of the defect and form a fibrocartilaginous repair tissue [36]. Regenerative procedures aim at recreating a hyaline-like tissue as similar as possible to the physiological one, and they are emerging as a potential therapeutic option also in cases of large lesions, where other procedures are less indicated [1, 32]. Several materials have been developed in the attempt to meet the requirements of cartilage regeneration. The rationale for using a scaffold is to offer a temporary 3-dimensional structure of biodegradable polymers to mimic cartilage architecture and favour cell growth [24]. However, despite the increasing number of publications every year confirming good outcomes also at midterm evaluations [1], the superiority of this technique versus BMS has not been clearly proven, and MF is still considered the gold standard that sets the reference point to measure the potential of new procedures [8].

The increasing awareness on the role of subchondral bone in the etiopathology of articular surface lesions may explain the limits of the current regenerative procedures, which have been developed to target only the chondral layer [9, 24]. Subchondral bone may be involved in the pathological process not only primarily, such as in osteochondritis dissecans (OCD), osteonecrosis and trauma, but also secondarily in degenerative cartilage lesions. In fact, even focal chondral defects, if left untreated, may increase in size over time and cause changes of the underlying subchondral bone [25, 33]. Thus, a surgical approach for both cartilage and bone reconstruction that would address articular surface lesions by restoring the properties of the entire osteochondral unit should be considered. Among the many different multilayer scaffolds developed to reproduce both bone and cartilage [35], only a few have been investigated in clinical studies [22]. While safety and promising results have been suggested [21, 22], there are no trials proving the real potential of the osteochondral regenerative approach compared to the gold standard in clinical practice.

The aim of this study is to assess the benefit provided by a nanostructured collagen-hydroxyapatite (coll-HA) multilayer scaffold in a randomized controlled trial. The hypothesis is that this comparative evaluation will demonstrate safety and a superior clinical benefit of this osteochondral scaffold.

The study was performed in compliance with the protocol and in accordance with the Declaration of Helsinki, the International Conference of Harmonization Guidelines for Good Clinical Practice, standards from International Organization for Standardization (ISO) and valid international and national regulations. Patients referring to specialized orthopaedic centres in the following countries were included: Italy, Sweden, Belgium, Switzerland, Austria, Germany, Norway, Poland, South Africa. From 2011 to 2013, 145 patients affected by chondral and osteochondral knee lesions were screened; 124 patients were randomized (safety population).

The selected ITT population consisted of 118 patients, while the PP population resulted in 100 patients, due to protocol violators and dropouts (Fig. 1). Study and BMS group presented, respectively, a mean age of 34.0 ± 10.9 and 35.2 ± 10.2 years old; a male/female ratio of 36/15 and 31/18; a body mass index of 25.6 ± 3.3 and 25.2 ± 3.2 and a mean defect size of 3.4 ± 1.5 and 3.5 ± 1.6 cm². Further demographics data have been reported in detail in Table 1. These patients were evaluated for up to 2 years of follow-up with both clinical and imaging examination.

A statistically significant improvement of all clinical scores was obtained from baseline to 2-year follow-up in both treatment groups.

The main finding of this study is that comparable results were found in the overall population, whereas the coll-HA osteochondral scaffold offered significant clinical better results compared to BMS in the treatment of osteochondral lesions. BMS techniques confirm to be a treatment option for purely chondral lesions, but offer worse results for more complex lesions, where the biomimetic scaffold, designed to address also the subchondral area, showed to be a more suitable therapeutic solution.

The reason for this selected superiority lies in the rational of the treatments themselves, and the mechanisms leading to these different results reflect previous literature findings. Smaller chondral lesions are those benefiting more from BMS, whose potential to achieve good results over time finds the main limitation in the lesion size [13]. In fact, the repair tissue response is often fibrous and cannot address successfully larger defects [31]. This, together with the damage produced by the technique at the subchondral bone level [30], may explain the inferior results obtained with BMS in more complex lesions. On the other hand, the composite scaffold has been developed to mimic the biochemical and biophysical properties of different layers of the native osteochondral structure [37] to favour the repair of the entire osteochondral unit [34]. The properties of this scaffold have been shown to produce good results in preclinical models [23, 34]. The scaffold ability to induce an in situ repair through cells coming from the surrounding bone marrow in the animal model [18] allowed to introduce it in the clinical practice as a cell-free approach.

Safety and preliminary experiences with human application have been documented in case series published by different groups. Delcogliano et al. obtained an encouraging outcome at 24-month follow-up on 19 patients affected by large-sized defects [5], confirmed by Verdonk et al. evaluating at the same follow-up 38 patients affected by osteochondral lesions [39]. These positive findings were also confirmed by Berruto et al. evaluating 49 patients in a multicentre study which supported the potential of this scaffold also for the treatment of large osteochondral lesions [2]. This biomimetic scaffold was also tested as salvage procedure in young patients for the treatment of unicompartmental OA, where subchondral cysts, oedema and stiffening may be addressed by the osteochondral scaffold, an attempt to provide an alternative solution to metal resurfacing that showed promising short-term results [12, 27]. Finally, some encouraging findings have also been documented in the only available comparative evaluation versus a chondral scaffold for the treatment of complex lesions, with a better outcome in the osteochondral treatment group [11].

This multicentre study confirms safety and potential of the osteochondral scaffold, showing for the first time, with a randomized study design, the improvement provided compared to BMS when targeting complex osteochondral knee lesions. This benefit is supported by the clinical outcome documented through a subjective functional evaluation. More adverse events were reported in the study group. However, most of them were minor post-op symptoms that spontaneously resolved within 45 days. The higher rate of adverse events may be explained by the different invasiveness of coll-HA and BMS surgical approaches, i.e. arthrotomic versus arthroscopic, which have been shown to affect the early recovery of the patients [19]. Summarizing, this trial shows how to address the entire osteochondral unit and the usefulness of this biomimetic implant.

Nonetheless, the imaging evaluation confirms some concerns previously risen in the literature on the slow restoration of the subchondral bone area. Even though the slow bone mineralization did not correlate with a lower clinical outcome, their significance and impact in terms of tissue regeneration and clinical outcome remains to be determined. Suboptimal imaging findings, not correlating with clinical results, have been shown at short-term evaluation by Verdonk et al. and Christensen et al. in particular in terms of subchondral bone formation [2, 4]. On the other hand, Brix et al. underlined a limited cartilage quality but a successful osteoconduction [3]. While further studies need to explore the clinical impact of the observed alterations, the only available midterm evaluation of this osteochondral scaffold shows stable clinical results, as well as a significant MRI improvement between 2 and 5 years of follow-up [20]. This may suggest that the maturation required by an osteochondral scaffold may be longer than what expected for chondral scaffolds (as confirmed by this RCT). This is probably due to the greater complexity of a scaffold aiming at restoring the entire osteochondral unit and also by the bigger size of the tissues integrating and regenerating, being not just a thin layer but rather a thick layer of articular surface. Moreover, controversial findings may be also partially explained by MRI itself, which in a recent meta-analysis was shown to fail a correlation in the majority of the parameters observed, with only 28% of the studies presenting a correlation between MRI and treatment outcomes [6]. Since the scores available for cartilage studies were designed for the evaluation of cartilage treatments, it is easy to understand that it is even more difficult to expect a correlation in case of an osteochondral treatment evaluation. Therefore, caution should be recommended when interpreting MRI findings and greater importance should be placed on the patient clinical evaluation, until more reliable imaging instruments will be available.

This study presents some limitations. The number of originally planned patients has not been reached, which could have hampered the possibility to detect further significant differences. Nonetheless, the study groups allowed to document the superiority of the osteochondral scaffold in deep osteochondral lesions, which is in line with previous preliminary literature findings [11] and what one may expect from an osteochondral biomimetic implant. Patients were rather heterogeneous in terms of age, lesion dimension, location and aetiology, and one-third required combined surgeries. This, on the one hand, may represent a confounding factor, but it also allows to have a study population more representative of the general population commonly requiring cartilage treatments, which increases the generalizability of these study findings. Despite its heterogeneity, a tendency to treat small lesions was observed in this study, and this might have entailed a bias in favour of the BMS group, as it is possible that coll-HA yields more favourable results in bigger lesions, while MF is more suitable for smaller lesions. Thus, the superiority found for complex lesions regardless of this limitation further confirms the potential of this scaffold. The study presents also other limitations: the 2-year follow-up does not allow to detect the outcome decline of BMS shown by previous literature [8, 13], neither to understand the clinical meaning and evolution of the abnormal MRI findings observed in the scaffold implant area, which warrants further evaluation at longer follow-up. To this regard, the only study currently available at midterm follow-up showed stable clinical outcome and a trend towards slow imaging improvement, even though MRI abnormalities persisted even at 5-year follow-up [20]. The understanding of the tissue and clinical evolution over time is therefore particularly important because it could allow to better identify both treatment potential and most suitable indications. In fact, many questions remain still open, athletes tend to present good early results with several procedures, thus the benefit shown in this study should be confirmed by longer-term evaluations, and it will be important to understand whether bone repair, which may explain the better results in osteochondral lesions, is coupled by sufficient cartilage repair in order to allow good and stable results overtime in young active patients.

Future studies should therefore focus on osteochondral healing mechanisms and on further improvement of the regenerative potential of this scaffold. To this regard, technical advancements have been already identified, such as the possibility to enhance the scaffold fixation with the use of fibrin glue or to use thinner scaffolds for shallower lesions, thus prompting the possibility to further increase the outcome of the scaffold [10]. Moreover, while preclinical evaluations did not show any benefit in chondrocyte seeding [18], further studies should explore whether other augmentation strategies could improve the osteochondral regenerative potential either with chondro/osteoinductive molecules or taking advantages from the combination with other cell sources to promote a faster healing with a higher quality tissue and in the end better clinical results.

While the clinical applicability of the aforementioned augmentation procedures may be affected by economic and regulatory requirements, this study highlighted the safety and potential of the biomimetic coll-HA implant showing, for the first time with a randomized trial, its superiority versus BMS approach in the treatment of osteochondral defects. Thus, this procedure can be considered as an option to address the damaged articular surface when subchondral bone is also involved, with the advantage of being an off-the-shelf product requiring a one-step surgery and offering good results for the treatment of the osteochondral surface.

This RCT showed safety and an overall positive clinical outcome provided by this coll-HA biomaterial. While comparable results were found in the overall population, the osteochondral scaffold offered significant clinical better results compared to BMS in the treatment of osteochondral lesions. BMS techniques confirm to be a treatment option for purely chondral lesions, but offer worse results for more complex lesions, where the biomimetic scaffold, designed to address also the subchondral area, showed to be a more suitable therapeutic solution.