Keywords osteochondral allografts - immunology - transplantation - patient outcomes
Fresh osteochondral allograft (OCA) transplantation is a treatment option for large
articular cartilage lesions, particularly in young, active patients who may not be
ideal candidates for nonsurgical treatments or arthroplasty.[1 ] When successful, OCA transplantation can relieve pain and restore joint function
in patients with symptomatic articular defects in the knee, hip, ankle, and shoulder.[2 ]
[3 ]
[4 ]
[5 ]
[6 ]
[7 ]
[8 ]
[9 ] Currently, most OCA transplantations are performed to treat large (>2 cm2 ), full-thickness symptomatic chondral or osteochondral lesions in the knee. While
recent technological advances have been associated with increased OCA availability
and improved clinical success, the best current evidence supports 5- to 15-year functional
OCA survival rates of 68 to 75%.[2 ]
[3 ]
[4 ]
[5 ]
[6 ]
[7 ]
[8 ]
[9 ]
[10 ]
[11 ]
[12 ]
[13 ]
[14 ]
[15 ]
[16 ]
[17 ]
[18 ]
[19 ]
[20 ] As such, all components of OCA transplantation need to be further optimized to improve
patient outcomes and meet the increasing clinical demand for this important treatment
option for the growing number of patients affected by these complex joint problems.
Recognized factors contributing to functional graft survival after OCA transplantation
include OCA-related factors (storage time, preservation methods, chondrocyte viability
at the time of implantation, and preimplantation preparation and surgical techniques)
and recipient-related variables (age, sex, body mass index [BMI], nicotine use, joint
pathology, and patient adherence with prescribed postoperative restrictions and rehabilitation
protocols) that influence cartilage integrity, rate and extent of allograft osteointegration,
and joint health.[2 ]
[3 ]
[4 ]
[5 ]
[6 ]
[7 ]
[8 ]
[9 ]
[10 ]
[11 ]
[12 ]
[13 ]
[14 ]
[15 ]
[16 ]
[17 ]
[18 ]
[19 ]
[20 ] With recent advances in tissue preservation, transplantation techniques, and patient
management tools and strategies that have reportedly mitigated related modes of failure,
prolonged and incomplete OCA osteointegration has become a primary mechanism for OCA
treatment failure.[18 ]
[19 ]
[20 ]
[21 ]
[22 ] Preimplantation OCA preparation techniques, including subchondral bone drilling,
thorough irrigation, and autogenous bone marrow aspirate concentrate saturation, may
dampen immune responses and improve OCA osteointegration.[11 ]
[14 ]
[18 ]
[19 ] Still, insufficient OCA osteointegration remains a significant cause of treatment
failure, and studies suggest that recipient immunological responses may influence
the process of creeping substitution and contribute to undesirable outcomes.[21 ]
[22 ]
[23 ]
[24 ]
[25 ]
[26 ]
[27 ]
[28 ]
[29 ]
[30 ]
[31 ]
[32 ]
[33 ]
[34 ]
[35 ]
[36 ]
Unlike other forms of allogeneic transplantation, fresh OCAs are considered “immunoprivileged”
such that human leukocyte antigen or ABO blood group donor–recipient matching is not
required for transplantation.[22 ]
[23 ]
[24 ]
[25 ]
[26 ]
[27 ]
[28 ]
[29 ]
[30 ]
[31 ]
[32 ]
[33 ]
[34 ]
[35 ]
[36 ] Further, OCA transplant patients receive no immunosuppressive medications. This
approach has proven safe and effective based on the lack of immediate graft rejection
responses and the documented functional graft survival rates. However, short-term
treatment failures still occur, which suggests that immune system processes other
than those associated with rejection, or “subrejection responses,” may directly or
indirectly influence OCA transplantation outcomes.[22 ] These responses and their impacts on OCA transplantation treatment failures have
not been fully characterized. This gap in knowledge poses a limitation to further
optimization of OCA transplantation outcomes. Therefore, the objective of this study
was to further characterize the potential immune system contributions to OCA transplantation
treatment failures by analyzing donor–recipient ABO and Rh-factor mismatches as well
as histological and immunohistochemical assessments of transplanted OCA tissues recovered
from revision surgeries. The study was designed to test the hypothesis that treatment
failures would be associated with lymphocyte phenotype profiles, potentially associated
with cell-mediated subrejection immune responses independent from donor–recipient
blood type mismatches.
Methods
With institutional review board (IRB) approval and documented informed consent, a
defined subset of age-, BMI-, and joint-matched OCA transplant recipients enrolled
in our institution's lifelong OCA patient outcomes registry such that there was a
2:1 ratio of successful versus failed outcomes was analyzed. Cases performed by one
of four surgeons were included based on the following criteria:
Complete donor and recipient ABO and Rh factor data were available. (Donor blood type
data were obtained from American Association of Tissue Banks [AATB]-accredited tissue
banks [MTF Biologics, Joint Restoration Foundation, RTI Donor Services], and recipient
blood type data were retrieved from the patient's electronic medical records.)
OCA transplantation was performed according to our institution's standard of care
for preimplantation preparation, transplantation techniques, and prescribed postoperative
management, as previously described.[18 ]
[19 ]
Treatment failure data were documented. (Patients who required OCA revision surgery
or artificial arthroplasty at any time point were defined as treatment failures. A
successful outcome was defined as the documented absence of any revision or arthroplasty
surgery for a minimum of 2 years following the index OCA transplantation surgery.)
Cases not fulfilling all of the inclusion criteria for this analysis were excluded.
Osteochondral Allograft Tissue Recovery
With IRB approval and documented informed consent, resected osteochondral tissues
that would otherwise be discarded after standard-of-care revision surgeries to treat
OCA treatment failures in the knee were recovered from patients. For controls, osteochondral
tissues from unused portions of allografts that would otherwise be discarded after
standard-of-care OCA transplantation surgeries were recovered, processed, and analyzed
in the same ways as resected tissues.
Histological Assessments
Sections of OCA tissues were fixed in 10% neutral-buffered formalin for 1 week and
then placed in 10% ethylenediaminetetraacetic decalcifying solution until the bone
was suitable for sectioning (∼3 weeks). After decalcification, specimens were routinely
processed, sectioned (5 µm), and stained with hematoxylin and eosin (H&E). Subjective
histological assessments were performed on H&E-stained, and toluidine blue-stained
sections by a pathologist blinded to patient demographics, outcomes data, and tissue
source. Histological immune response was defined by the aggregation of lymphocytes
and plasma cells surrounding small blood vessels in the OCA subchondral bone and neighboring
soft tissues. OCA tissues with evidence for this immune response were then processed
for immunohistochemical assessments.
Immunohistochemistry
Antigen markers CD3, CD4, CD8, and CD20 were chosen to characterize lymphocytes (T-cell
marker, cytotoxic T-lymphocytes, and B-cell marker, respectively) to provide a broad
view of immune cell populations and allow for subjective assessment of relevant immune
cell mechanisms. Paraffin-embedded specimens were deparaffinized, rehydrated, and
quenched of endogenous peroxidase. Antigen retrieval was performed with Diva Decloaker
10X (CD3, CD4, and CD8) or Borg Decloaker 1X (CD20). Background Sniper was applied,
and primary antibodies were incubated: CD3 (Dako rabbit polyclonal; 1:150 dilution)
for 45 minutes, CD4 (Abcam rabbit monoclonal; 1:750 dilution) for 30 minutes, CD8
(LifeSpan BioSciences mouse monoclonal; 1:1000 dilution) for 30 minutes, and CD20
(Laboratory Vision rabbit polyclonal; 1:300 dilution) for 40 minutes, all at room
temperature. To confirm cross-reactivity, appropriate positive controls were included
in each immunostaining protocol with human lymph node and tonsil tissue. Conjugated
goat anti-mouse secondary antibody (MACH 2 Universal HRP-Polymer Detection; CD8) or
conjugated goat anti-rabbit secondary antibody (Dako EnVision+ System- HRP; CD3, CD4,
and CD20) and Romulin AEC Red Chromogen were used for detection of primary antibodies,
and sections were counterstained with CT hematoxylin. Immunohistochemistry staining
for NCR-1/NKP46 (natural killer cells) was attempted with multiple different primary
antibodies. Either no to minimal staining or nonspecific staining was detected in
control and experimental tissues; therefore, the results were excluded from this study.
Statistical Analysis
A subset of age-, BMI-, and joint-matched OCA transplant recipients were defined from
the registry enrollees such that there was a 2:1 ratio of documented successful versus
failed outcomes for analyses based on the case–control-matched experimental design.
R for Statistical Computing software was used for all statistical analyses. Descriptive
statistics were calculated to determine means, medians, ranges, and percentages. All
case–control comparisons were made using the Fisher's exact tests to assess for significant
differences in proportions. When significant differences in proportions were noted,
odds ratios were calculated. A p -value of ≤0.05 was used to define statistical significance.
Results
Study Population and Osteochondral Allograft Tissue Recovery
A subset of registry patients meeting all inclusion criteria was selected from 414
OCA transplant recipients enrolled at the time of study initiation ([Fig. 1 ]; [Table 1 ]). All treatment failures meeting inclusion criteria were first identified, which
yielded 33 patients (15 female; mean age = 38.1, range = 16–60 years; mean BMI = 28.8,
range = 18–37 kg/m2 ; 30 knees, 2 hips, 1 ankle). Then, an age-, BMI-, and joint-matched cohort with documented
successful outcomes and meeting inclusion criteria were identified, which yielded
70 patients, for a total of 103 patients (45 female; mean age = 36.9, range = 16–60
years; mean BMI = 27.5, range = 17–37 kg/m2 ; 96 knees, 4 hips, 3 ankles) included for analyses. Of the 33 treatment failures,
adequate OCA tissues were recovered from 18 patients for immune response assessments.
Portions of seven OCAs not used during standard-of-care transplant surgeries were
recovered as controls.
Fig. 1 CONSORT flow chart for study population and tissue recovery allocation and analyses.
OCA, osteochondral allograft.
Table 1
Key variables assessed for osteochondral allograft transplantation cases in treatment
success and treatment failure matched cohorts
Variable
Treatment success (N = 70)
Treatment failure (N = 33)
Age, y, mean, range
36.9, 18–58
38.1, 16–60
Sex, N (%), female
30 (42.9)
15 (45.5)
BMI, kg/m2 , mean, range
27.5, 17–35
28.8, 18–37
Previous surgeries, mean, range
2.8, 1–11
2.9, 1–9
OCAT type, N-joint, technique
66 knees (27 plug, 39 shell)
2 hips (1 plug, 1 shell)
2 ankles (2 shell)
30 knees (13 plug, 17 shell)
2 hips (1 plug, 1 shell)
1 ankle (1 shell)
Concomitant procedures, N-type
5 ligament recon
18 osteotomy
4 ligament recon
8 osteotomy
Follow-up duration, mo, mean, range
49.1, 24–70
26.2, 7–56
Abbreviations: BMI, body mass index; OCAT, osteochondral allograft transplantation.
Notes: plug = cylindrical osteochondral allografts press-fit into sockets; shell = patient-specific
custom-cut osteochondral allografts stabilized in custom-cut recipient beds using
screws, pins, or nails; concomitant procedures = autograft or allograft ligament reconstruction(s)
for ligament-related instability in the affected knee and/or distal femoral osteotomy,
high tibial osteotomy, or tibial tuberosity osteotomy in ipsilateral lower extremity
for malalignment or maltracking; follow-up duration = the time point for which functional
graft survival or nonsurvival was documented for each included case.
Treatment failure mechanisms were attributed to allograft bone necrosis and/or collapse
(n = 14, 42.4%), damage to new nontransplanted areas in the knee (n = 10, 30.3%), meniscus tear and/or extrusion (n = 8, 24.2%), or unknown reasons (n = 1, 3.0%). Treatment failures occurred at a mean and median of 26.2 and 23 months
after primary OCA transplantation, respectively, with a range of 7 to 56 months.
Blood Type Matching
No statistically significant differences in proportions for treatment success (n = 70) versus failure (n = 33) based on mismatches for ABO type (n = 73), Rh factor (n = 30), or both (n = 85) were noted ([Table 2 ]). No statistically significant differences in proportions for histological immune
response presence or absence based on mismatches for ABO type (n = 10), Rh factor (n = 2), or both (n = 10) were noted ([Table 3 ]).
Table 2
Proportions of treatment successes and failures for osteochondral allograft transplant
patients based on Rh-factor and ABO blood type matching
Rh Fisher's exact—failure
Rh
Success
Failure
p -Value = 0.352
Same
52
21
Opposite
18
12
ABO Fisher's exact—failure
ABO
Success
Failure
p -Value = 0.821
Same
21
9
Different
49
24
ABO and Rh Fisher's exact—failure
ABO and Rh
Success
Failure
p -Value = 0.412
Same
14
4
Different
56
29
Table 3
Proportions of histologic immune responses for failed osteochondral allograft transplants
based on Rh-factor and ABO blood type matching
Rh Fisher's exact—immune response
Rh
Success
Failure
p -Value = 1
Same
5
11
Opposite
1
1
ABO Fisher's exact—immune response
ABO
Success
Failure
p -Value = 0.312
Same
4
4
Different
2
8
ABO and Rh Fisher's exact—immune response
ABO and Rh
Success
Failure
p -Value = 0.312
Same
4
4
Different
2
8
Histological Assessments
No control OCA tissues were noted to include lymphocyte aggregations in the subchondral
bone or neighboring soft tissues ([Fig. 2A ]), whereas 12 (67%) of the failed OCA tissues contained lymphocyte aggregations in
the subchondral bone, suggestive of a cell-mediated immune response ([Fig. 2B ]). Interestingly, the mechanisms of failure for each of these 12 tissues involved
insufficient OCA osteointegration, characterized by subchondral bone fracture, collapse,
subsidence, and/or necrosis, whereas only 2 of the “no immune response” OCAs failed
by these mechanisms.
Fig. 2 Tissues from osteochondral allograft (OCA) revision surgeries were stained with hematoxylin
and eosin (H&E) to identify immune reactions of interest. (A ) Healthy tibial plateau OCA used as a negative control. (B ) Failed tibial plateau OCA (inclusion criteria): lymphocyte aggregation around blood
vessels. (Black arrows indicate blood vessel, scale = 100 µm).
Immunohistochemical Assessments
For the 12 OCA tissues histologically defined to have a cell-mediated immune response
and further characterized immunohistochemically, combinations of CD3 + , CD4 + , CD8 + ,
and CD20+ lymphocytes comprised the perivascular aggregations in each case ([Fig. 3 ]).
Fig. 3 Observed aggregation of lymphocytes around small blood vessels in a failed hip osteochondral
allograft. (A ) Hematoxylin and eosin (H&E) [aggregate of blue cells: lymphocytes], (B ) CD3+ [red: positive staining; blue: counterstain], (C ) CD4+ [red: positive staining; blue: counterstain], (D ) CD8+ [red: positive staining; blue: counterstain], and (E ) CD20+ [red: positive staining; blue: counterstain] stained lymphocytes. [scale = 100 µm].
Discussion
The results of this study suggest that cell-mediated subrejection immune responses
may play roles in some mechanisms of treatment failure after OCA transplant surgeries.
Interestingly, OCA transplant treatment failures and cell-mediated immune responses
in failed OCAs could not be attributed to ABO blood type or Rh-factor mismatches.
Immune responses in OCA subchondral bone recovered from failed transplants were consistently
characterized by perivascular lymphocyte aggregations comprised of CD3 + , CD4 + ,
CD8 + , and CD20+ phenotypes. Interestingly, the mechanisms of failure for OCA transplants
with these cell-mediated immune responses involved insufficient osteointegration,
characterized by subchondral bone fracture, collapse, subsidence, and/or necrosis.
Considering the results, these data support the possibility that mixed-aggregate T-
and B-cell infiltrates directly or indirectly prohibit timely and complete OCA osteointegration,
posing a threat to the functional survival of the transplanted tissues. Based on the
lack of association with blood type mismatches and the nature and location of the
lymphocyte aggregates, remaining donor bone marrow elements are the most likely culprits
for initiating this subrejection immune response.
Previous studies have also suggested the potential for subrejection immune responses
contributing to OCA failures.[24 ]
[25 ]
[26 ]
[27 ]
[28 ]
[29 ]
[30 ]
[31 ]
[32 ] Humoral and cell-mediated immune responses to osteochondral, cortical, and corticocancellous
allograft bone have been documented in preclinical and clinical studies. These responses
and related outcomes have been attributed to graft type and volume, blood type mismatches,
and sex mismatches.[22 ]
[24 ]
[25 ]
[26 ]
[27 ]
[28 ]
[29 ]
[30 ]
[31 ]
[32 ]
[33 ]
[34 ] In the present study, only fresh OCAs were considered, humoral immune responses
were not assessed, and graft volume and donor–recipient sex were not analyzed as variables.
Therefore, these initial results are limited to characterizing localized cell-mediated
immune responses in failed OCAs. Based on these data, blood type mismatching was not
considered a primary culprit for inciting subrejection immune responses associated
with treatment failures. While intact hyaline cartilage is well documented to be immunoprivileged
and OCA bone is not expected to contain viable cells, donor cell debris and proteins
are immunogenic.[23 ]
[24 ]
[25 ]
[26 ]
[27 ]
[28 ]
[29 ]
[30 ]
[31 ]
[32 ]
[33 ]
[34 ]
[35 ]
[36 ] As such, the study results point to residual donor bone marrow elements as a potential
stimulus for mixed T- and B-cell aggregations in the subchondral bone of OCAs recovered
from revision and arthroplasty surgeries for failed transplants (see [Tables 1 ]
[2 ]
[3 ] and [Figs. 1 ]
[2 ]
[3 ]).
In this study, mechanisms of failure for the OCAs noted to be associated with cell-mediated
immune responses involved insufficient OCA osteointegration, whereas the majority
of the unsuccessful OCAs with no noted immune responses failed by other mechanisms.
Insufficient OCA osteointegration is characterized by subchondral bone fracture, collapse,
subsidence, and/or necrosis, all of which can be influenced by cell-mediated immune
responses. Successful OCA integration requires an intricate balance between degradative
and regenerative immune responses during the process of creeping substitution. The
immunogenic potential of OCA bone marrow elements may trigger responses that favor
degradative mechanisms, regulated by cytotoxic T-cells, weakening the subchondral
bone of the graft and subjugating the OCA to biomechanical failure. Cycles of continuous
exposure to OCA marrow elements during bone turnover may stimulate prolonged proinflammatory
responses capable of deterring adequate responses for regeneration.
Limitations of this study should be considered when interpreting and applying the
data. The study population comprised a relatively small number of patients selected
from a single institution's OCA transplantation registry to include all documented
treatment failures with complete blood type and outcomes data matched 1:2 with patients
documented to have successful outcomes. As such, selection bias is likely, type II
statistical errors are possible, and the results are not generalizable. Further selection
bias is likely with respect to the limited number of adequate OCA tissues available
for assessment. In addition, immune response characterizations were subjective, and
only nonimplanted OCA controls could be included based on ethical considerations for
human subjects research that makes recovery of transplanted OCAs from patients with
successful outcomes unfeasible. Finally, only short-term outcomes were evaluated,
and patient, surgical, and biomechanical mechanisms for failure were not considered
in this study. From these data, only potential associations and possible mechanisms
related to localized cell-mediated subrejection immune responses can be used for further
research aimed at preventing OCA osteointegration failure toward optimizing outcomes
for patients undergoing OCA transplantation.
Conclusion
The results of this study suggest that T- and B-cell-mediated subrejection immune
responses may play roles in OCA transplant treatment failures independent of donor–recipient
blood type mismatch effects. Residual donor bone marrow elements are a potential stimulus
for these immune responses, which were associated with OCAs that failed due to insufficient
osteointegration.
