Based on the considerations above, a focused effort was undertaken to develop a xenograft ligament device from a section of the porcine patellar tendon with bone blocks and based on the immunogenicity mechanisms learned from the previous series of primate studies. Grafts were sourced from a porcine stifle and processed into a bone-tendon-bone (BTB) configuration. The graft was exposed to a series of chemical treatments. First, decellularization treatment to remove intact porcine cells and cellular components, followed by exposure to recombinant α-galactosidase enzyme solution to cleave α-gal epitopes from the graft. Removal of these α-gal epitopes was confirmed by ELISA testing with a monoclonal anti-Gal antibody that demonstrated that essentially 100 % of these epitopes were removed from the tendon portion of the construct. Following enzyme treatment, low-level glutaraldehyde cross-linking treatment was employed, with the intention of attenuating the anti-non-Gal-mediated destruction by the host immune system. We determined empirically incubating tendons in low-level glutaraldehyde followed by glycine quench of aldehyde yields for optimal conditions for macrophage infiltration allowing for the gradual remodeling of implants.

The final stage of treatment included packaging and exposure of the hydrated graft to 17.8 kGy of e-beam irradiation, a low level of irradiation intended to provide graft sterility while minimizing the degradative effects of radiation.

In order to mechanically characterize treated and sterilized porcine grafts, we used clinically relevant controls for comparative biomechanical evaluations and used standardized static testing methods. The two test groups included treated porcine device (pPT) and human bone-patellar tendon-bone allograft (hPT) cut to 9-mm width. Ten grafts were used in each assessment group. All testing used fresh-frozen grafts stored frozen and thawed just before testing. Porcine ligament graft specimens were immunochemically processed, while human patellar tendon grafts were sourced from accredited tissue banks as humans use graded specimens.

Structural properties were determined from load displacement curves: ultimate load, yield load, ultimate displacement, yield displacement, and axial stiffness. Axial stiffness was calculated from linear slope. Conversion of these tensile properties was accomplished by normalization of stress vs. strain plots and specimen cross-sectional area. Structural and material properties were derived for all specimens. A retrospective comparison of our porcine and human patellar tendon results from Noyes 1976 study evaluating young and old human anterior cruciate ligaments is shown for physical, structural, and material properties [27]. The biomechanical characterization of the specimens compared processed xenograft with human allograft. Human ACL construct groups are presented from Noyes 1976 study. Cross-sectional area and bone-to-bone length of the tested graft groups closely approximate the ACL. The structural properties of ultimate load, yield load, ultimate displacement, yield displacement, and stiffness in ten grafts are presented in Table 32.1.

No significant differences were found between xenograft device and human patellar tendon test groups in the structural parameters of yield load, ultimate displacement, yield displacement, stiffness, or yield strength. No significant differences were found between these test groups in the material parameters of ultimate strain, yield strain, or modulus. The xenograft device exhibited significantly greater ultimate load, ultimate strength, and yield strength as compared to human patellar tendon grafts. Xenograft device material property results compare favorably to ACLs of young donors as reported in the literature and resulted in an activity force limit 1.3 times the ultimate strength of young donor ACLs [26].

An ACL reconstruction study in primates was performed in 20 rhesus monkeys to evaluate the feasibility of the graft as an ACL reconstruction graft [34]. Testing involved a unilateral primate ACL reconstruction model with 2-, 6-, and 12-month sacrifice time points and clinical, histological, and biomechanical assessments. Control groups included primates implanted with an unprocessed porcine graft and allograft-implanted primates. Evaluation methods included objective functional assessment in life, radiology, serological testing of anti-Gal and anti-non-Gal antibody response, and clinical and serum chemistry. Post-sacrifice assessments included gross pathology, organ histopathology, implant histology, and postmortem biomechanics. All animals returned to normal function by 7 weeks postoperatively. Range of motion and laxity were assessed at 6 and 12 months by manual manipulation and comparison to contralateral, unoperated limbs. All range of motion and laxity measurements were considered to be clinically acceptable considering the surgical and anatomical complexities of the small primate knee. No clinically significant differences between allograft and treated porcine grafts were noted in either clinical end point. Two-month histology presented in Figs. 32.3, 32.4, 32.5, and 32.6 demonstrated parallel development between rhesus autograft and treated porcine groups in contrast to frank rejection in the untreated porcine group.