DISCORDANT XENOISLETS FROM A LARGE ANIMAL DONOR UNDERGO ACCELERATED GRAFT FAILURE RATHER THAN HYPERACUTE REJECTION - IMPACT OF IMMUNOSUPPRESSION, ISLET MASS, AND TRANSPLANT SITE ON EARLY OUTCOME

Background. It is known whether discordant, free, nonvascularized xeno islets-akin to discordant, vascularized, solid xenoorgans - are hypera cutely rejected. Quantitative xenoislet requirements and the optimal t ransplant site also remain to be defined. Methods. We studied these qu estions with a discordant dog-to-diabetic Lewis rat xenoislet model, u sing (1) functional (cure of streptozotocin-induced diabetes) and (2) histologic (biopsies of intraportal grafts) parameters. WF-to-Lewis al loislet recipients served as histologic controls. Results. (1) We foun d that 5000 xenoislet equivalents (IEs) transplanted into the portal v eins of nonimmunosuppressed rats never functioned. Peritransplant comb ination therapy (rapamycin, cyclosporin A, anti-rat lymphocyte serum) significantly prolonged graft survival of 5000 intraportal IEs (median , 3 days) but not of 2500 intraportal or of 5000 intraperitoneal or re nal subcapsular IEs. (2) By means of immunofluorescence (at 1 hour aft er transplantation), we noted immunoglobulin M (IgM) and IgG binding t o islets in xenografts but not allografts; we noted complement and fib rinogen binding in both xenografts and allografts. Insulin-positive is let cells within intact xenoislets were demonstrated in nonimmunosuppr essed rats up to 48 hours after transplantation. Cellular xenograft in filtration and inflammation, beginning at 6 hours, were observed even in immunosuppressed rats. (3) Thus, in spite of IgM and IgG binding, i ntraportal discordant xenoislets were not hyperacutely rejected and de stroyed. Nevertheless, universal xenoislet nonfunction in nonimmunosup pressed rats was immune mediated. A large xenoislet mass (more than 10 ,000 IEs/kg), the intraportal site, and combination therapy were absol ute prerequisites for immediate function. But even if the prerequisite s were all fulfilled, accelerated xenoislet graft failure occurred. Co nclusions. This outcome suggests that the specific binding if IgM and IgG to xenoislets, in conjunction with the binding of complement and f ibrinogen, contributed to accelerated graft failure. Thus distinction between discordant and concordant species combinations is important fo r free, nonvascularized xenoislet transplants. These findings and the steroid-free combination protocol (rapamycin, cyclosporin A, anti-T-ce ll therapy) warrant further testing in preclinical discordant xenoisle t studies.