Protocol biopsies were performed to define the histologic response to tacrolimus therapy in patients with refractory acute renal allograft rejection. Renal allograft biopsies were performed at defined intervals after initiation of tacrolimus therapy. Protocol biopsies were performed before tacrolimus therapy (within 48 hr of initiation of therapy) and after 1 week of therapy. If the 1-week biopsy did not show rejection reversal, repeat protocol biopsies were obtained at 1- to 2-week intervals, until histologic reversal was observed. Additional biopsies were obtained at 4 weeks and at 8-12 weeks after initiation of tacrolimus therapy. Indicated biopsies were also performed to evaluate increases in serum creatinine. A total of 92 biopsies were performed in 23 patients (average 4.0 biopsies/ patient). Biopsies were performed in each patient immediately before starting tacrolimus therapy (23 biopsies), and 69 biopsies (3.0 biopsies/patient) were performed during tacrolimus therapy. Rejection diagnosis was based on strict Banff criteria. Pretacrolimus biopsies demonstrated mild acute rejection in 64% of patients and moderate acute rejection in 36%. One week after initiation of tacrolimus therapy, protocol biopsies revealed the following: no rejection (60%), improvement (13%), no change (20%), and worsening rejection (7%). Histologic changes at 1 week did not correlate with changes in renal function, as 63% of patients that showed histologic improvement or reversal during the first 2 weeks of therapy did not show improvement in serum creatinine. A lack of histologic improvement (or worsening) at 1 week was demonstrated in a significant proportion of patients (27%); increased tacrolimus dosing provided rejection reversal or improvement in 1-2 weeks in each of these patients. Recurrent rejection was diagnosed on eight biopsies in seven patients, however six episodes were diagnosed by protocol biopsies alone (i.e., in the absence of an elevation in serum creatinine). Delayed improvement in renal function, despite histologic reversal, was likely due to physiologic effects of tacrolimus (i.e., afferent arteriolar vasoconstriction), as histologic evidence of tacrolimus toxicity was not observed during the first 2 weeks of therapy. Histologic evidence of tacrolimus nephrotoxicity (nodular arteriolar hyalinosis) was found in 21% (15 of 69) of biopsies in 39% of patients (9 of 23) at a median time of 60 days (range 12-150 days). Tacrolimus dose and blood levels (by IMx assay) did not correlate with development of clinically silent or clinically evident nephrotoxicity.
In conclusion: 1) protocol biopsies provide information that allows individualization of tacrolimus rejection therapy, 2) histologic resolution of rejection often precedes biochemical improvement, 3) histologic evidence of tacrolimus nephrotoxicity is seldom observed in the first 2 weeks of therapy, and 4) clinically silent recurrent rejection and clinically silent tacrolimus nephrotoxicity are observed with significant frequency during tacrolimus therapy for refractory renal allograft rejection.