Introduction The ideal IS post LTx is unknown and the balance between graft protection against alloimmune injury vs the risks of infection and malignancy is a delicate one. Our unit adopted induction immunosuppression with ATG followed by cyclosporin (CyA), MPA and prednisolone maintenance in 2000 with good outcomes. However, a high burden of leucopenia and concern over increasing numbers of PTLD lead to protocol change in August 2010 where de nevo azathioprine (AZA) replaced MPA. We now compare short term outcomes of the two strategies.
Methods Two cohorts were compared: cohort 1 (January 2000-July 2010) vs cohort 2 (August 2010-July 2012). ATG 2mg/kg was adminstered on 3 consecutive post operative days. CyA dosing was guided by trough (C0) and peak (C2) levels. Patients were established on MPA 2–3mg daily in divided doses in the first year and AZA titrated to a maximum of 2mg/kg daily as tolerated. We documented the incidence of biopsy proven PTLD and ACR grade ≥A2, correcting for confounders. T-tests, Chi squared, multivariate regression and Kaplan Meier statistics were applied.
Results 181 vs 52 underwent LTx in the 2 study periods. There were no differernces in age (mean 48yrs  vs 49 ; gender (M:F 110:71 vs 31:21), type of surgery (Double:Single lung 110:71 vs 34:18). Donor age, cause of death and ischaemia time were similar in the 2 cohorts. 169/181 and 3/52 received MPA vs 12/181 and 49/52 had AZA respectively. CyA dosing and serum levels did not differ between the 2 cohorts. 29 (16%) and 18 (34%) experienced ACR respectively (P<0.05). 15 patients in cohort 1 developed PTLD (mean age 54 [range 26–63], mean interval from surgery at diagnosis 286 days [73–790], 11/15 developed PTLD in the first post-op year) vs none in cohort 2. Donor:recipient EBV serostatus was similar in the 2 cohorts. Bacterial and fungal infections were documented in 94 vs 25 and 41 vs 7 respectively. Pseudomonas was isolated from a similar percentage in the 2 groups (59 vs 17). CMV viraemia necessitating pre-emptive therapy with valganciclovir was observed in 56 (32%) vs 10 (19%), P<0.05 despite similar distribution of donor:recipient CMV serostatus.
Conclusion ATG/MPA combination may better protect lung allografts against acute rejection but at the expense of a higher burden of lymphoma and infection. Valganciclovir enhances MPA bioavailability and the high prevalence of CMV activation after ATG/MPA may be a contibutor to our observations. Longer term data, including other malignancies, would further inform this debate.