Low Incidence of Post-Transplant EBV-Related Disease After Alemtuzumab-Mediated T Cell Depletion Is Explained by the Differential Susceptibility to Alemtuzumab of B Cells and Protective CD8 and CD4 T Cells

Abstract 2330

In vivo T cell depletion with anti-thymocyte globulin (ATG) or alemtuzumab (anti-CD52) before reduced intensity allogeneic stem cell transplantation (alloSCT) in combination with in vitro T cell depletion with alemtuzumab reduces the risk of GVHD. Detectable levels of circulating antibodies are present up to several months after the alloSCT, leading to a delayed immune reconstitution which is associated with an increased incidence of opportunistic infections and early relapses. Prior to 2007, combined in vitro (Alemtuzumab 20 mg added “to the bag”) and in vivo T cell depletion with horse-derived ATG (h-ATG) resulted in good engraftment without GVHD in the absence of GVHD prophylaxis after reduced intensity alloSCT using conditioning with fludarabine and busulphan. Due to the unavailability of h-ATG, rabbit-derived ATG (r-ATG) 10–14 mg/kg was introduced in the conditioning regimen in 2007. Strikingly, in this cohort of patients, early EBV reactivation and EBV-associated post-transplantation lymphoproliferative disease (PTLD) was observed in 10 out of 18 patients at a median time of 6 weeks after alloSCT (range 5 to 11 weeks) in the absence of GVHD or immunosuppressive treatment. Analysis of T and B cell recovery early after transplantation revealed preferential depletion of T cells as compared to B cells, thereby allowing unrestricted proliferation of EBV infected B cells. Due to this unacceptable high incidence of EBV-related complications, in the conditioning regimen r-ATG was replaced by low dose alemtuzumab (15 mg i.v. day -4 and -3) in 2008. In this cohort of 60 patients, only 2 patients experienced transient EBV reactivation during the first 3 months after alloSCT and one patient developed an EBV-associated lymphoma 4 weeks after alloSCT. To investigate the mechanisms underlying the low incidence of EBV reactivation using alemtuzumab for T cell depletion, we studied the in vivo and in vitro effects of alemtuzumab on different lymphocyte subsets. First, lineage-specific reconstitution was studied in 20 patients from the alemtuzumab cohort with known CD52 negative diseases (11 AML and 9 multiple myeloma) to exclude the confounding effect of antibody absorption by malignant cells. Whereas at 3 weeks after alloSCT detectable numbers of circulating NK cells and T cells were observed (medians 71 (range 6–378), and 12 (range 1–1164)E6/L, respectively), no circulating B cells could be detected (median 0, range 0–1 E6/L). At 6 weeks after alloSCT, NK and T cell numbers further increased (medians 212 (52-813), and 130 (range 25–1509)E6/L, respectively), whereas B cell numbers still remained low in the majority of patients (median 15, range 0–813E6/L). In all patients, T cells were detectable before the appearance of circulating B cells. Furthermore, the expression of CD52 and the sensitivity to alemtuzumab-mediated complement-dependent cell lysis (CDC) of B cells, T cells and NK cells was measured in vitro. The highest CD52 expression was observed on B cells (mean fluorescence intensity (MFI) 120), resulting in 95% lysis after incubation with 10ug/mL alemtuzumab and rabbit complement. NK cells showed a significantly lower CD52 expression (MFI 41), which was also reflected by a lower susceptibility to alemtuzumab-mediated CDC (62% lysis). Interestingly, differential expression of CD52 was observed on CD4 and CD8 T cells (MFI 120 and 101, respectively). Cytotoxicity analysis revealed relative protection of CD8 compared to CD4 T cells against alemtuzumab-mediated CDC, resulting in 52% and 90% lysis, respectively. Based on these results, we investigated in detail the presence and phenotype of the CD4 and CD8 subsets and EBV-specific CD8 T cells using tetramer staining at 6 weeks after alloSCT. In accordance with the in-vitro expression and susceptibility data, circulating CD52+ CD8 T cells including EBV-specific T cells were detectable. Interestingly, the majority of circulating CD4 T cells (64-93%, n=4) lacked CD52 expression, explaining their capacity to persist in the presence of alemtuzumab. We conclude that in vivo and in vitro T cell depletion with alemtuzumab is associated with a relatively low risk of EBV-associated PTLD because of efficient B cell depletion and persistent EBV immunity allowed by the relative insusceptibility for alemtuzumab of CD8 T cells and the development of CD52 negative escape variants of CD4 T cells.