Selective Depletion of T Cell Alloresponses Using a Photodepletion Strategy Preferentially Targets CD4+ T Cells.

Abstract 1467

P-glycoprotein (Pgp) is the product of the multidrug-resistance-1 (MDR1) gene and actively transports various molecules across the extracellular membrane. Previous studies demonstrated that activated lymphocytes decrease Pgp activity and preferentially retain the photosensitizing agent 4, 5-dibromorhodamine 123 (TH9402, Kiadis Pharma, NL). We evaluated a photodepletion technique to achieve selective depletion (SD) of graft-versus-host disease (GVHD) alloreacting T cells in 24 HLA-identical sibling stem cell transplants. Donor lymphocytes were activated by 72 hr exposure with irradiated in-vitro expanded recipient T lymphocytes and pulsed with TH9402. Alloactivated T cells preferentially retaining the photosensitizer were eliminated by light exposure. The transfused cell products were then analyzed for T cell subset frequency. After SD preferential CD4+ T cell depletion occurred with an inversion of the CD4+/CD8+ ratio. Within the CD4+ compartment a significant depletion of naïve (CD27+ CD45RO-), central memory (CD27+ CD45RO+), and effector (CD27-CD45RO+) populations was noted. Conversely, CD8+ naïve and effector subsets was relatively conserved with depletion occurring predominantly in the central memory population. Additionally, an 80% reduction in B cells occurred, with a 60% reduction of cytotoxic NK cells (CD56+CD16+) and a relative expansion of regulatory NK cells (CD56+CD16-). We compared outcomes of 24 SD transplants with 34 patients receiving a T-cell depleted transplant. All patients with hematological malignancies were conditioned with fludarabine, cyclophosphamide, and total body irradiation and received a CD34-selected stem cell allograft from an HLA identical sibling. SD recipients also received 5×10⁶/kg SD donor T cells. Low-dose cyclosporine was used as the only post-transplant immunosuppression. Median follow up was 14.5 months. The probability of severe acute GvHD (grade III-IV, 13%) and relapse (25%) was low for all patients. On day 100 post-transplant SD transplant recipients had lower absolute CD4+ counts, and increased rates of CMV reactivation requiring treatment (median 2 vs 1 reactivations within 100 days, p<0.01), and more chronic GvHD (70% vs 31%, p=0.04). Previous studies have shown that Pgp efflux of rhodamine-123 in quiescent T cells varied between subsets, with a preferential retention occurring in the CD4+ and central memory compartments (Br J Haematol. 1998 Jun;101(4):722-7). Our data is in accordance with these observations. In conclusion, while SD appeared to eliminate alloactivated T cells resulting in low frequencies of severe acute GVHD, differences in Pgp activity and dye retention led to the preferential non-specific elimination of CD4+ and central memory T cells which may have contributed to greater viral reactivation and chronic GVHD.