Graft vs. HTLV-I Virus Effect Can Be Induced Independently of Graft vs. Leukemia Effect in Patients with Adult T-Cell Leukemia Who Received Allogeneic Hematopoietic Stem Cell Transplantation.

[Background/Purpose] CTL against HTLV-I virus related antigens, which virus is the causative agent for adult T-cell leukemia (ATL), have been considered of potential targets for immunotherapy against ATL. For example, CTL- specific for immunogenic protein Tax (Tax-specific CTL), have been suggested to play an important role in graft vs. leukemia effect (GVL) among ATL patients after allogeneic hematopoietic stem cell transplantation (allo-HSCT). However, it is well known that primary ATL cells largely do not express Tax. Thus we prospectively examined Tax-specific CTL in ATL patients after allo-HSCT, to evaluate the correlation between Tax-specific CTL and GVL.

[Methods] PBMC serially obtained from 8 ATL patients who received allo-HSCT were examined by tetramers containing HLA-A*0201 or A*2402 restricted Tax derived peptides (Tax_11-19, Tax_301-309, respectively) with other parameters.

[Result] Tax-specific CTL were developed in 3 of 8 (37.5%) patients (0.4%, 0.32%, and 3.2% of CD8 for case #1, 4 and 5, respectively). case#1 and #4 received HLA-mismatched and case#5 received HLA-matched allograft from their children who were negative for anti-HTLV-I antibody. Case#1 and #4 developed Tax-specific CTL at 18mo. and 4.5 mo. after allo-HSCT, when both cases documented extranodal relapse and received interventions. Case#5 developed the CTL at 6 mo. at GVHD. Those CTL were all donor origin. Tax-specific CTL of case #4 were expanded during skin GVHD following donor lymphocyte infusion (DLI), and were positive for tetramer of HLA-A*2402 restricted peptide Tax_301-309, strikingly recipient was negative for HLA-A*2402, which meant this newly developed anti-HTLV-I virus CTL had no cytotoxicity against HLA-A*2402 negative recipient tumor cells. Furthermore, in remaining 2 cases, those virus-specific CTL were positive for co-inhibitory molecule; programmed death-1 (PD-1), which indicated that those CTL are functionally impaired. As for the other 5 cases without developing Tax-specific CTL, case #7 and #8; survived in CR for longer than 1.5 yr. without Tax-specific CTL induction, case #3 and #6; clinically showed evidence of GVL irrelevant to Tax-specific CTL. 7 of 8 (88%) patients but case #2 who had an early relapse and died of ATL, showed acute and/or chronic GVHD.

[Summary] Development of anti-HTLV-I CTL after allo-HSCT for ATL patients was less frequent than that of GVHD (37.5% vs. 88%). Case #4 clearly demonstrated the anti-HTLV-I CTL without cytotoxicity against tumor cells, as is just a determinant epitope spreading phenomenon following the tumor destruction by DLI. CTL of case #1 and 5 were functionally impaired by expression of PD-1. Irrespectively of the Tax-specific CTL induction, 7 of 7 long survivors (100%) had GVHD. Furthermore case #3 and #6 clearly showed GVL in the absence of Tax-specific CTL. GVL for ATL is largely produced in corporation with GVHD-related alloresponses, but not necessarily with CTL against HTLV-I virus.