DLI in mixed chimeras leads to eradication of CML-like disease. (A) Serial flow cytometric analysis of PBL chimerism (top row) and CML-like leukemia (bottom row) from a representative mouse with mixed chimerism and CML-like leukemia treated with DLI beginning day 25 after transplantation. Blood sampling and analysis were performed on days 25, 29, 60, and 69 after BMT, with the PBL count indicated. DLI was administered on days 25, 30, and 40, with a total dose of 1.2 × 10⁸ allogeneic splenocytes. Note the disappearance of GFP⁺ cells by day 60, accompanied by conversion to full allogeneic chimerism. (B) Earlier initiation of DLI improves the antileukemic response. Serial analysis of chimerism and leukemia were conducted as in panel A, from a representative mouse with mixed chimerism and CML-like myeloproliferative disease treated with early DLI. Blood sampling and analysis were performed on days 13, 16, 19, and 25 after BMT, and DLI was administered on days 14, 17, 20, and 23 with a total of 1.7 × 10⁸ allogeneic splenocytes. (C) Persistent mixed chimerism in some recipients cured of CML-like leukemia by DLI. Serial analysis of chimerism and leukemia were conducted as in panel A, in a recipient treated with early DLI (days 14, 18, 23, and 26 with a total of 1.3 × 10⁸ allogeneic splenocytes). Note the eradication of GFP⁺ leukemia cells but persistence of mixed chimerism with predominantly syngeneic (H-2^d+) leukocytes. (D) Plot of PB counts (y-axis, logarithmic scale) as a function of time (x-axis) after BMT (blue circles indicate chimeras with CML-like disease but no DLI treatment; red triangles, chimeras with CML-like disease and early treatment with DLI). (E-F) Cumulative mortality as a result of leukemia (E) and overall survival (F) for recipients of TCD BCR-ABL–transduced and TCD B6 BM, either without DLI treatment (blue, n = 6), with administration of allogeneic splenocytes at the time of BMT (green, n = 7), or with early DLI treatment of established mixed chimeras with CML-like leukemia (red, n = 7), in a representative transplantation cohort (1 of 4 independent experiments). The frequency of fatal CML in the DLI-treated group was significantly lower than in untreated chimeras (P = .021, Fisher exact test) and the DLI-treated group's survival was significantly longer (P = .041, Mantel-Cox test). (G) DLI in chimeric mice leads to eradication of BCR-ABL proviral clones. Genomic DNA from the indicated tissues (spl indicates spleen; asc, ascites; liv, liver) was analyzed by Southern blot with a GFP gene probe to detect distinct bands from each provirus integration site (upper panel), a Cadherin-11 gene probe that distinguishes between Balb/c- or B6-derived DNA (middle panel), and an ABL probe that allows determination of the total proviral content of each sample (bottom panels). Con1 and con2 are control DNAs from cell lines that each contained a single BCR-ABL provirus, whereas p210 spl is spleen DNA from a nonchimeric mouse with BCR-ABL–induced CML-like disease. Lanes 6 to 10 contain DNA from normal Balb/c (lane 4) or B6 spleen (lane 5) or mixtures of Balb/c and B6 DNA at the indicated ratios. Lanes 11 to 13 are from mice that received BCR-ABL–transduced BM only and were not chimeras, lanes 14 to 16 are from mice that were chimeric but not treated with DLI, and lanes 17 to 28 are from chimeric mice treated with DLI. Brackets indicate samples from the same individual mouse. Note that in 3 mixed chimeric mice with CML-like leukemia, DLI led to the disappearance of BCR-ABL proviral clones and the representative tissues (liver or spleen) are either chimeric (lane 18) or completely donor-derived (lanes 20 and 28).