Adoptive immunotherapy evaluating escalating doses of donor leukocytes for relapse of chronic myeloid leukemia after bone marrow transplantation: separation of graft-versus-leukemia responses from graft-versus-host disease.

Infusions of large numbers (> 10(8)/kg) of donor leukocytes can induce remissions in patients with chronic myeloid leukemia (CML) who relapse after marrow transplantation. We wanted to determine if substantially lower numbers of donor leukocytes could induce remissions and, if so, whether this would reduce the 90% incidence of graft-versus-host disease (GVHD) associated with this therapy. Twenty-two patients with relapsed CML were studied: 2 in molecular relapse, 6 in cytogenetic relapse, 10 in chronic phase, and 4 in accelerated phase. Each patient received escalating doses of donor leukocytes at 4- to 33-week intervals. Leukocyte doses were calculated as T cells per kilogram of recipient weight. There were 8 dose levels between 1 x 10(5) and 5 x 10(8). Lineage-specific chimerism and residual leukemia detection were assessed using sensitive polymerase chain reaction (PCR) methodologies. Nineteen of the 22 patients achieved remission. Remissions were achieved at the following T-cell doses: 1 x 10(7) (n = 8), 5 x 10(7) (n = 4), 1 x 10(8) (n = 3), and 5 x 10(8) (n = 4). To date, 15 of the 17 evaluable patients have become BCR-ABL negative by PCR. The incidence of GVHD was correlated with the dose of T cells administered. Only 1 of the 8 patients who achieved remission at a T-cell dose of 1 x 10(7)/kg developed GVHD, whereas this complication developed in 8 of the 11 responders who received a T-cell dose of > or = 5 x 10(7)/kg. Three patients died in remission, 1 secondary to marrow aplasia, 1 of respiratory failure and 1 of complications of chronic GVHD. Sixteen patients who were mixed T-cell chimeras before treatment became full donor T-cell chimeras at the time of remission. Donor leukocytes with a T-cell content as low as 1 x 10(7)/kg can result in complete donor chimerism together with a potent graft-versus-leukemia (GVL) effect. The dose of donor leukocytes or T cells used may be important in determining both the GVL response and the incidence of GVHD. In many patients, this potent GVL effect can occur in the absence of clinical GVHD.