Of 2947 children with acute lymphocytic leukemia (ALL), treated during three consecutive studies of the Pediatric Oncology Group (1974-1986), 52 (1.8%) had Down's Syndrome (DS). Comparison of clinical and laboratory characteristics showed no significant differences in leukocyte count, racial distribution, sex ratio, platelet count, incidence of mediastinal mass, lymphadenopathy or hepatosplenomegaly, or percentage of blood or bone marrow blasts for children with ALL with or without Down's Syndrome (DS-ALL or NDS-ALL, respectively). However, children with DS-ALL were slightly older at the time of presentation and had higher hemoglobin values. The relative frequency of each major immunophenotype (early pre-B, pre-B, T, or B) was also comparable for patients with or without DS. For this report, treatment regimens were categorized as either conventional (no consolidation therapy) or intensive. Cox regression analysis revealed that the presence of DS, a higher leukocyte count, black race, or age older than 10 years was independently associated with a poorer event-free survival (EFS) for children treated with conventional chemotherapy. However, for the cohort of children who received intensive chemotherapy, DS was no longer an independent risk factor. In fact, event-free survival (EFS) was markedly improved to a level comparable with that observed in the children diagnosed as having NDS-ALL. On the other hand, serious toxicity, requiring interruption of treatment, was significantly more frequent in the intensively treated children with DS compared with similarly treated patients with NDS-ALL, although deaths resulting from toxicity occurred infrequently.

Natural killer (NK) cells and NK cell activity were determined in three groups (newly diagnosed [n = 21], on therapy [n = 21], and off therapy [n = 18]) of children with various types of malignant solid tumors and in a control group (n = 26) by means of Leu-7 and Leu-11b monoclonal antibodies and a 4-hour 51Cr-release assay, respectively. The erythroleukemia cell line K562 was used as a target cell. The newly diagnosed group included eight patients with localized disease (Stage I-II), ten with bulky but nonmetastatic disease (Stage III), and three with metastases (Stage IV). The mean percent of NK cell activity in the newly diagnosed group was significantly higher than that of the control group. Children with Stage III tumors at diagnosis had higher mean NK cell function than those with Stage I-II and Stage IV. On therapy patients had significantly fewer NK cells and lower NK cell cytotoxicity than those in the other groups studied. We also studied the following: (1) the in vitro effect of recombinant interferon-alpha (rIFN-alpha) and recombinant interleukin-2 (rIL-2) on NK cell function of peripheral blood lymphocytes (PBL) from children with solid malignancies; and (2) the susceptibility of neuroblastoma-derived (CHP-126 and SKNSH) and rhabdomyosarcoma-derived (A-204) cell lines to NK cell lysis. Both rIFN-alpha and rIL-2 enhanced NK cell activity of PBL from children with malignancies and healthy children against K562 and solid tumor cell lines. The enhancing effect or rIL-2 was greater than that of rIFN-alpha. CHP-126 and SKNSH cell lines were susceptible to NK cell lysis mediated by the PBL of children with neuroblastoma and the control group. The A-204 cell line was less sensitive than K562 to NK cell cytotoxicity. Our results suggest a potential therapeutic role for both cytokines in the treatment of malignant solid tumors of childhood.

Several investigators have reported decreased natural killer cell activity of the peripheral blood from children with acute leukemia both at diagnosis and in relapse. This study was to determine whether this decreased natural killer activity is the result of a diminished number or defective function of natural killer cells or both. The percentage of Leu 19+ (NKH-1, CD,,) and the lytic activity of a purified Leu 19+ population from the peripheral blood of children with acute lymphocytic leukemia as well as from healthy children were determined. This cell population mediates most natural killer and natural killer-like activity in the peripheral blood. Twenty-three children with acute lymphocytic leukemia (16 at diagnosis, seven in relapse) as well as six control children were studied. The percentage of Leu 19+ cells was significantly lower in the blood of children either at initial diagnosis (median, 3%; P < 0.005) or in relapse (median, 1%; P < 0.01) than that for the control children (median, 12.5%). The cytotoxicity of peripheral blood mononuclear cells (PBMC) against K562 in %hour 'lCr release assay was significantly lower for patients either at diagnosis (median, 8.5 lytic units[LU]/1o7 cells; P < 0.005) or in relapse (median, 3 LU/107 cells; P < 0.005) than that for normal controls (median, 97 LU/107 cells). The lytic activity of a fluorescent activated cell sorter (FACS) purified Leu 19+ population was evaluated. The Leu 19+ cells were sorted from PBMC of 11 children with acute lymphocytic leukemia who had more than 2% Leu 19+ cells in their PBMC (nine at diagnosis and two in relapse). The lytic activity of sorted Leu 19+ cells was significantly (P < 0.04) lower for patients at initial diagnosis (median, 45 LU/i07 cells) than that for normal control children (median, 317 LU/io7 cells). One of the two children studied at relapse had low Leu 19+ lytic activity. The authors concluded that (1) The percentage of Leu 19+ cells in the peripheral blood of the majority of children with acute lymphocytic leukemia either at diagnosis or in relapse is below normal: (2) The lytic activity of PBMC from the majority of children with acute lymphocytic leukemia is reduced both at diagnosis and in relapse: and (3) The lytic activity of Leu 19f purified PBMC is reduced in the majority of children with acute lymphocytic leukemia. These findings confirm the defective natural killer cell profile in children with acute lymphocytic leukemia both at the time of diagnosis and in relapse.

No abstract