Long-term surviving mice can induce an antigen-specific secondary response to tumor rechallenge. (A) Long-term surviving mice (100 days after primary MDA-MB-435-erbB2 tumor inoculation) treated with combined transfer of CD4^{+-transduced and CD8+-transduced T cells were able to reject a secondary intravenous injection of MDA-MB-435-erbB2 tumor cells (5 × 106 cells; □). Normal mice did not reject intravenous injection of MDA-MB-435-erbB2 tumor cells (5 × 106; ▪; *P < .001, Mann-Whitney test). (B) As a further test of specificity, long-term surviving mice received subcutaneous injections of 5 × 104 mouse mammary carcinoma 4T1.2-erbB2 (▪) or 4T1.2 parental cells (□) and their growth was statistically compared (*P < .005, Mann-Whitney test). As controls, normal scid mice received either 5 × 104 4T1.2-erbB2 (▴) or 4T1.2-parental tumor cells (▵). (C) Survival of long-term surviving mice given subcutaneous injections of 5 × 104} 4T1.2-erbB2 (□) or 4T1 parental cells (▵) was compared (*P < .005, Mann-Whitney test). As controls, normal scid mice were challenged with 5 × 104 4T1.2-erbB2 (▪)or 4T1 parental tumor cells (▴). (D) To test the ability of mice to eradicate a lower dose of 4T1.2-erbB2 tumor, long-term surviving mice were challenged subcutaneously with either 5 × 104 cells (▵) or 5 × 103 4T1.2-erbB2 tumor cells (□) and survival was statistically compared (*P < .001, Mann-Whitney test). As controls the growth of 4T1.2-erbB2 tumor cells in normal scid mice at 5 × 104 cells (▴)or5 × 103 cells (▪) was assessed. All results are calculated as the percentage of each group surviving or represented as the mean size (mm2) ± SEM and are representative of 2 experiments.