Twenty patients with relapsed B-ALL after HSCT were treated with CAR T cell therapy and were evaluated for efficacy and safety. Twelve patients previously received haploidentical transplantation, while 8 patients received HLA-matched transplantation. The median relapse time was 12 months (range, 4 to 72). Thirteen patients received autologous CAR T cells, and 7 patients received allogeneic CAR T cells, which were derived from transplant donors. The median infusion dose was 2.9×106/kg (range, 0.33 to 12×106/kg). Nineteen patients were evaluated for efficacy, among which 17 patients (89.5%) achieved MRD negative. The CR rates in the HLA-matched transplantation group and haploidentical transplantation group were 100% (7/7) and 83.3% (10/12), respectively. The median follow-up time was 9.80 months (range, 2.40 to 64.97). Ten patients (50%) died of relapse, 3 patients (15%) died of infection, and 1 patient (5%) died of aGVHD. Fifteen patients (75%) developed CRS, including 3 (20%) grade 1 CRS, 6 (40%) grade 2 CRS, and 6 (40%) grade 3 CRS. Ten patients (50%) developed aGVHD, including 1 (10%) grade I aGVHD, 6 (60%) grade II aGVHD, and 3 (30%) grade III aGVHD. The log rank test showed that CAR T cell origin was correlated with aGVHD occurrence in the haploidentical transplantation group (P = 0.005). The authors’ study indicated that the initial efficacy and safety of CAR T cell therapy for patients with post-transplant relapse were satisfactory. However, aGVHD was a concern in patients with a history of haploidentical transplantation occupied with allogeneic CAR T cells, which warrants clinical attention.