This subject was a 22-year-old man with B-ALL who had third bone marrow (BM) relapse before enrollment on to our compassionate clinical protocol using TanCAR-T 19/22 cells. He was diagnosed with B-ALL with more than 100 × 109/L WBC count and normal karyotype in January 2016. After complete remission (CR) 2, he underwent haplo-HSCT from his father 10 months after the original diagnosis. He had suffered hemorrhagic cystitis and stage 1 gastrointestinal acute GVHD within 2 months post haplo-HSCT, which resolved with 15 daily doses of methylprednisolone 50 mg followed by 5 daily doses of methylprednisolone 100 mg. Three months after discontinuation of the cyclosporine A and methylprednisolone, his disease relapsed with 6.4% marrow blasts when he still had full donor chimerism, then rapidly progressed with 56.5% marrow blasts by flow cytometry 10.6 months post haplo-HSCT, and undetectable donor chimerism was noted at the same time. He received salvage chemotherapy with MOEP (3 daily doses of mitoxantrone 10 mg, vindesine 4 mg, 3 daily doses of etoposide 100 mg, and 5 daily doses of dexamethasone 15 mg) and had severe bone marrow depression and no response with 65.4% marrow blasts 1 month after the first cycle of MOEP. Then, he was treated on our haplo-CAR-T 19 cell protocol. He received cytoreduction chemotherapy with vindesine and methylprednisolone plus hydroxyurea and lymphodepleting therapy with daunorubicin and cyclophosphamide, and his marrow blasts dropped to 12.7% prior to haplo-CAR-T 19 cell infusion. Haplo-CAR-T 19 cells at a dose of 4.91 × 106/kg (2.89 × 107 T cells/kg, 17% transfection efficiency) were administered and induced MRD-negative CR (MRD-CR) and full donor chimerism within 2 weeks after infusion. The infused haplo-CAR-T 19 cells exhibited rapid expansion and peaked with 15,281 copies per microgram DNA within the first 2 days after infusion, but dropped from 3374 copies per microgram DNA at day 7 to 468 copies per microgram DNA at day 12; methylprednisolone 160 mg and dexamethasone 5 mg were used at day 11 for treatment of the infusion-related grade 3 cytokine release syndrome (CRS). He experienced stage 3 skin acute GVHD within 1 month after haplo-CAR-T 19 cell infusion, which was under control with 5 daily doses of methylprednisolone 40 mg plus cyclosporin A 80 mg administered from day 31 after haplo-CAR-T 19 cell infusion. However, 1 month after obtaining MRD-CR, his disease exhibited florid progression with WBC count increasing from 1.59 × 109 to 12.52 × 109/L and corresponding percentage of circulating blasts increasing from 1.39 to 67.37% within 2 weeks; his bone marrow exhibited highly active cellular proliferation with 59.67% blasts that had the expression pattern CD19+ CD34+ CD10+ CD22+ CD38+ CD58+ CD33+ CD20− CD13− CD15−. At the same time, undetectable haplo-CAR-T 19 cells and donor chimerism were documented. In this case, other therapies including TanCAR-T 19/22 cells rather than salvage chemotherapy or reinfusion of CAR-T 19 cells could be a potential treatment option for this patient due to the poor response to salvage chemotherapy and poor persistence of infused CAR-T 19 cells. However, higher tumor burden and short-term interval post discontinuation of steroid greatly increased the risk of failure of the generation of autologous CAR-T cells; florid progression of the disease made waiting till steroid tapered off was less feasible. Donor-derived TanCAR-T 19/22 cell therapy was an optimal approach to overcome this problem, but as well known, haplo-CAR-T cell therapies were not to be advocated routinely in the setting of prior GVHD requiring steroid mainly due to raised concern for the high risk of GVHD reactivation. After more careful consideration of the clinical benefits and risks of the second haplo-CAR-T cell infusion, he was enrolled on to our compassionate clinical protocol using haplo-TanCAR-T 19/22 cells. His father underwent apheresis, and the peripheral blood mononuclear cells (PBMCs) were used to prepare the TanCAR-T 19/22 cells. He received cytoreduction chemotherapy with vindesine 4 mg and five daily doses of methylprednisolone 80 mg and three daily doses of hydroxyurea 3 g followed by a lymphodelpeting chemotherapy with idarubicin at a total dose of 30 mg and cyclophosphamide at a total dose of 3 g. Planned bone marrow aspiration after the abovementioned chemotherapy and prior to haplo-TanCAR-T 19/22 cell infusion was not performed due to poor compliance of the patient. Two days later, he was treated with haplo-TanCAR-T 19/22 cells at a total dose of 4.72 × 106 TanCAR-T 19/22 cells per kilogram (3.05 × 107 T cells per kilogram, 15% transfection efficiency) administered via fractioned dosing (D0, 30%; D1, 70%) for safety consideration. The materials and methods used in TanCAR-T 19/22 production have been described previously [–], with the exception of the construct of the CAR and the source of PBMCs used for manufacturing the TanCAR-T 19/22 cells. TanCAR-19/22 was a tandem CAR molecule, consisting of an anti-CD22 scFv derived from mouse m971 mAb [] and anti-CD19 scFv derived from the mouse FMC63 mAb [], joined in tandem, human CD8α hinge and transmembrane domain, and human CD137 and CD3ζ signaling domains. A schematic of the TanCAR-19/22 is shown in Fig. a. PBMCs used for manufacturing the TanCAR-T 19/22 cells were collected by leukapheresis rather than fresh peripheral blood (PB). Flow cytometry was used for the determination of the TanCAR-19/22 transfection efficiency and quantification of the haplo-TanCAR-T 19/22 cells in clinical specimens using a Biotin-SP-AffiniPure Goat Anti-Mouse IgG, F (ab') 2 Fragment Specific (Jackson ImmunoResearch, USA) and PE Streptavidin antibody (BD Biosciences, USA). Haplo-TanCAR-T 19/22 cells in clinical specimens also were measured by qPCR as described []. The extent of donor engraftment in clinical specimens was assessed by using short tandem repeat amplification and fluorescence labeling multiplex PCR combined with capillary electrophoresis as described []. Serum interleukin (IL)-2, IL-6, IL-8, and IL-10 and tumor necrosis factor-α levels were batch analyzed as described []. BM before haplo-TanCAR-T 19/22 cell protocol showed predominant blast cells with an absence of normal BM precursors. BM flow cytometry at day 14 after haplo-TanCAR-T 19/22 cell infusion indicated that there were 0.73% residual marrow blasts. Of note, those residual leukemic blasts exhibited the expression pattern CD34+ CD10+ CD22+ CD38+ CD33+ CD19− CD20−, which were undetected by flow cytometry by day 28 in the absence of further therapy. Given the incomplete recovery of platelet and absolute neutrophil count by day 28, this patient achieved a MRD-CRi by day 28 after infusion. There was no evidence of blasts in BM either by BM smear or by flow cytometry at serial time points thereafter for 14 months. BM had reconstitution of normal hematopoiesis by day 56 with the exception of platelet count that still unrecovered at a level of 36 × 109/L as the time of this report. Full donor chimerism was established at day 14 post infusion and remained stable thereafter. After infusion, haplo-TanCAR-T 19/22 cells expanded and peaked at a level of 30.7% of circulating T cells at day 12 followed by a contraction phase with a low level of 0.45% of circulating T cells by day 28. This was coincident with the elimination of circulating B cells that were almost undetected at day 28 by flow cytometry. Haplo-TanCAR-T 19/22 cells were still measurable with a low level of 2.29% of circulating T cells and the circulating B cells still had not recovered as the time of this report. Haplo-TanCAR-T 19/22 cells were also present by flow cytometry at all the response evaluation time points in BM obtained at response evaluation, and chronic B cell aplasia was documented. An overall concordance between the expansion and persistence of haplo-TanCAR-T 19/22 cells in PB measured by flow cytometry and qPCR was observed. As the time of this report, TanCAR-19/22 DNA remained detectable on qPCR with 1134 and 396 copies per microgram DNA in PB and BM, respectively. After haplo-TanCAR-T 19/22 cell infusion, he experienced grade 3 CRS graded according to the UPenn grading scale [, ]. Fever of up to 38.8 °C occurred within 24 h after haplo-TanCAR-T 19/22 cell infusion, lasting 11 days and becoming afebrile at day 12 after treated with a lower dose of tocilizumab at 160 mg (1.6 mg/kg) and etanercept 50 mg at day 8. Multiple serum cytokines had markedly increased 7 days post infusion and almost returned to baseline values by day 41, where interleukin (IL)-6 levels peaked at 3377 pg/mL (88-fold over baseline) at day 11. Aspartate aminotransferase and lactate dehydrogenase significantly elevated 8 to 10 days after infusion, peaked at 1529.1 U/L (38-fold over upper limit of normal) and 2027.8 U/L (13-fold over baseline) at day 12, respectively, and returned to baseline values by day 21 with best support care. He also exhibited coagulation dysfunction with prolonged activated partial thromboplastin time, elevated D-dimer, and fallen fibrinogen concentrations, as well as capillary leak with grade 2 hypoalbuminemia in spite of intensive supplementation of protein during the CRS, which resolved by day 23. The prior stage 3 skin acute GVHD that was under control was reactivated and rapidly progressed to stage 4 skin GVHD with new-onset local skin ulcerations particularly in the scrotal skin and mouth mucosa 11 days after haplo-TanCAR-T 19/22 cell infusion. The concentration of serum total bilirubin continually elevated from day 12 and increased to 134 μmol/L at day 21. Given the rapidly progressive skin GVHD manifestations and liver involvement, lower-dose methylprednisolone at 20 mg daily as the initial dose with subsequent tapering in an effort to balance the benefits and risks of systemic immunosuppression was implemented from day 21 and discontinued by day 39. Skin rash and serum total bilirubin improved significantly after those treatments. However, stage 3 gut GVHD manifestations mainly including diarrhea occurred from day 50, and the serum total bilirubin elevated again, suggesting a grade 3 acute GVHD. Sixteen doses of methylprednisolone 20 mg per day were administered again from day 78, significantly controlling diarrhea and serum total bilirubin. This patient subsequently exhibited moderate chronic GVHD mainly manifested as scleroderma, diarrhea, and weight loss. Persistent thrombocytopenia with platelet count ranging from 15 × 109 to 43 × 109/L without platelet transfusion could be acknowledged as a manifestation of chronic GVHD in the setting of reconstitution of normal hematopoiesis. The systemic immunosuppressive treatment was tapered within 2 months with methylprednisolone 4 mg every other day and methotrexate 5 mg once a week and sirolimus 1 mg daily as a minimum maintenance dose from day 154 to the time of this report, keeping the chronic GVHD under good control.