Biol Blood Marrow Transplant 25 (2019) 14981504 Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org Blinatumomab for Acute Lymphoblastic Leukemia Relapse after Allogeneic Hematopoietic Stem Cell Transplantation € kbuget3, Ralf Bargou4, Mark R. Litzow5, Anthony S. Stein1,*, Hagop Kantarjian2, Nicola Go Alessandro Rambaldi , Josep-Maria Ribera , Alicia Zhang8, Zachary Zimmerman8, 6 7 Gerhard Zugmaier9, Max S. Topp10 1 Gehr Family Center for Leukemia Research, City of Hope, Duarte, California 2 The University of Texas M.D. Anderson Cancer Center, Houston, Texas 3 Department of Medicine II, Goethe University, Frankfurt, Germany 4 Comprehensive Cancer Center Mainfranken, University Hospital Wu € rzburg, Wu€ rzburg, Germany 5 Mayo Clinic, Rochester, Minnesota 6 Department of Oncology and Hemato-Oncology, University of Milan and Azienda Pope John XXIII Hospital, Bergamo, Italy 7 Catalan Institute of Oncology-Hospital Germans Trias i Pujol, Jose Carreras Leukemia Research Institute, Badalona, Spain 8 Amgen Inc, Thousand Oaks, California 9 Amgen Research Munich, Munich, Germany 10 €rzburg, Wu Medical Clinic and Polyclinic II, University Hospital Wu € rzburg, Germany Article history: A B S T R A C T Received 14 December 2018 Patients with relapsed/refractory (R/R) acute lymphoblastic leukemia (ALL) following allogeneic hematopoietic stem Accepted 10 April 2019 cell transplantation (alloHSCT) have a poor prognosis, and alternative therapies are needed for this patient popula- tion. Blinatumomab, a bispecific T cell engager immunotherapy, was evaluated in an open-label, single-arm, phase II Key Words: study of adults with R/R Philadelphia chromosome-negative B cell precursor ALL and resulted in a rate of complete Blinatumomab remission (CR) or CR with partial hematologic recovery of peripheral blood counts (CRh) of 43% within 2 treatment Philadelphia chromosome-nega- cycles. We conducted an exploratory analysis to determine the efficacy and safety of blinatumomab in 64 patients tive B precursor ALL who had relapsed following alloHSCT before enrollment in the phase II study. Forty-five percent of the patients (29 Efficacy of 64) achieved a CR/CRh within the first 2 cycles of treatment, 22 of whom had a minimal residual disease (MRD) Safety response (including 19 with a complete MRD response). After 1 year and 3 years of follow-up, the median relapse- Allogeneic hematopoietic stem cell transplantation free survival was 7.4 months for patients who achieved CR/CRh in the first 2 cycles, and the median overall survival was 8.5 months; overall survival rate (Kaplan-Meier estimate) was 36% at 1 year and 18% at 3 years. Grade 3 and 4 adverse events were reported in 20 patients (31%) and 28 patients (44%), respectively, with grade 3 and 4 neurologic events in 8 and 2 patients, respectively, and grade 3 cytokine release syndrome in 2 patients. Eight patients had fatal adverse events, including 5 due to infections. Seven patients had grade  3 graft-versus-host disease during the study, none of which resulted in the discontinuation of blinatumomab or hospitalization. Our data suggest that bli- natumomab is an effective salvage therapy in this patient population. © 2019 American Society for Blood and Marrow Transplantation. This is an open access article under the CC BY-NC- ND license. (http://creativecommons.org/licenses/by-nc-nd/4.0/) INTRODUCTION received salvage therapies, including a second HSCT, donor lym- The outcomes of patients with relapsed/refractory (R/R) acute phocyte infusion with or without previous chemotherapy, radia- lymphoblastic leukemia (ALL) following allogeneic hematopoietic tion therapy, cytoreductive chemotherapy, mild chemotherapy, stem cell transplantation (alloHSCT) are poor. A retrospective or palliative care, showed 1- and 2-year overall survival rates of analysis of adults with ALL who relapsed after first alloHSCT and 17% and 10%, respectively [1]. Overall survival rates from a simi- lar analysis were 16% at 2 years and 8% at 5 years [2]. Two studies Financial disclosure: See Acknowledgments on page 1504. that focused on outcomes of second alloHSCT found overall sur- The results of this analysis were presented in part at the 57th American Soci- vival rates of 23% at 1 year [3] and 11% and 35% at 3 years [3,4]. ety of Hematology Annual Meeting and Exposition, Orlando, FL, December 5 to Thus, there is a need for more effective salvage therapies for 8, 2015. Qualified researchers may request data from Amgen clinical studies. Com- patients with ALL who relapse after alloHSCT. plete details are available at http://www.amgen.com/datasharing. CD19 is expressed in almost all (> 90%) tested B lineage ALL * Correspondence and reprint requests: Anthony S. Stein, MD, Gehr Family Cen- cells [5,6] and is a target for immunotherapy in ALL. Blinatu- ter for Leukemia Research, City of Hope, 1500 E Duarte Rd, Duarte, CA 91010. momab is a bispecific T cell engager (BiTE) immunotherapy E-mail address: astein@coh.org (A.S. Stein). https://doi.org/10.1016/j.bbmt.2019.04.010 1083-8791/© 2019 American Society for Blood and Marrow Transplantation. This is an open access article under the CC BY-NC-ND license. (http://creativecommons.org/licenses/by-nc-nd/4.0/) A.S. Stein et al. / Biol Blood Marrow Transplant 25 (2019) 14981504 1499 with dual specificity for CD19 and CD3 [7,8]. Blinatumomab Statistical Analysis simultaneously binds CD3+ cytotoxic T cells and CD19+ B cells Patients who had undergone alloHSCT before enrollment into the phase II study were included in the exploratory analysis. The proportion of patients and redirects T cells to lyse malignant and normal B cells [7]. who achieved responses (with 95% confidence intervals [CIs]) are presented; In a phase II study of 189 adults with R/R Philadelphia chro- relapse-free survival and overall survival are described with Kaplan-Meier mosome (Ph)-negative B cell precursor ALL, treatment with estimates. Statistical analysis was performed using SAS version 9.3 (SAS Insti- single-agent blinatumomab resulted in a rate of complete tute, Cary, NC). The analysis of relapse-free survival was restricted to patients who had remission (CR) or CR with partial hematologic recovery of CR or CRh and was calculated relative to the date of bone marrow aspiration peripheral blood counts (CRh) of 43% within the first 2 cycles when CR or CRh was first achieved. The date of detection of hematologic of treatment [9]. The median relapse-free survival was 5.9 relapse, extramedullary disease, or death due to any cause, whichever was months, and median overall survival was 6.1 months [9]. earlier, served as the event date for relapse-free survival. A patient with CR or In the present exploratory analysis, we evaluated the effi- CRh who did not have hematologic relapse and did not die was censored on the date of the last available bone marrow aspiration or the last survival fol- cacy and safety of blinatumomab in the 64 patients who had low-up visit, whichever was later. Overall survival was calculated relative to relapsed following alloHSCT before enrollment in the phase II the start of blinatumomab treatment. Patients alive were censored on the last blinatumomab study. documented visit date or last contact date. METHODS RESULTS Study Design and Patients Patients and Treatment Details of the design of the phase II study and study participants have Sixty-four of the 189 patients enrolled and treated with bli- been published previously [9]. In brief, this open-label, single-arm, phase II study was conducted in Europe and the United States. For the current analy- natumomab in the phase II study had relapsed after previous sis, eligible patients were age 18 years with R/R Ph-negative B cell precursor alloHSCT, of whom 67% were male and 63% were age <35 ALL who had relapsed following alloHSCT before enrollment in the study; years (Table 1). Similar proportions of patients had either a alloHSCT must have been performed at least 3 months before the start of matched sibling (47%) or an unrelated (50%) donor (20.3% treatment. Additional eligibility criteria included 10% bone marrow blasts matched, 12.5% unmatched, 3.1% cord blood, and 31% with with or without measurable extramedullary disease, Eastern Cooperative Oncology Group performance status of 2, no active acute (grade II-IV) or unavailable data). Fifty-five patients (86%) had received sal- active chronic graft-versus-host disease (GVHD) or systemic treatment for vage therapy after alloHSCT and before treatment with blina- GVHD within 2 weeks before treatment start, and no history or presence of tumomab. Thirty-four patients (53%) had received a clinically relevant central nervous system (CNS) pathology. This study is reg- myeloablative conditioning regimen, 10 (16%) had received a istered at ClinicalTrials.gov (NCT 01466179). Patients received up to 5 cycles of blinatumomab with an end-of-study reduced-intensity conditioning regimen, 6 (9%) had undergone visit 30 days after the last treatment cycle and efficacy follow-up visits at 3, 6, 2 previous transplantations before receiving blinatumomab 9, 12, 18, and 24 months after the start of treatment. Blinatumomab was and had received myeloablative and reduced-intensity condi- administered by continuous i.v. infusion at 9 mg/day for the first 7 days of the tioning regimens, and 14 (22%) had received regimens that cycle 1, 28 mg/day from day 8 through the end of cycle 1, and 28 mg/day in the subsequent cycles. A cycle consisted of continuous i.v. infusion of blinatu- were not specified or not available. Twelve patients (19%) were momab over 4 weeks, followed by a 2-week treatment-free interval. Blinatu- reported to have received antithymocyte globulin (ATG) ther- momab treatment was interrupted if a patient experienced a grade 3 apy as part of their conditioning regimen, between 103 and CNS-related adverse events or other clinically relevant grade 3 or 4 adverse 574 days before enrollment. events and was restarted when the adverse event had resolved to grade 1 or baseline. Blinatumomab was discontinued if treatment was interrupted for As of the data cutoff date (June 20, 2014), 50 patients (78%) more than 2 weeks due to an adverse event, a grade 4 CNS event, or at least 2 had discontinued the study, including 47 (73%) who had died, seizures. 1 (1.6%) who was lost to follow-up, and 2 (3.1%) who withdrew Patients who achieved CR or CRh within the first 2 cycles of treatment from the study. Causes of death were disease progression in 33 could receive up to 3 additional cycles of consolidation treatment and could patients (53%); sepsis in 5 patients (8%), including 2 with fatal also undergo alloHSCT. Patients with hematologic relapse during the follow- up period (at least 3 months after the completion of treatment) could receive adverse events of sepsis; respiratory insufficiency/failure in 4 up to 3 additional cycles of treatment (9 mg/day for the first 7 days) for a max- patients (6%); brain hemorrhage, gastrointestinal hemorrhage, imum of 8 cycles. and grade IV GVHD in 1 patient each (1.6%): and unknown cause in 2 patients (3.1%). Outcomes of Patients with Previous AlloHSCT Study endpoints included CR/CRh rate within the first 2 cycles of treat- Efficacy of Blinatumomab in Patients with Previous AlloHSCT ment with blinatumomab, alloHSCT after CR/CRh, minimal residual disease Forty-five percent of patients (29 of 64) with previous (MRD) response (prespecified exploratory endpoint), incidence of adverse events, relapse-free survival, and overall survival. alloHSCT achieved a CR/CRh within the first 2 cycles of treat- CR was defined as bone marrow blasts 5% with no evidence of disease ment (Table 2). The rate of CR/CRh was 43% (54 of 125) among and full recovery of peripheral blood counts (platelet count >100,000/mL and patients who did not have previous alloHSCT. absolute neutrophil count >1,000/mL). CRh was defined as bone marrow Among the patients with previous alloHSCT, CR/CRh within blasts 5% with no evidence of disease and partial recovery of peripheral blood counts (platelet count >50,000/mL and absolute neutrophil count the first 2 cycles was achieved by 6 of 12 patients (50%) who >500/mL). received previous ATG, compared with 23 of 52 patients (44%) MRD response was defined as MRD <10¡4 leukemic cells as measured by who did not receive previous ATG. Among the 29 patients with polymerase chain reaction (PCR) at a central laboratory. Complete MRD previous HSCT who achieved a CR/CRh within the first 2 cycles, response, a subset of MRD response, was achieved if no PCR amplification of individual rearrangements of immunoglobulin genes or T cell receptor genes 7 underwent subsequent alloHSCT after treatment with blina- was detected. Assay requirements were sensitivity of at least 10¡4 and quan- tumomab while in remission, and 3 underwent subsequent titative measurement range of at least 10-4. alloHSCT after relapse. In addition, 22 of the 29 patients who achieved CR/CRh within the first 2 cycles had an MRD Determination of T Cell Subsets at Study Baseline response, including 19 with a complete MRD response Samples of lymphocyte subsets were obtained at screening from all (Table 2). MRD response rates were similar in patients who did patients to investigate any effect of previous transplantation on the number not undergo alloHSCT before receiving blinatumomab (Supple- of T cells at study baseline. Flow cytometry was used to determine the num- bers of CD4+ and CD8+ T cells, including CD4+ naïve T cells, central memory T mentary Table S1). cells (TCM cells), and effector memory T cells (TEM cells). and CD8+ naïve T Among the 5 patients who died of sepsis during the study, 3 cells, TCM cells, TEM cells, and effector memory RA T cells (TEMRA cells). did not respond to blinatumomab, 1 had an unknown response 1500 A.S. Stein et al. / Biol Blood Marrow Transplant 25 (2019) 14981504 Table 1 Table 2 Demographic Data and Baseline Disease Characteristics of the Patients with Clinical Outcomes in the Patients with Previous AlloHSCT Previous AlloHSCT (N = 64) Outcome Value Characteristic Value Hematologic response (N = 64), n (%) (95% CI) Male sex, n (%) 43 (67) CR/CRh in first 2 cycles 29 (45) (33-58) Age, yr, median (range) 32 (1974) CR in first 2 cycles 18 (28) (18-41) Age group, n (%) CRh in first 2 cycles 11 (17) (9-29) 18 to <35 yr 40 (63) Blast-free hypoplastic/aplastic bone marrow 4 (6) (2-15) 35 to <65 yr 22 (34) Progressive disease 12 (19) (10-31) 65 yr 2 (3) Failure to respond to therapy 12 (19) (10-31) Donor type, n (%) No response assessment 7 (11) Haploidentical 2 (3) Molecular response*, n (%) (N = 29) Sibling 30 (47) MRD response 22 (76) Unrelated 32 (50) Complete MRD response 19 (66) Degree of matching MRD nonresponse 4 (14) Matched* 13 (20) No MRD response assessment 3 (10) Mismatchedy 8 (12) * Based on 29 patients who had achieved CR/CRh within the first 2 cycles. Cord blood (both donors 4/6 1 (2) match) Unavailable 10 (16) responded to blinatumomab, and the patient who had grade IV Previous salvage after alloHSCT, 55 (86) acute GVHD underwent an on-study alloHSCT and died 6 before blinatumomab, n (%) months after completing blinatumomab treatment. Number of previous salvage therapies before blinatumomab, n (%) The median relapse-free survival was 7.4 months (95% CI, None 9 (14) 5.0 to 10.1 months) in patients who achieved CR/CRh in the 1 23 (36) first 2 cycles, with a median duration of follow-up of 12.4 2 12 (19) months (95% CI, 11.5 to 18.0 months) (Table S2; Figure 1A). 3 20 (31) The median relapse-free survival was 11.4 months (95% CI, 2.3 Conditioning regimens for previous alloHSCT, n (%) to 24.9 months) in patients who achieved CR/CRh and had 1 Myeloablative 34 (53) previous relapse and 6.2 months (95% CI, 3.8 months to not Reduced intensity 10 (16) estimable [NE]) for patients who had at least 2 previous relap- Myeloablative and reduced 6 (9) ses (Supplementary Table S2). The median overall survival was intensityz 8.5 months (95% CI, 4.2 to 11.2 months), with a median follow- Not specified/not available 14 (22) up of 16.6 months (95% CI, 12.4 to 23.3 months) (Supplemen- ATG therapy as part of conditioning 12 (19) tary Table S2; Figure 1B). Similar to relapse-free survival, regimen, n (%) overall survival was approximately twice as long for patients History of GVHD, n (%) 19 (30) with 1 previous relapse compared with patients with at least 2 Chronic 4 (6) previous relapses (14.3 months [95% CI, 4.0 to 23.1 months] Acute 1 (2) versus 6.5 months [95% CI, 3.5 to 9.3 months]) (Supplementary Not specified 14 (22) Table S2). The median relapse-free survival and overall survival Time from previous alloHSCT to first 9.7 (3.3-40.2) at 3 years was 7.4 months (95% CI, 5.0 to 10.1 months) and 8.5 blinatumomab dose, mo, median months (95% CI, 4.2 to 1.2 months), respectively (median fol- (range) low-up: relapse-free survival, 34.9 months [95% CI, 34.4 to Number of previous relapsesx, n (%) 35.6 months]; overall survival, 36.0 months [95% CI, 35.6 to 1 23 (36) 36.4 months]) (Supplementary Figure 1). The overall survival 2 24 (38) rate (Kaplan-Meier estimate) was 36% (95% CI, 24% to 48%) at 1 >2 17 (27) year and 18% (95% CI, 9% to 29%) at 3 years. Baseline bone marrow blast category (central laboratory), n (%) Of the 19 patients with previous HSCT who achieved a CR/ 10% to <50% 19 (30) CRh during the first 2 cycles of treatment and had a complete 50% 42 (66) MRD response, the median relapse-free survival was 7.6 Lymphocyte count, £ 109/L, mean 7.6 (15.8) months (95% CI, 3.8 to 11.2 months) (n = 16) and the median (SD) overall survival was 23.1 months (95% CI, 9.4 to 25.3 months) Neutrophil count, £ 109/L, mean (SD) 2.1 (2.7) (n = 10), with a median follow-up of 16.6 months (95% CI, 12.4 * Matched patients: 10/10, 8 patients; 11/11, 1 patient; fully matched, 4 to 19.0 months). Of the 10 patients who achieved a CR/CRh patients. during the first 2 cycles of treatment and did not have a com- y Mismatched patients: 9/10, 4 patients; 8/10, 2 patients; 4/6, 1 patient; and plete MRD response, the median relapse-free survival was 6.1 mismatched (not specified), 1 patient. z months (95% CI, .1 month to NE) (n = 7) and the median overall Underwent 2 transplantations before receiving blinatumomab. x Before receiving blinatumomab. survival was 12.7 months (95% CI, 4.0 months to NE) (n = 6), with a median follow-up of 12.8 months (95% CI, 11.2 to 23.3 months). to blinatumomab, and 1 developed sepsis after alloHSCT. Among the 4 patients who died of respiratory insufficiency/ Safety failure, 1 relapsed and 3 did not respond to blinatumomab. The All 64 patients with previous alloHSCT (100%) experienced patient who died of brain hemorrhage had relapsed, the at least 1 adverse event during the study, with grade 3, grade patient who died of gastrointestinal hemorrhage had not 4, and serious adverse events reported in 20 (31%), 28 (44%), A.S. Stein et al. / Biol Blood Marrow Transplant 25 (2019) 14981504 1501 Figure 1. Relapse-free survival and overall survival for patients with previous alloHSCT. (A) Relapse-free survival for patients who achieved CR/CRh. (B) Overall sur- vival. Vertical lines represent patients who were censored. and 45 (70.3%) patients, respectively (Table 3). Eight patients the GVHD events was considered serious, none resulted in the (12.5%) had a grade 3 neurologic events, and 2 patients (3%) discontinuation of blinatumomab, and none resulted in hospi- had grade 3 cytokine release syndrome. Two patients had a talization. All patients were treated with steroids, and 1 patient grade 4 neurologic event, and there were no grade 4 events of also received tacrolimus (Table 4). cytokine release syndrome. The 3 most common grade 3 Seventeen of 19 patients with a history of GVHD before adverse events were neutropenia, febrile neutropenia, and treatment with blinatumomab did not develop reactivation of anemia. Eight patients had fatal adverse events, including 5 GVHD following treatment with blinatumomab; only 2 due to infections (2 from sepsis, 1 with Fusarium infection, 1 patients with a history of GVHD developed GVHD during the with bacteremia, and 1 with Candida infection considered study. related to blinatumomab treatment); none of these patients had achieved CR/CRh while receiving blinatumomab. There T Cell Populations at Screening were no fatal neurologic or cytokine release events. Blood samples for the determination of T cell numbers were In general, the incidences of adverse events were similar available for 61 patients (95%) in the previous alloHSCT group among patients with and without previous alloHSCT. A notable and for 114 patients (91%) in the no previous alloHSCT group. exception was the incidence of grade 4 adverse events, which The distribution of CD4+ and CD8+ cell subsets measured at was 2-fold higher in patients who had undergone previous- screening was similar in the 2 groups of patients, although alloHSCT (Table 3). there was an apparent trend toward higher numbers of TEM Seven patients (11%) experienced GVHD during the study and TEMRA cells in the previous alloHSCT group (Figure 2). (Table 4), including 6 after receiving blinatumomab and 1 after undergoing subsequent on-study alloHSCT. One patient had GVHD of the mouth (grade 1, chronic), 1 had GVHD of the skin DISCUSSION (grade 2 chronic); 2 had cutaneous GVHD (one grade I and one The results from this exploratory analysis demonstrate that grade III, both acute), 1 had GVHD events involving joints blinatumomab is effective in patients with R/R Ph-negative B (grade 2 acute arthralgia) and liver (grade 3), 1 had grade II cell precursor ALL who had relapsed following alloHSCT, a acute unspecified GVHD, and 1 had grade II acute GVHD of the patient population with historically poor outcomes. Of note, oral mucosa. There were no grade 4 or 5 GVHD events. None of 65% of the patients in our study had 2 or more previous 1502 A.S. Stein et al. / Biol Blood Marrow Transplant 25 (2019) 14981504 Table 3 Adverse Events Adverse Events* Patients with Previous AlloHSCT (N = 64), n (%) Patients without Previous AlloHSCT (N = 125), n (%) All events, any grade 64 (100.0) 124 (99.2) Grade 3 20 (31.3) 52 (41.6) Neurologic events 8 (12.5) 14 (11.2) Cytokine release syndrome 2 (3.1) 1 (.8) Infections 14 (21.9) 26 (20.8) Grade 4 28 (43.8) 27 (21.6) Neurologic events 2 (3.1) 1 (.8) Cytokine release syndrome 0 (.0) 0 (.0) Infections 6 (9.4) 6 (4.8) Serious 45 (70.3) 78 (62.4) Fatal 8 (12.5) 20 (16.0) Sepsis 1 (1.6) 3 (2.4) Disease progression 1 (1.6) 1 (.8) Fusarium infection 1 (1.6) 1 (.8) Septic shock 1 (1.6) 1 (.8) Candida infectiony 1 (1.6) 0 (.0) Enterococcal bacteremia 1 (1.6) 0 (.0) Gastrointestinal hemorrhage 1 (1.6) 0 (.0) Respiratory failure 1 (1.6) 0 (.0) Grade 3 adverse events occurring in 5 patients Febrile neutropenia 13 (20.3) 36 (28.8) Neutropenia 14 (21.9) 16 (12.8) Anemia 11 (17.2) 16 (12.8) Pneumonia 6 (9.4) 11 (8.8) Thrombocytopenia 9 (14.1) 7 (5.6) Leukopenia 6 (9.4) 9 (7.2) Elevated alanine aminotransferase 8 (12.5) 6 (4.8) Pyrexia 7 (10.9) 7 (5.6) Hypokalemia 6 (9.4) 7 (5.6) Decreased white blood cell count 5 (7.8) 4 (3.2) Elevated aspartate aminotransferase 5 (7.8) 3 (2.4) Elevated serum bilirubin 5 (7.8) 3 (2.4) Encephalopathy 5 (7.8) 1 (.8) Hypotension 5 (7.8) 1 (.8) Adverse events were graded using Common Terminology Criteria for Adverse Events, version 4.0. * During treatment until 30 days after treatment or before alloHSCT if applicable. y Considered by the investigator to be related to blinatumomab treatment. relapses, implying an especially dismal outcome with standard and the median overall survival was 6.1 months [9]. The out- of care therapy [2,3]. comes of patients who underwent previous alloHSCT were Among the total patient population of the phase II study, similar; 45% achieved a CR/CRh within the first 2 cycles, the 43% of patients achieved a CR/CRh within the first 2 cycles of median relapse-free survival was 7.4 months, and the median treatment, the median relapse-free survival was 5.9 months, overall survival was 8.5 months. The survival rates of 36% at 1 Table 4 Patients with GVHD Events* Patient History of GVHD GVHD Event Worst Grade Chronic/Acute Medication for GVHD Continuation of Blinatumomab 1 No Mouth 1 Chronic Budesonide Yes 2 No Skin 2 Chronic Not reported Yes 3 Yes Cutaneous 3 Acute Dexamethasone Yes 4 No Cutaneous 1 Acute Dexamethasone Yes 5y Yes Arthralgia 2 Acute Dexamethasone Methylprednisolone, Not applicablez Liver 3 Acute tacrolimus 6 No Not specified 2 Acute Dexamethasone Yes 7 No Oral mucosa 2 Chronic Methotrexate Yes * During treatment until 30 days after treatment; graded using Common Terminology Criteria for Adverse Events, version 4.0. y This patient had 2 different GVHD events. z GVHD occurred after treatment with blinatumomab ended, within 30 days of the last dose. A.S. Stein et al. / Biol Blood Marrow Transplant 25 (2019) 14981504 1503 Figure 2. T cell subsets at baseline in patients with and without previous alloHSCT. The numbers of CD4+ and CD8+ T cells were determined by flow cytometry. year and 18% at 3 years are comparable to those obtained after respectively); relapse-free and overall survival of patients with a second alloHSCT [3]. Patients who experienced more than 1 previous HSCT was not reported [17]. relapse after their previous alloHSCT and before treatment The adverse events reported in the present study were con- with blinatumomab had shorter relapse-free survival and sistent with those reported in previous blinatumomab clinical overall survival compared with those with only 1 relapse. A studies and were generally balanced between the previous decreasing survival rate after subsequent lines of salvage ther- alloHSCT and no previous alloHSCT groups. An exception was apy has previously been reported in patients with relapsed ALL the incidence of grade 4 adverse events, which was 2-fold [10]. higher in the previous-alloHSCT group, a heavily pretreated Similar proportions of patients in the previous alloHSCT population of patients. Fatal adverse events were due mainly and no previous alloHSCT groups had a complete MRD to infections and occurred in patients with uncontrolled leuke- response (66% versus 62%), providing additional evidence that mia. Apart from relapse, there were no other neoplastic causes blinatumomab is effective in patients with more aggressive of death, and no cases of post-transplantation lymphoprolifer- ALL. As expected, patients who had an MRD response had lon- ative disease occurred after blinatumomab therapy and during ger relapse-free survival and overall survival compared with follow-up. The incidence of grade 3 neurologic events was patients who did not have an MRD response. similar in the 2 groups. Neurologic events of grade 3 were Our results are comparable to the findings in a blinatumo- managed by treatment interruption or discontinuation, which mab arm of a phase III study that compared the efficacy and did not appear to substantially impact the response to blinatu- safety of blinatumomab with chemotherapy. In that study, 34% momab [9]. The low incidence of grade 3 cytokine release of patients had received previous alloHSCT, and the treatment syndrome was likely due to premedication with dexametha- benefit with blinatumomab was consistent among patients sone and the use of a stepped-dosage regimen [9]. A low inci- with and patients without previous alloHSCT. Remission rates dence of cytokine release syndrome was also observed in the also favored blinatumomab in both groups of patients [11]. phase III study using these mitigation strategies [11]. Other novel therapeutic agents that have demonstrated effi- Blinatumomab treatment did not appear to be associated cacy in patients with R/R ALL include anti-CD19 chimeric with an increase in GVHD among patients with previous antigen receptor T cell (CAR T-cell) therapy [1214] and the alloHSCT. All but 2 cases of GVHD (which was observed in 11% anti-CD22 antibody-calicheamicin conjugate inotuzomab ozo- of the patients) were mild or moderate in severity; none gamicin [1517]. A subgroup analysis of patients with previous resulted in hospitalization, and none affected blinatumomab transplantation (n = 46) and without previous transplantation treatment (ie, blinatumomab administration was not inter- (n = 29) treated with the CAR T cell agent tisagenlecleucel rupted). Of note, only 2 of the 7 patients who experienced showed a slightly higher rate of CR/CRh in the former group; GVHD after treatment with blinatumomab had a history of the CR/CRh rate was 81% for the overall patient population [12]. GVHD; one-third of patients who had undergone previous Other efficacy outcomes were not reported for this subgroup of alloHSCT had a history of GVHD. The incidence of GVHD in our patients. Two additional studies reported CR rates of 100% study cohort is lower than that observed in patients who (n = 8) [13] and 84% (n = 19) [14] among patients with previous receive donor lymphocyte infusion for relapse after alloHSCT. HSCT treated with CAR T cell therapy. Those studies did not Up to 60% of patients who receive donor lymphocyte infusions report the survival of patients with previous HSCT, however. For develop acute GVHD, approximately one-third develop clini- the overall patient populations, 1-year survival rates were 61% cally significant GVHD, and GVHD is the leading cause of mor- [13] and 76% [14], and after long-term follow-up, the median tality in these patients [18,19]. Blinatumomab treatment was overall survival among patients with low versus high disease not associated with any fatal cases of veno-occlusive disease, burden was 20.1 months versus 12.4 months [13]. In the phase which is considered a major toxicity in patients treated with III study of inotuzomab ozogamicin versus standard of care, inotuzumab ozogamicin [1517,20]. In addition, we did not patients with and without previous HSCT treated with inotuzo- observe any cases of Epstein-Barr virus-associated post- mab ozogamicin had similar CR/CRh rates (76.5% and 81.5%, transplantation lymphoproliferative disorder. 1504 A.S. Stein et al. / Biol Blood Marrow Transplant 25 (2019) 14981504 The proportions of specific T cell populations at baseline stockholder of Amgen. M.S.T. has received a research grant and were similar in the previous allo HSCT and no previous personal fees from Amgen. alloHSCT groups, even though approximately one-third of patients had received ATG before alloHSCT. These data reflect SUPPLEMENTARY MATERIALS the similar outcomes observed in both groups of patients. Supplementary material associated with this article can be Although this was not investigated in the present study, it has found in the online version at doi:10.1016/j.bbmt.2019.04.010. been reported that a cutoff of 8.52% regulatory T cells at base- line identified 100% of patients who will respond to blinatumo- REFERENCES mab and excluded 70% of those who will not respond to 1. Poon LM, Hamdi A, Saliba R, et al. Outcomes of adults with acute lympho- blastic leukemia relapsing after allogeneic hematopoietic stem cell trans- blinatumomab [21]. plantation. Biol Blood Marrow Transplant. 2013;19:1059–1064. Our present study has some limitations. This phase II study 2. Spyridonidis A, Labopin M, Schmid C, et al. Outcomes and prognostic fac- was not specifically designed to prospectively evaluate the tors of adults with acute lymphoblastic leukemia who relapse after alloge- neic hematopoietic cell transplantation. An analysis on behalf of the Acute outcomes of patients who underwent previous alloHSCT, Leukemia Working Party of EBMT. Leukemia. 2012;26:1211–1217. potentially limiting the capture of transplantation data. In 3. Poon LM, Bassett Jr R, Rondon G, et al. Outcomes of second allogeneic addition, this retrospective analysis was performed on a small hematopoietic stem cell transplantation for patients with acute lympho- blastic leukemia. Bone Marrow Transplant. 2013;48:666–670. subset of patients enrolled in a single-arm, open-label study. 4. Aljasem HA, Messner HA, Lipton JH, et al. Outcome following second allo- In summary, blinatumomab appears to be a feasible and geneic hematopoietic cell transplantation: a single-center experience. Eur effective salvage therapy for patients with R/R ALL after J Haematol. 2018;100:308–314. alloHSCT and is not associated with a higher rate of severe side 5. Raponi S, De Propris MS, Intoppa S, et al. Flow cytometric study of poten- tial target antigens (CD19, CD20, CD22, CD33) for antibody-based immu- effects compared with patients who do not undergo alloHSCT. notherapy in acute lymphoblastic leukemia: analysis of 552 cases. Leuk Several ongoing studies in patients with R/R ALL are evaluating Lymphoma. 2011;52:1098–1107. blinatumomab in combination with other agents, including 6. Piccaluga PP, Arpinati M, Candoni A, et al. Surface antigens analysis reveals significant expression of candidate targets for immunotherapy in adult inotuzumab ozogamicin, the anti-PD-1 immunotherapy agents acute lymphoid leukemia. Leuk Lymphoma. 2011;52:325–327. pembrolizumab and nivolumab, and the anti-CTLA-4 immuno- 7. Lo€ ffler A, Gruen M, Wuchter C, et al. Efficient elimination of chronic lympho- therapy agent ipilimumab. Further improvements in the out- cytic leukaemia B cells by autologous T cells with a bispecific anti-CD19/ anti-CD3 single-chain antibody construct. Leukemia. 2003;17:900–909. comes of patients with R/R ALL await the results of these 8. Bargou R, Leo E, Zugmaier G, et al. Tumor regression in cancer patients by studies. very low doses of a T cell-engaging antibody. Science. 2008;321:974–977. 9. Topp MS, Go € kbuget N, Stein AS, et al. Safety and activity of blinatumomab for adult patients with relapsed or refractory B-precursor acute lympho- ACKNOWLEDGMENTS blastic leukaemia: a multicentre, single-arm, phase 2 study. Lancet Oncol. 2015;16:57–66. The authors thank all the participating patients and their 10. Go€ kbuget N, Dombret H, Ribera JM, et al. International reference analysis families. They also thank the following individuals from Uni- of outcomes in adults with B-precursor Ph-negative relapsed/refractory versity Hospital Schleswig-Holstein: Heinz Horst, for providing acute lymphoblastic leukemia. Haematologica. 2016;101:1524–1533. 11. Kantarjian H, Stein A, Go €kbuget N, et al. Blinatumomab versus chemother- expert analysis of cytomorphology; Wolfram Klapper, for pro- apy for advanced acute lymphoblastic leukemia. N Engl J Med. viding expert analysis of histopathology; and Monika Bru € gge- 2017;376:836–847. mann, for providing expert analysis of MRD response. 12. Maude SL, Laetsch TW, Buechner J, et al. Tisagenlecleucel in children and young adults with B-cell lymphoblastic leukemia. N Engl J Med. Financial disclosure: This study was funded by Amgen Inc. 2018;378:439–448. Medical writing support was funded by Amgen Inc and pro- 13. Park JH, Rivie re I, Gonen M, et al. Long-term follow-up of CD19 CAR ther- vided by Kathryn Boorer, PhD, of KB Scientific Communica- apy in acute lymphoblastic leukemia. N Engl J Med. 2018;378:449–459. tions, LLC. 14. Wei G, Hu Y, Pu C, et al. CD19 targeted CAR-T therapy versus chemother- apy in re-induction treatment of refractory/relapsed acute lymphoblastic Conflict of interest statement: A.S.S. has served on speakers leukemia: results of a case-controlled study. Ann Hematol. 2018;97: bureaus for Amgen and Celgene. H.K. has received research 781–789. funding from Abbvie, Agios Pharmaceuticals, Amgen, Ariad 15. Kantarjian H, Thomas D, Jorgensen J, et al. Inotuzumab ozogamicin, an anti-CD22-calecheamicin conjugate, for refractory and relapsed acute Pharmaceuticals, Astex Pharmaceuticals, Bristol-Myers Squibb, lymphocytic leukaemia: a phase 2 study. Lancet Oncol. 2012;13:403–411. Cyclacel Pharmaceuticals, Daiichi-Sankyo, ImmunoGen, Jazz 16. Kantarjian H, Thomas D, Jorgensen J, et al. Results of inotuzumab ozogami- Pharmaceuticals, Novartis, and Pfizer and honoraria from Abb- cin, a CD22 monoclonal antibody, in refractory and relapsed acute lym- phocytic leukemia. Cancer. 2013;119:2728–2736. vie, Actinium Pharmaceuticals, Agios Pharmaceuticals, Amgen, 17. Kantarjian HM, DeAngelo DJ, Stelljes M, et al. Inotuzumab ozogamicin ver- ImmunoGen, Orsinex, Pfizer, and Takeda. N.G. has received sus standard therapy for acute lymphoblastic leukemia. N Engl J Med. research funding, personal fees, and travel support from Amgen, 2016;375:740–753. 18. Roddie C, Peggs KS. Donor lymphocyte infusion following allogeneic Novartis, and Pfizer and has participated on advisory boards for hematopoietic stem cell transplantation. Expert Opin Biol Ther. Amgen, Novartis, Pfizer, and Celgene. R.B. has received consul- 2011;11:473–487. tancy honoraria from Amgen, Cellex, GEMoaB Monoclonals, and 19. Frey NV, Porter DL. Graft-versus-host disease after donor leukocyte infu- Novartis. M.R.L. has received clinical trial support from Amgen. sions: presentation and management. Best Pract Res Clin Haematol. 2008;21:205–222. A.R. has received personal fees from Amgen, Celgene, Gilead, 20. Jabbour E, Ravandi F, Kebriaei P, et al. Salvage chemoimmunotherapy with Novartis, and Pfizer. J.-M.R. has received research grants from inotuzumab ozogamicin combined with mini-hyper-CVD for patients with Amgen, Novartis, and Pfizer and personal fees from Amgen, Cel- relapsed or refractory Philadelphia chromosome-negative acute lympho- blastic leukemia: a phase 2 clinical trial. JAMA Oncol. 2018;4:230–234. gene, Jazz Pharmaceuticals, Novartis, Pfizer, Shire, and Takeda. 21. Duell J, Dittrich M, Bedke T, et al. Frequency of regulatory T cells deter- A.Z. is an employee and stockholder of Amgen. Z.Z. is an ex- mines the outcome of the T-cell-engaging antibody blinatumomab in employee and stockholder of Amgen. G.Z. is an employee and patients with B-precursor ALL. Leukemia. 2017;31:2181–2190.