Bispecific T-cell engaging (BiTE®) antibody constructs recruit T cells to target antigens independent of their T-cell receptor specificity. Blinatumomab, a CD19xCD3 BiTE® antibody construct, is approved for the treatment of relapsed/refractory (r/r) B-cell precursor acute lymphoblastic leukemia (BCP-ALL), including patients with persistence or reoccurrence of measurable residual disease (MRD). Despite superior anti-leukemic efficacy compared to cytostatic agents, a majority of patients do not respond to treatment. Biomarkers for the identification of non-responders prior to or early during treatment are unknown. However, the definition of robust biomarkers for efficacy is of high importance for clinical decision-making and might also indicate ways to improve response rates. We therefore established a comprehensive immune-monitoring program for blinatumomab patients.

We sequentially analysed peripheral blood of patients with r/r or MRD-positive disease receiving blinatumomab prior to the start of infusion and once weekly over the course of the first 28-day cycle. We determined CD3+ T cell counts and their subset distribution (CD4, CD8, naïve [TN], central memory [TCM], effector memory [TEM], and effector memory CD45RA-positive [TEMRA] T cells) by multiparameter flow cytometry (MPFC). Additionally, patient-derived T cells were cocultered with BCP-ALL cell lines (SEM and REH) at an effector: target ratio of 1:3 in presence of blinatumomab or a control BiTE® (0.5 ng/ml, respectively) for 3 days to assess their cytotoxic capacity. Blinatumomab-mediated cytotoxicity was determined by MPFC. T-cell proliferation was evaluated by MPFC (Far Red Cell tracer) after incubation with CD3/CD28 beads for 6 days.

A total of 16 patients were enrolled. Four of these patients suffered from MRD disease, whereas the remaining 12 had overt relapse of BCP-ALL. Response rates for patients with morphological relapse were 50% (4 CRMRD- and 2 CRMRD+ after the first cycle) and 75% for patients with MRD disease (3 MRD conversions, 1 MRD persistence). Absolute lymphocyte counts were not significantly different between responders and non-responders before treatment initiation (0.9 G/l and 0.7 G/l, respectively). Whereas the percentage of CD3+ T cells (of all lymphocytes) did not significantly differ between responders and non-responders on day 0, non-responders had a significantly reduced CD3+ percentage on day 7 (81.2% vs 92.9%, p=0.03). Until day 28 of the first cycle, CD3+ percentages of responders and non-responders re-converged (80.2% and 81.4%, respectively). There were no significant differences for CD4+ and CD8+ T-cell percentages prior to and over the course of the first cycle. This was also true for TN, TCM, TEM and TEMRA subset distributions.

Additionally, CD19-BiTE®-mediated cytotoxicity of patient-derived T cells was assessed against CD19 expressing target B cell lines in vitro. Interestingly, specific lysis did not differ between responders and non-responders on day 0 (89.6% vs 79.8%, p=0.89), but decreased for non-responders over the course of the first cycle (normalized AUC for cytotoxicity 149 vs 403 in responders, p=0.03). However, this observation was only true for the patients with refractory morphologic relapse, as the non-responding MRD patient maintained T cell cytotoxicity over the first cycle of therapy. Impaired T cell proliferation (leading to reduced E:T ratios) might contribute to the observed dysfunctional cytotoxicity in non-responders, as the percentage of proliferating T cells after CD3/CD28 bead-based stimulation tended to be reduced for non-responders (17.9% vs 36.1%, p=0.07).

In summary, lymphocyte counts, T cell percentage and T cell subset distributions do not allow for a response prediction prior to treatment start for patients receiving blinatumomab. BCP-ALL patients with morphological relapse/persistence who do not achieve a remission with blinatumomab therapy show reduced cytotoxicity in vitro in comparison to patients responding to treatment. As this observation could not be confirmed with one MRD non-responder in our cohort, the burden of disease might contribute to the observed T cell dysfunction, possibly by interfering with T cell proliferation. Evaluation of immune checkpoint expression on effector and target cells over the course of the therapy is currently ongoing, as are analyses of the T cell transcriptome of responders and non-responders.


Kischel:AMGEN Research (Munich) GmbH: Employment, Equity Ownership. Subklewe:Celgene: Consultancy, Honoraria; Oxford Biotherapeutics: Research Funding; Miltenyi: Research Funding; Roche: Consultancy, Research Funding; Janssen: Consultancy; Gilead: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria; Morphosys: Research Funding; AMGEN: Consultancy, Honoraria, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.