Functional Defects of T Cells of NHL Patients after Different Chemotherapy Regimens Activated By CD19/CD3 Tetravalent Bispecific TandAb^® AFM11

Introduction:

T-cell engaging immunotherapies such as bispecific T-cell recruiting antibodies such as the recently approved blinatumomab, or chimeric antigen receptor T-cells (CAR-T) have emerged as highly active therapeutics in patients with refractory or relapsed hematological malignancies such as ALL or NHL. However, a relevant number of heavily pretreated patients progress or do not respond to these novel therapies. The objective of this study was to determine whether patients´ treatment history impacts the T-cell engagement of novel immunotherapies by analyzing activity of AFM11 - a CD19-directed tandem diabody (TandAb^®) construct with T-cells obtained from patients after different chemotherapeutic regimens (R-Bendamustine, R-CHOP, HD-BEAM).

Methods:

T-cells were isolated and enriched from NHL patients 4-6 weeks after different therapeutic regimens and characterized side by side with T-cells enriched from healthy volunteers by flow cytometry. The responsiveness of T-cells from NHL patients to AFM11 was compared with T-cells from healthy volunteers in proliferation and cytokine release assays. In addition, enriched T-cells were used as effector cells at limiting effector-to-target (E:T) ratios in heterologous cytotoxicity assays with NALM-6 target cells in the presence of AFM11 or an appropriate control TandAb.

Results:

We found less CD3⁺ cells (median 148.1 cells/µL in patient group vs. 1232.2 cells/µL in healthy donors), less NK-cells (median 35.9 in patient group vs. 217.8 cells/µL in healthy donors), and no B-cells in the selected patients. Among memory T-cells, the percentages of both central (CD45RO⁺/CD62L⁺) and effector memory (CD45RO⁺/CD62L^-) CD4⁺ and CD8⁺ cells were significantly increased in patient PBMC. Finally, we examined the percentages of regulatory T-cells (Treg) among the CD4⁺ T-cells of patients and healthy donors. Overall, patients contained higher proportions of Treg. Patients who received R-Bendamustine or HD-BEAM treatment had more Treg, while the percentage of Treg in patients after R-CHOP was comparable to healthy donors.

Whereas CD8⁺ T-cells from patients and healthy volunteers showed similar proliferative response upon AFM11 stimulation in the presence of target cells, CD4⁺ T-cells from patients proliferated significantly more than CD4⁺ T-cells from healthy volunteers. No difference was observed among the patient groups with different prior chemotherapeutic regimens.

T-cells from patients produced significantly higher amounts of IFN-γ than T-cells of healthy donors upon stimulation with AFM11 and target cells. In contrast, the production of IL-10 was significantly lower by T-cells from patients when compared to healthy donors. Surprisingly, most of the T-cells from healthy donors and patients did not produce IL-6.

In heterologous cytotoxicity assays T-cells from healthy donors induced statistically significant higher specific lysis of NALM-6 in the presence of AFM11. There was no statistically significant difference between T-cells isolated from patients after different chemotherapies at an E:T ratio of 1:2.

However, at lower E:T ratios there was statistically significant difference among T-cells from patients after different chemotherapy regimens. T-cells isolated from patients who received R-CHOP treatment had the strongest lysis capacity compared to the two other groups. The lowest efficacy of AFM11-mediated target cell lysis was observed with T-cells from patients after treatment with HD-BEAM.

Conclusions:

In conclusion, T-cells of NHL patients after different chemotherapy regimens are reduced in number and have functional defects. However, AFM11 is able to activate patient T-cells for potent target cell lysis with similar efficacy as T-cells from healthy volunteers at higher E:T ratios. Only at limiting effector cell counts a lower efficacy was observed for T-cells from patients. T-cells from R-CHOP treated patients are the most active among the tested treatment groups and display similar responsiveness as T-cells from healthy donors. Lower activity was measured with T-cells from R-Bendamustine pretreated patients and substantial lower cytotoxic activity for T-cells from patients after HD-BEAM treatment. The correlation of these findings with in vivo response will be evaluated in an ongoing Phase I trial with AFM11.