Treatment of secondary CNS lymphoma using CD19-targeted chimeric antigen receptor (CAR) T cells

Abstract

Background: Aggressive B cell lymphoma with secondary central nervous system (CNS) involvement (SCNSL) carries a dismal prognosis. Chimeric antigen receptor (CAR) T cells (CAR-T) targeting CD19 have revolutionized the treatment for B cell lymphomas; however, only single cases with CNS manifestations successfully treated with CD19 CAR-T have been reported.

Methods: We prospectively enrolled 4 patients with SCNSL into our study to assess clinical responses and monitor T cell immunity.

Results: Two of four SNCSL patients responded to the CD19-targeted CAR-T. Only one patient showed a substantial expansion of peripheral (PB) CAR-T cells with an almost 100-fold increase within the first week after CAR-T. The same patient also showed marked neurotoxicity and progression of the SNCSL despite continuous surface expression of CD19 on the lymphoma cells and an accumulation of CD4⁺ central memory-type CAR-T cells in the CNS. Our studies indicate that the local production of chemokine IP-10, possibly through its receptor CXCR3 expressed on our patient's CAR-T, could potentially have mediated the local accumulation of functionally suboptimal anti-tumor T cells.

Conclusions: Our results demonstrate expansion and homing of CAR-T cells into the CNS in SNCSL patients. Local production of chemokines such as IP-10 may support CNS infiltration by CAR-T cells but also carry the potential of amplifying local toxicity. Future studies investigating numbers, phenotype, and function of CAR-T in the different body compartments of SNSCL patients receiving CAR-T will help to improve local delivery of "fit" and highly tumor-reactive CAR-T with low off-target reactivity into the CNS.

Keywords: B cell lymphoma; CAR-T cells; CNS lymphoma; Cytokines; Neurotoxicity; T cells.

Fig. 1

Antigen expression on tumor tissue. A Immunohistochemical analyses performed on tissue from the mandibular buccal mucosa of patient 2 at approximately 2 years prior to CAR-T cell therapy. Hematoxylin- and eosin-stained sections (left; 400 × magnification) are shown. Arrows indicate areas with large cells with frequent mitoses. The tumor cells were stained for expression of CD19 (right; 400 × magnification). B Analysis of CNS-infiltrating B cell lymphoma cells by flow cytometry in patient 3 at approximately two months prior to CAR-T cell therapy. The infiltrating tumor cells were stained for surface expression of CD19 antigen. C Analysis of CNS-infiltrating B cell lymphoma cells by flow cytometry in patient 4 at day + 9 (left) and day + 21 (right) after CAR-T cell therapy. Again, the infiltrating tumor cells were stained for surface expression of CD19 antigen

Fig. 2

CAR-T cell expansion and persistence in patients with secondary CNS lymphoma post CD19 CAR-T cells. Time course of A reconstitution of absolute neutrophil counts (ANC) and absolute lymphocyte counts (ALC), B serum C-reactive protein (CRP) and ferritin levels, and C CAR-T cell numbers after lymphodepleting chemotherapy and CAR-T cell infusion. D Dot plots showing peripheral blood CAR-T cells in 3 of 4 patients at different timepoints post CAR-T infusion. CAR-T cells were identified by staining of the expression of the CAR on the cell surface and costaining with anti-CD3 and other T cell markers (Supplemental Table 1)

Fig. 3

CNS infiltration by CD19-targeted CAR-T cells in a SCNSL patient with ICANS. A Proportions of CD4⁺ and CD8⁺ CAR-T cells were determined in the CSF (upper panel) and in the PB (lower panel) of patient 4 shortly after onset of potentially immune-mediated CNS toxicity using flow cytometry. Dot plots show CNS-infiltrating CAR-T cells at day + 8 (CSF) and day + 9 (PB), respectively. CAR-T cells were identified by staining of the expression of the CAR on the cell surface and costaining with anti-CD3, anti-CD4, anti-CD8, and other T cell markers (Supplemental Table 1). CAR-T cell memory subtypes were determined by costaining for CD45RA and CD62L. Central memory (CM) CAR-T cells are shown in the right lower quadrant. B Concentrations of 22 different T cell-related cytokines/chemokines were determined in our patient on days + 8 (left) and + 9 (right), respectively, at the onset of potentially immune-mediated CNS toxicity using CodePlex Secretome technology. Results are shown as absolute concentrations in pg/mL. C Histograms show surface expression of receptors involved in CNS-directed homing of T cells on PB CAR-T cells (red histograms) and non-CAR-T (gray histograms) from our patient measured on day + 7. The dot plot on the right shows proportions of peripheral blood CAR-T cells expressing α4β1 integrin required for the entry of T cells into the CNS. D Increased levels of CD27 and CD127 were found on PB CAR-T cells (red histograms) vs. non-CAR-T (gray histograms) as measured on day + 9. In addition, cytoplasmic granzyme B and surface levels of CXCR3 were determined on day + 14 post CAR-T cell treatment in PB CAR-T cells. E Expression of exhaustion markers on PB CAR-T cells (red histogram) from our patients compared to their own non-CAR-T (gray histogram) on day + 9 post CAR-T cell treatment

Fig. 4

CNS imaging results post CD19 CAR-T cells. MRIs were taken on A patient 3 and B patient 4 at the timepoints indicated. In the case of patient 3, repeat imaging showed a decrease in the size of the brain lesion post CAR-T. For patient 4, imaging showed a new lesion in the left posterior parietal lobe