Ibrutinib-based therapy reinvigorates CD8⁺ T cells compared to chemoimmunotherapy: immune monitoring from the E1912 trial

E1912 was the first frontline phase 3 study to compare ibrutinib-based therapy (ibrutinib with rituximab) with the chemoimmunotherapy fludarabine, cyclophosphamide, and rituximab (FCR).¹ Long-term follow-up has demonstrated superior progression-free survival (PFS) and overall survival for ibrutinib-rituximab relative to those for FCR.² Nevertheless, clinical challenges remain, including the current need for continuous therapy, tolerability, and residual and progressive disease. Immunotherapy represents a powerful combination or alternative therapy to tackle resistant disease and deepen responses.^3,4 However, T-cell exhaustion is a major barrier for optimal immunotherapy.^5,6 Dysfunctionality of T cells is characterized by increased chronic lymphocytic leukemia (CLL)-associated T-cell subsets expressing inhibitory checkpoint molecules.⁷ Furthermore, helper CD4⁺ T cells have a tumor-promoting capacity, whereas impaired immune synapse formation contributes to suppressed CD8⁺ T-cell cytotoxicity.⁸ Correlative studies have revealed that BTKis reduce abnormally high T-cell numbers and checkpoint receptor expression while reducing malignant B cells.^9-13 However, the impact of BTKis on T-cell function and association with clinical response is less well defined. Here, we leverage pretreatment and on-treatment peripheral blood patient samples serially collected from the E1912 trial and report on the impact of ibrutinib-rituximab vs FCR on T cells using immune monitoring functional assays.

Viable peripheral blood mononuclear cells at baseline and 6-, 12-, and 18-month time points from the Eastern Cooperative Oncology Group–American College of Radiology Imaging Network (ECOG-ACRIN)-E1912 Cancer Research Group (supplemental Figure 1, available on the Blood website) were biobanked for longitudinal comparative immune analysis. supplemental Table^1,2 summarizes the samples studied (ibrutinib-rituximab, n = 89; FCR, n = 62).

Immune monitoring and functional assays are detailed in supplemental Methods.

All patient samples were obtained after written informed consent, in accordance with the declaration of Helsinki, and approved by the National Research Ethics Committees, ECOG-ACRIN and the National Institutes of Health.

We initially investigated the impact of therapy on T cells and explored the association with clinical outcome (Figure 1A). Flow cytometry measured the absolute numbers of naive (CD45RA⁺/CCR7⁺), central memory (CD45RA⁻/CCR7⁺), effector memory (T_EM; CD45RA⁻/CCR7⁻) and terminally differentiated effector memory (CD45RA⁺/CCR7⁻) subsets in patients at baseline and 6- and 12-month treatment time points (supplemental Table 3). This analysis revealed a reduction in the majority of CD4⁺ and CD8⁺ T-cell subsets, including naive and effectors, during ibrutinib-rituximab treatment (Figure 1B; supplemental Figure 2A,C), consistent with T-cell normalization, as previously reported for monotherapy.^10,13 Expectedly, we observed a marked decrease of subsets after FCR, with evidence of immune reconstitution at 12 months (Figure 1B; supplemental Figure 2B,D).¹⁴ The frequencies of subsets remained relatively stable during ibrutinib-rituximab treatment, whereas FCR caused naive and central memory subsets to contract, whereas T_EM expanded (supplemental Figure 2E-F). Both therapies reduced the number of regulatory T cells, T helper 17 (T_H17) cells, and natural killer (NK) cells compared with those at baseline, but an increased the regulatory T-cell to CD4 ratio after FCR was observed (supplemental Figure 3A-C).¹⁵ Strikingly, patients on ibrutinib-rituximab with higher T-cell numbers, including PD-1⁺ effector CD8⁺ and CD4⁺ subsets, at baseline had longer PFS (Figure 1D,F), suggesting the importance of an existent but exhausted immune response before therapy. Interestingly, higher levels of PD-L1–expressing CLL cells at baseline correlated with favorable PFS (Figure 1D,F; multivariable analysis in supplemental Table 4). Furthermore, an elevated frequency of effector CD8⁺ T cells at the 6-month ibrutinib-rituximab time point associated with favorable PFS, whereas no association was detectable at 12 months (Figure 1D,F). Conversely, higher T-cell numbers correlated with worse PFS in the FCR arm, whereas increased NK-cell frequency at baseline associated with favorable outcome (supplemental Figure 4A). Consistent with tumor-mediated exhaustion, greater numbers of PD-1⁺ and PD-L1⁺ CD8⁺ T-cell subsets associated with higher measurable residual disease (MRD) during ibrutinib-rituximab treatment (supplemental Figure 5A). In contrast, elevated frequencies of T-cell subsets not expressing checkpoint molecules, including CD8⁺ terminally differentiated effector memory and NK cells, correlated with low MRD during ibrutinib-rituximab treatment, in keeping with reduced exhaustion. An association between T cells and MRD was less evident in the FCR arm, except for increased checkpoint-expressing T cells at 12 months, which correlated with higher MRD (supplemental Figure 5B). Ibrutinib’s inhibition of interleukin-2–inducible T-cell kinase (ITK) enhanced T_H1 polarization,^16,17 but both therapeutic arms reduced T_H1 and T_H2 numbers and T_H1:T_H2 ratios (supplemental Figure 6A-B). Nevertheless, an increased frequency of T_H2 and CD4:CD8 ratio (baseline and 6 months) associated with unfavorable PFS and incidence of infection, respectively, during ibrutinib-rituximab treatment (Figure 1F-G). However, increased effector CD8⁺ T-cell numbers and CD16⁺ NK cells at 6 months were associated with no infections during ibrutinib-rituximab treatment (Figure 1G). Conversely, increased T cells after FCR correlated with infections (supplemental Figure 4B). In sum, higher CD8⁺ T-cell numbers at baseline and early on-therapy, associated with favorable clinical responses, whereas PD-1⁺ and PD-L1⁺ subsets associated with greater MRD during ibrutinib-rituximab treatment.

Next, we characterized the cytolytic function of therapy-reshaped T cells against baseline CLL cells (Figure 2A-B). T cells from both 6- and 12-month ibrutinib-rituximab time points showed enhanced killing function compared with those at pretreatment levels. In contrast, T cells after FCR showed no cytolytic improvement. Notably, patients who experienced grade 3 infections during ibrutinib-rituximab treatment showed lower anti-CLL T-cell cytotoxic function (Figure 2C). Hypothesizing altered T-cell:CLL interactions, we then performed conjugation assays. T cells during ibrutinib-rituximab treatment showed augmented formation of polarized F-actin synapses with baseline CLL cells (Figure 2D; supplemental Figure 7A). In comparison, T cells after FCR exhibited distinctly nonpolarized synapses (Figure 2E). Given the opposing roles of patient CD4⁺ and CD8⁺ T cells,⁸ we examined these subsets and detected an increased frequency of granzyme B⁺ CD8⁺ T-cell:CLL synapses at both ibrutinib-rituximab time points compared with that at baseline, in which CD4⁺ T-cell:CLL synapses dominated. This switch in the CD4⁺:CD8⁺ synapse balance was not detected after FCR (Figure 2F-G; supplemental Figure 7B-D). Interestingly, in keeping with protumoral CD4⁺ T cells, increased formation of CD4⁺ T-cell:CLL F-actin–positive synapses at baseline correlated with unfavorable PFS and grade 3 infections during ibrutinib-rituximab administration (Figure 2H; supplemental Figure 7E). Together, these data demonstrate that ibrutinib-rituximab promotes previously exhausted CD8⁺ T-cell activity, which could provide a gateway for immunotherapy.

Ibrutinib is known to reduce PD-1 expression on patient T cells.^9,10,18 Here, we detected a reduced frequency of PD-1–expressing T-cell subsets during ibrutinib-rituximab administration as well as PD-L1–expressing T cells except for CD8⁺ T_EM at 6 months (Figure 1D-E; supplemental Figure 8). In contrast, T-cell PD-1/PD-L1 expression was relatively unaffected after FCR. This prompted us to investigate for a checkpoint blockade in our cytotoxicity assay (Figure 2I-K; supplemental Figure 9). Both ibrutinib-rituximab– and FCR-exposed T cells were insensitive to anti–PD-1. However, anti–PD-L1¹⁹ increased anti-CLL T-cell cytotoxicity only at the 6-month ibrutinib-rituximab time point, suggesting a narrow window for the checkpoint blockade activity.¹⁷ This led us to investigate whether the T-cell–engaging bispecific antibody glofitamab (CD20 × CD3)²⁰ could trigger improved cytolytic responses. T cells from all ibrutinib-rituximab time points tested up to 18 months showed increased anti-CLL T-cell killing after treatment with glofitamab, compared with those at baseline (Figure 2L-N). However, T cells after FCR did not respond to glofitamab, including at the later time point. Overall, these data support the ability of ibrutinib-based therapy to enhance T-cell–mediated cytotoxicity induced by bispecific immunotherapy.

In summary, our data highlight the importance of T cells during ibrutinib-rituximab therapy, with higher T-cell numbers and rejuvenated cytotoxicity accompanying favorable clinical responses. Our exploratory findings that increased levels of PD-1–expressing T cells as well as PD-L1–expressing CLL cells before therapy associate with longer PFS suggests that ibrutinib-rituximab appears to capitalize on T-cell–mediated immune surveillance in patients. Strikingly, opposing associations were found in the chemoimmunotherapy arm, and T cells showed no functional improvement after FCR. Previous studies have reported CD8⁺ T clonotype expansion during ibrutinib therapy,^21,22 likely reflecting active immunosurveillance. Taken together, tumor debulking and alleviation of T-cell exhaustion during BTKi-based therapy^9-13 may promote CD8⁺ T-cell activity. The switch from CD4⁺ T-cell:CLL interactions at baseline to CD8⁺ lytic synapses during ibrutinib-rituximab therapy supports this concept. Although ibrutinib-rituximab did not increase T_H1 numbers, we do not exclude ITK inhibition contributing to beneficial immunomodulation.¹⁸ Furthermore, our longitudinal assays designed to evaluate changes in T-cell cytolytic function with therapy, revealed that revitalized cytotoxicity during ibrutinib-rituximab therapy could be maximized with glofitamab, further supporting combination immunotherapy approaches.^23-25 Overall, this report underscores the importance of trial-associated science to understand how BTKis modulate T cells and supports the development of immunotherapy-based therapies.

The authors thank the Nikon Imaging Facility at King’s College London. This study was coordinated, in part, by the ECOG-ACRIN Cancer Research Group (Peter J. O'Dwyer and Mitchell D. Schnall, joint group chairs).

This study was supported by the National Institutes of Health, National Cancer Institute awards RO1CA193541, U10CA180820, UG1CA232760, UG1CA233290, U10CA180794, and RO1CA251801.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Contribution: D.P. designed, performed experimental work, analyzed data, and wrote the manuscript; X.V.W. performed data and association analysis; C.E.L. and N.I. performed experimental work; G.D.R. contributed to image analysis; T.D.S., N.E.K., and M.S.T. designed and supervised the study and managed data; S.H., M.B., and C.K. provided the bispecific antibodies and advised on the study; A.G.R. designed and supervised the study and wrote the manuscript.

Conflict-of-interest disclosure: S.H., M.B., and C.K. are employees of and declare patents and stock ownership with Roche Glycart AG. A.G.R. has received research support from Roche Glycart AG to Institution as part of a research collaboration. The remaining authors declare no competing financial interests.

Correspondence: Alan G. Ramsay, Lymphoma Immunology, Innovation Hub, Guy’s Cancer Centre, Great Maze Pond, London SE1 9RT, United Kingdom; email: alan.ramsay@kcl.ac.uk.