Initial treatment of chronic lymphocytic leukemia with low-dose high-frequency rituximab followed by addition of acalabrutinib

Abstract

Abstract Introduction: Rituximab is effective at removing circulating CLL cells by inducing antibody-dependent cellular phagocytosis (ADCP). This ADCP could be limited by low CLL cell CD20 levels (innate or acquired), intrinsic CLL cell resistance to ADCP, inadequate serum levels of antibody or complement, or finite innate immune cytotoxic capacity. We conducted correlative studies on samples from our phase II clinical trial (ClinicalTrials.gov NCT03788291) for treatment-naïve CLL patients with progressive disease starting with low-dose, high-frequency rituximab monotherapy followed by addition of the second-generation covalent binding Bruton tyrosine kinase inhibitor (BTKi) acalabrutinib to obtain data to improve the performance of this regimen. Methods: Thirty-eight untreated CLL patients with progressive disease received 50 mg intravenous (IV) rituximab on cycle 1 day 1 (C1D1), followed by 50 mg of subcutaneous (SQ) rituximab 48h later (C1D3), and then 50 mg SQ rituximab twice weekly for 23 weeks. Acalabrutinib (100 mg every 12h) was initiated on C1D8 and continued for at least one year. Peripheral blood samples were collected pre-treatment on day 1 (C1D1pre), one hour after start of IV infusion (C1D1-1h), at completion of infusion (C1D1post), on day 3 before and after SQ rituximab (C1D3pre, C1D3post), on day 8 before and after therapy (C1D8pre, C1D8post), on day 15 before and after therapy (C1D15pre, C1D15post), and on day 1 of each subsequent 4-week cycle until C7D1 and then on C9D1. These samples were assessed for CLL cell counts, complement C3 fragment deposition, and cell marker levels (e.g. CD20) by flow cytometry. Serum complement (CH50 titer) and rituximab levels were assayed. Single-cell RNA sequencing was performed on seven patient samples. Lymph node volume was measured clinically. In vitro ADCP assays evaluated CLL cell sensitivity to antibody-mediated cytotoxicity. Wilcoxon signed-rank tests were used for statistical analysis. Results: Low-dose IV rituximab rapidly reduced median circulating CLL cells to 15% of baseline (95% CI 8%-27%) within 1h, with no further decrease at infusion completion (p=0.94). By C1D3pre, CLL counts rebounded to 76% of baseline (95% CI 70%-83%; p<0.0001) with a slight reduction (68% of baseline) at C1D3post (95% CI 60%-80%; p=0.004), but no significant reduction (78% of baseline) at the end of the first week (C1D8pre, 95%CI 71%-90%; p=0.32). Serum rituximab was 2.2 mg/ml (95%CI 1.3-2.7) at 1h and significantly higher 10.3 mg/ml (95%CI 5.3-13.4) at C1D1post (p<0.0001). CLL cell CD20 levels decreased by C1D1-1h to 67% of baseline (95%CI 53%-73%; p=0.002) and to 42% (95%CI 33%-55%) of baseline at C1D1post (p<0.0001). CH50 decreased throughout the rituximab infusion to a nadir of 55 units/ml (95%CI 47-64; p<0.0001), which is within normal range (38.9–89.9 units/ml). Palpable lymph node volume decreased to 34% (95%CI 22%-48%) of baseline after one week. C1D8 addition of acalabrutinib mobilized CLL cells into circulation with an increase to 210% of baseline by C1D15 (p<0.0001). CLL cell counts subsequently decreased to 2.7% of baseline by C9D1 (p<0.0001). Compared to C1D8pre, CLL CD20 levels decreased to 56% (95% CI 44%-71%; p=0.0004) after 1 week of acalabrutinib therapy (C1D15pre). During this week, expression of CLL cell MS4A1 (coding for CD20) decreased to 48%(95%CI 0.44-0.53; p<0.0001). CD20 levels remained stable over the subsequent duration of rituximab therapy (C2D1 to C7D1; p=0.68) and thenCD20 levels increased following cessation of rituximab therapy (C7D1 to C9D1; p=0.004). Conclusion: One week of low dose rituximab effectively reduced CLL tumor burden. Clearance of circulating CLL cells stopped within 1h of initiation of first rituximab IV dose. This was not due to insufficient serum rituximab, low serum complement, or loss of CLL cell CD20. Alternative explanations include limited innate immune cytotoxic capacity and innate CLL cell resistance to ADCP. Resistance to ADCP is currently being tested in vitro on these samples. Addition of acalabrutinib was associated with improved treatment response and a significant decrease in expression of MS4A1 by circulating CLL cells with progressive decrease in CLL cell CD20 levels until cessation of rituximab therapy. This was likely due to BCR pathway signaling inhibition by acalabrutinib. These data suggest that BTKi could decrease the efficacy of antibody therapy and could be informative about the design of future clinical trials.