Addition of Ianalumab (VAY736) to Ibrutinib in Patients With Chronic Lymphocytic Leukemia on Ibrutinib Therapy: Results From a Phase Ib Study

Abstract

Introduction:

Ianalumab (VAY736), an anti-B-cell activating factor receptor monoclonal antibody, combined with ibrutinib, significantly improved survival and reduced tumor burden in preclinical chronic lymphocytic leukemia (CLL) models.

Methods:

This phase Ib dose-escalation/expansion trial (NCT03400176) enrolled patients with CLL who did not achieve a complete response (CR) with ibrutinib or had developed resistance mutations. Patients received intravenous ianalumab (escalation: 0.3–9.0 mg/kg; expansion: 3.0 mg/kg) Q2W and continued ibrutinib (420 mg) once daily for up to 8 cycles of 28 days. The study aimed to evaluate the safety, tolerability, recommended dose, and antitumor activity of this combination.

Results:

Thirty-nine patients were treated (escalation: n=15; expansion: n=24). No dose-limiting toxicities were observed. 38.5% of the 39 patients were in CR or CR with incomplete marrow recovery at cycle 9 (C9). At C9 day 1 (D1), 17 patients (43.6%) achieved undetectable measurable residual disease in blood or bone marrow. Grade ≥3 adverse events occurred in 16 patients (41.0%), which were treatment-related in 9 (23.1%). No on-treatment deaths were reported; 1 patient died due to COVID-19 during the post-treatment period. Seventeen patients (43.6%) discontinued ibrutinib at or after C9D1 and remained off therapy for 12.1 to 24.5 months. Preliminary RNA-sequencing and flow cytometry data support both natural killer (NK) and T-cell activation with ianalumab.

Conclusions:

The combination was well tolerated, with 43.6% of patients discontinuing ibrutinib therapy. Biomarker data suggests that ianalumab increased NK and T-cell activation. These data support further evaluations of ianalumab in combination with Bruton’s tyrosine kinase inhibitors for patients with CLL.

Translational Relevance

Chronic lymphocytic leukemia (CLL) is the most prevalent adult leukemia. A current standard of care for CLL includes indefinite treatment with Bruton’s tyrosine kinase inhibitors (BTKi), such as ibrutinib, acalabrutinib or zanubrutinib, which may result in long-term toxicity. In preclinical studies, ianalumab (VAY736), a human monoclonal antibody targeting B-cell activating factor receptor–positive B cells, showed antileukemic activity; when combined with ibrutinib, it significantly reduced tumor burden and improved survival in animal models. We conducted a phase Ib dose-escalation/expansion study of ianalumab with ibrutinib in 39 patients with CLL. The combination was well tolerated with an acceptable safety profile in patients with CLL. Seventeen patients (43.6%) attained a response with undetectable measurable residual disease in blood or marrow. Biomarker data suggested that ianalumab increased activation of natural killer and T cells. These data provide evidence in favor of further studies investigating ianalumab in patients with CLL.

Introduction

Bruton’s tyrosine kinase (BTK) is a nonreceptor tyrosine kinase involved in B-cell antigen receptor signaling in both normal and transformed B lymphocytes (1). The B-cell receptor signaling pathway is critical in chronic lymphocytic leukemia (CLL), as evidenced by the success of BTK inhibitors (BTKi) in the therapy for CLL. The first-in-class approved BTKi (ibrutinib) and subsequent (acalabrutinib, zanubrutinib, and pirtobrutinib) BTKi, have demonstrated dramatic impact on the outcomes (2,3). However, BTKi therapy requires continuous, indefinite treatment, which may result in long-term toxicity including cardiac toxicities such as atrial fibrillation, bleeding and hypertension (4). In a real-world analysis trial of 616 patients treated with ibrutinib in the US, 41% of patients had discontinued ibrutinib at a median follow-up of 17 months, with the main reason being toxicity, including arthralgia, atrial fibrillation, and rash (5). Moreover, patients treated with BTKi may develop resistance, leading to disease relapse or progression. In approximately 70% of patients developing resistance to BTKi, the resistance has been attributed to mutations in BTK at the C481 drug-binding site, other sites in the BTK protein, or the downstream effector phospholipase C gamma 2 (PLCγ2) (1,6–10).

Combination therapies with ibrutinib, such as with rituximab, have shown prolongation of progression-free survival (PFS) and overall survival (OS) compared with standard chemoimmunotherapy. However, in a randomized phase 3 study for patients with untreated CLL that had both ibrutinib and ibrutinib + rituximab arms, there was no PFS benefit with the addition of rituximab (11). In contrast, the CD20 antibody, obinutuzumab, when added to acalabrutinib, improved complete response (CR) rate and PFS over acalabrutinib monotherapy in patients with previously untreated CLL (12). An update of this study also demonstrated that acalabrutinib and obinutuzumab demonstrated a prolongation of OS over chemoimmunotherapy, which was not observed with acalabrutinib alone. These data suggest that an Fc-engineered antibody with improved effector cell activation might impart the best opportunity for an antibody to be effective with a BTKi. However, the depth of response (e.g., CR or undetectable measurable residual disease) remains low in these combinations and BKTi generally is dosed continuously, highlighting the need for investigating a mechanism for improving responses to discontinue BTKi (13,14).

B-cell activating factor receptor (BAFF-R) is expressed on most B-cell receptor–positive B cells, and BAFF-R signaling has been shown to be involved in the activation, including proliferation, maturation, and survival of normal and malignant B cells (15,16). Furthermore, BAFF-R signaling plays a role in B-cell–mediated human diseases such as autoimmune disorders and B-cell malignancies, and some polymorphisms in BAFF/BAFF-R genes have been associated with increased risk of developing CLL (17).

Ianalumab (VAY736) is a human monoclonal antibody targeting BAFF-R–positive B cells for elimination via antibody-dependent cell-mediated cytotoxicity (ADCC) (18). Ianalumab has several mechanisms of action: (1) by blocking BAFF ligand from binding to BAFF-R, thus blocking BAFF-R and downstream NF-kappa B signaling, which prevents B-cell differentiation, proliferation, and survival; (2) by depleting B cells by mediating potent ADCC through binding to both NK cells in the periphery and in the CLL tumor microenvironment; and (3) by triggering macrophage antibody-dependent cellular phagocytosis of CLL cells (18,19). Ianalumab has demonstrated antileukemia activity in preclinical CLL models that is significantly superior to that of anti-CD20 monoclonal antibodies, including rituximab and obinutuzumab, which specifically bind to B cell-specific cell surface antigen CD20 and induce killing of normal and malignant B cells via ADCC (18). BAFF/BAFF-R interactions activate an alternative NF-κB–dependent survival signaling compared to the NF-κB signaling inhibited by BTK inhibitors such as ibrutinib (18,20,21), meaning that ianalumab can inhibit NF-κB synergistically with BTKi and target a different antigen other than CD20. This led to the hypothesis that a combination of ibrutinib with the BAFF-R inhibitor ianalumab may improve the efficacy of ibrutinib treatment.

In in-vitro models of human CLL cells and in vivo CLL mouse models, ianalumab demonstrated antileukemia activity that, when combined with ibrutinib, could significantly reduce tumor burden and improve the survival in the mouse leukemia model, suggesting that this combination may augment the antileukemia response. This potentially could allow patients to discontinue ibrutinib and reduce the toxicity related to continuous therapy (18). This paper presents the results from the first phase Ib dose-escalation/expansion study of ianalumab in combination with ibrutinib in patients with CLL (NCT03400176).

Patients and Methods

Study design

This is a phase Ib, open-label, multicenter (from 7 centers in the US) dose-escalation and dose-expansion study evaluating the safety, tolerability, recommended dose for expansion (RDE), pharmacokinetics (PK), pharmacodynamics, and antitumor activity of the combination of ianalumab with ibrutinib in adult patients with CLL. Patients were enrolled from April 15, 2018, to April 12, 2022.

In the escalation part, patients received the standard, approved dose of oral ibrutinib (420 mg daily) in all cohorts. Ianalumab doses of 0.3, 1.0, 3.0, and 9.0 mg/kg, given intravenously, once every 2 weeks (q2w), were tested in consecutive cohorts, supported by Bayesian analysis conducted ahead of each new dose cohort. The dosing cycle was 28 days. Ianalumab starting dose of 0.3 mg/kg q2w in patients with CLL was selected based on simulations of PK–B-cell dynamics using prior ianalumab data in immunology indications, along with adjustments to B-cell attributes and rate constants for patients with CLL (22,23). An optimal dose of ianalumab 3.0 mg/kg in combination with ibrutinib was identified based on data and analysis gathered during the dose escalation.

In the expansion part, patients received ianalumab 3.0 mg/kg q2w intravenously and were enrolled into 2 arms depending on whether they had ibrutinib-resistance mutations present at enrollment (arm A, patients without ibrutinib resistance mutations; arm B, patients with ibrutinib resistance mutations in their CLL).

The on-treatment period was up to 8 cycles of 28 days (C8D28). Patients received ianalumab in combination with ibrutinib (420 mg or highest tolerated dose, defined prior to study enrollment) for up to 6 cycles of 28 days. After 6 cycles, patients with a CR discontinued ianalumab and received ibrutinib for an additional 2 cycles; patients without a CR after 6 cycles continued to receive ianalumab + ibrutinib for the remaining 2 cycles. In the original protocol, patients achieving undetectable measurable residual disease (uMRD) at C9D1 could discontinue ibrutinib at the investigator’s discretion; this was amended so that the treatment decisions were based on clinical response following International Workshop on CLL (iwCLL) response criteria (24) (patients with CR at C9D1 could discontinue ibrutinib)

Participants

The study recruited adult patients with CLL diagnosed as per World Health Organization classification of hematologic disorders or iwCLL guidelines (24,25). Eligible patients had received ibrutinib either as first-line treatment (single agent or in combination) or following relapse from another approved treatment; patients had either not achieved a CR after more than 1 year of ibrutinib treatment or developed a known ibrutinib-resistance mutation (mutations in BTK[C481S] and/or mutations in the immediate downstream target of BTK, PLCγ2) (26). Only the following PLCγ2 mutations were considered as functional mutations: R665, S707, A708, and L845. Patients had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2, and platelet levels of ≥25 × 10⁹/L without transfusion support within 7 days of the first dose of ianalumab. Detailed inclusion and exclusion criteria are listed in Supplementary Methods. The overall representativeness of this study population to the general CLL population is shown in Supplement Table S1.

In the dose-escalation part, patients who had been receiving ibrutinib 420 mg following relapse on another approved therapy and meeting the criteria for non-response or ibrutinib resistance mutations (BTK or PLCγ2) were included.

In the dose-expansion part, patients who had received ibrutinib at a dose <420 mg and whose response had been stable on that dose for 2 months prior to trial treatment were also included. The expansion part consisted of the following 2 arms: the first arm (arm A) included patients who had received ibrutinib either following relapse from another approved therapy or as first-line therapy (alone or in combination) and who had not achieved CR after 1 year; the second arm (arm B) included patients who had developed ibrutinib-resistance mutations after relapse or first-line therapy (alone or in combination).

Patients who received live vaccine within 4 weeks of starting ianalumab and those with a known history of HIV, or with active hepatitis B or C infection were excluded from the study.

Procedures

Outcomes

For the escalation part, the safety primary endpoints were dose-limiting toxicities (DLTs) and were reported within the first 28 days of treatment with ianalumab in combination with ibrutinib. For both escalation and expansion parts, the safety primary endpoints were the incidence and severity of AEs and serious AEs (SAEs). The tolerability primary endpoints included dose interruptions, reduction, and dose intensity (computed as the ratio of actual cumulative dose received and actual duration of exposure).

The key secondary endpoints were rate of CR/CRi (complete response with incomplete marrow recovery) at the beginning of C9 (C9D1), overall response rate (ORR) assessed by iwCLL criteria (24), and clearance of ibrutinib resistance mutations (BTK[C481S] and/or PLCγ2 hotspot), defined as less than 1% mutation -bearing alleles for the patients in the arm with ibrutinib resistance mutations at baseline. Mutation testing was performed using peripheral blood (and some bone marrow aspirate) samples for CD19+ leukocytes using Robosep-S (EasySep HLA chimerism Whole Blood CD19-positive kit, StemCell Technologies, Vancouver, Canada). Samples were collected at C1D1, C4D1, C7D1, C9D1, and after C9D1 were further assessed every three months for two years of follow-up for patients with CR, partial response (PR) or whose disease was assessed as stable disease (SD) at C9D1. Next-generation sequencing (NGS) for the coding region of BTK(C481S) and PLCγ2 was performed on all baseline samples using a fully validated custom AmpliSeq assay on the Ion Torrent S5 platform (Thermo Fisher Scientific, Waltham, Massachusetts, USA) at the CLIA-certified Ohio State University’s James Molecular Laboratory (Columbus, Ohio, USA), with bioinformatic analysis incorporating Ion Torrent suite tools, GO Workbench (Genom Oncology, Cleveland, Ohio, USA) and review of low-level variants in the Integrative Genomics Viewer (Broad Institute, Cambridge, Massachusetts, USA). Sensitivity of mutation detection was 0.5% variant allele fraction (VAF) for BTK(C481S) and PLCγ2 hotspot resistance–associated mutations (27). Subsequent samples from patients with PLCγ2 mutations were analyzed using next-generation sequencing. For monitoring of patients with BTK(C481S) mutation, BTK(C481S)-specific quantitative droplet digital polymerase chain reaction (ddPCR) was performed on the QX200 Droplet Digital PCR System (Bio-Rad Laboratories, Hercules, California, USA) using a custom assay performed in the same CLIA laboratory. Analysis and quantitation of ddPCR data were performed using QX Manager software (Bio-Rad Laboratories). A positive mutation result was reported at or above the validated sensitivity of 0.1% mutation proportion if at least 5 mutant events were observed. Separate probes recognizing the 2 common BTK(C481S) changes, c.1442G>C and the c.1441G>A, were included, as well as a wild-type probe at that location with the percentage of mutant to total BTK(C481S) events reported. Alternate droplet patterns, possibly indicative of other BTK p.C481X sequence changes, were further investigated using the BTK(C481S)-PLCγ2 sequencing assay.

Biomarker analyses

MRD in blood and bone marrow was assessed using 10-color multiparametric flow cytometry analysis (28,29). The following markers were used for the MRD assessment using flow cytometry with a 3-tube panel: CD2, CD3, CD5, CD7, CD9, CD10, CD13, CD14, CD19, CD20, CD22, CD38, CD43, CD45, CD56/16, CD79b, CD81, FMC-7, HLA DR, KAPPA, and LAMBDA. MRD in bone marrow was assessed at screening and C9D1. MRD in blood was assessed at screening, C1D1, C1D2, C1D8, C1D15, C2D1, C3D1, C4D1, C5D1, C6D15, and C9D1, and for patients who continued to take ianalumab treatment, it was assessed at C7D1 and C8D1. After C9D1, MRD was assessed every 3 months for 2 years for patients with CR, PR, or SD at C9D1. Patients were defined as having uMRD if they have blood or marrow with <1 CLL cell per 10,000 leukocytes.

Samples of human peripheral blood mononuclear cells (PBMCs) were collected for NK cell biomarker assessment by flow cytometry at C1D1 before the first dose of ianalumab treatment, 1 hour and 20 to 24 hours after the first dose, and at C9D1. PBMCs were cryopreserved and later stained for flow cytometry analysis. Briefly, up to 4 × 10⁶ cryopreserved peripheral blood mononuclear cells (PBMCs) were thawed quickly in a water bath at 37°C and washed by centrifugation with prewarmed RPMI medium containing 10% of fetal bovine serum (FBS). After washing, cells were stained with fixable viability dye eFluor^™506 (Thermo Fisher, San Diego, California, USA) at a dilution of 1:100 in wash buffer (phosphate-buffered saline containing 0.1% sodium azide and 2% FBS), followed by staining with fluorochrome-conjugated surface antibodies for 30 minutes at room temperature in the dark. Previously prepared and qualified antibody cocktails consisting of 12 surface antibodies were utilized. The list of antibodies used for this panel is provided in Supplement Table S2. (30)

After incubation, cells were washed once by centrifugation in wash buffer and fixed with 1× Fixation/Permeabilization Buffer from the Foxp3/Transcription Factor Staining Buffer Set (ThermoFisher, San Diego, California, USA) for 30 minutes at room temperature in the dark. After fixation, cells were washed once by centrifugation with 1× Permeabilization Buffer from the Foxp3/Transcription Factor Staining Buffer Set and resuspended in 1× Permeabilization Buffer. Fluorochrome-conjugated intracellular antibody (Ki-67) was added, and cells were incubated for 30 minutes at room temperature in the dark. Cells were then washed twice by centrifugation with 1× Permeabilization Buffer and resuspended in 0.5% formalin solution.

Cells were then acquired on a BD LSRFortessa X-20 equipped with 5 lasers (BD Biosciences, San Jose, California, USA), and data were analyzed using FlowJo software (FlowJo LLC, Ashland, Oregon, USA; RRID:SCR_008520). For data analysis of the natural killer (NK) cell proliferation assay, after exclusion of debris, doublets, and dead cells, NK cells were further gated for NK subsets (cluster of differentiation [CD] 56, CD16), as well as inhibition (CD159a) and activation (CD25, CD335, NK group 2D [NKG2D]) biomarkers. Monocytes were also assessed for CD14, CD16, and HLA-DR expression. Samples of PBMCs were collected for gene expression analysis of CLL cell and immune cell signaling at C1D1 before the first dose of ianalumab treatment and 1 hour and 20 to 24 hours after the first dose. Two PBMC batches were isolated and kept separately for analysis. PBMC cell isolation occurred at the clinical site for batch 1 and at the central lab for batch 2. PBMCs were cryopreserved and later sorted into CD4+ T cells, CD8+ T cells, CD56+ NK cells, and CD19+CD5+ CLL cells for in situ ultralow input full-length complementary DNA generation, followed by transcriptome library generation and sequencing (coverage-based limiting-cell experimental analysis for RNAseq [CLEAR]). For cell-based limiting-cell RANAseq, cryopreserved patient PBMC samples from 3 timepoints (predrug dose, 1 hour after dose, and 20 to 24 hours after dose) were quickly thawed in a water bath at 37°C. Cells were gently dispersed in 10 mL of media, then centrifuged at 1000 rpm for 10 minutes at 4°C. Supernatant was aspirated off and cell pellets resuspended in media for enumeration; then cells were resuspended with sterile PBS for antibody staining.

CD3, CD4, CD5, CD8, CD16, CD19, and CD56 antibodies were purchased from BD Pharmingen (PE-Cy7 Mouse Anti-Human CD3, Cat. #: 557749, RRID:AB_396855; FITC Mouse Anti-Human CD4, Cat. #: 555346, RRID:AB_395751; APC-Cy7 Mouse Anti-Human CD5, Cat. #: 563516, RRID:AB_2738249; BV510 Mouse Anti-Human CD8, Cat. #: 563256, RRID:AB_2738101; APC Mouse Anti-Human CD16, Cat. #: 561304, RRID:AB_10714780; PE Mouse Anti-Human CD19, Cat. #: 555413, RRID:AB_395813; PerCP-Cy5.5 Mouse Anti-Human CD 56, Cat. #: 560842, RRID:AB_2033964). PBMCs were stained as shown in Supplement Table S3. A portion of the cells, up to 1 million, were kept as unstained control. Healthy donor PBMCs that were obtained and cryopreserved were used across days as a staining control. Antibody-binding beads from Invitrogen (Ultra Comp eBeads Plus Compensation Beads, Reference #: 01–3333-42) were used for single color compensation controls Supplement Table S4).

Stained cells were briefly vortexed and incubated in the dark on ice for 30 minutes. Tubes were washed with 2 mL FACS buffer, then spun down at 1000 rpm for 10 minutes. After removal of the supernatant, the cells were resuspended with sterile PBS in final volumes of 500 μL and beads in 200 μL. They were stored on ice in the dark until FACS sorted.

For the In situ cDNA generation, library preparation, sequencing, and data analysis, the Clontech SMART-Seq HT (Takara Cat# 634437) kit was used for global complementary DNA generation. Cells (300 cells for most samples; 50–250 cells for a few patient cell types) were sorted directly into freshly diluted SMART-Seq HT CDS Sorting Solution (lysis component: Takara Cat# 634439; RNase Inhibitor: Takara Cat# 2313A). cDNA generation followed the manufacturer manual, except the reagents were miniaturized to 1/5th of the protocol volume for all enriched cell lysates. The quality (majority of cDNAs between 7 and 10 kb), and the quantity of purified cDNAs was assessed prior to sequencing library generation using Agilent BioAnalyzer HS DNA kit and the Invitrogen Qubit DNA HS Assay kit. Sequencing libraries were generated using the Illumina Nextera XT Kit and Nextera XT Index Kit v2 Set A following manufacturer instructions, except for the miniaturization of reagent volume to a quarter of the listed volume. Purified library products were sequenced on Illumina NovaSeq 6000 flow cells using paired-end 100 bp to a depth of 17 to 20 million clusters. FASTQ files generated for each library were trimmed using AdapterRemoval v2.2.0, RRID: SCR_011834 (31) ensuring that all sequencing adapters were removed and that the average quality score for each read was above Q20 (representing 1 in 100 Illumina base error rate). Reads, which were aligned by HISAT2 v2.2.1, RRID:SCR_015530 (32) against ribosomal RNAs, mitochondrial DNAs, or PhiX bacteriophage (Illumina spike-in control) sequences, retrieved from NCBI (RefSeq Human Release 109.20210514, RRID:SCR_003496) (33), were removed from each FASTQ file, as these do not represent gene expression signal of interest. All remaining reads were aligned against the reference human genome GRCh38p7 with HISAT2. The resulting BAM alignment files were sorted and indexed before further analysis.

Alignments were quantified using the featureCounts utility from the Subread package v2.0.3, RRID:SCR_012919 (34,35) in unstranded mode using NCBI RefSeq human genes (see above) in GTF format. Custom Python scripts were used to produce a formatted gene expression counts table from the raw output of featureCounts. RNAseq quality metrics were derived using a modification of the QuaCRS quality control workflow (36), which includes running RNA-SeQC v1.1.8.1, RRID:SCR_005120 (37) FASTQC v0.11.5, RRID:SCR_014583, and RSeQC v2.6.2, RRID:SCR_005275 (38). Finally, coverage maps of each BAM file were derived using the BEDTools “genomecov” utility v2.28.0, RRID:SCR_006646 (39).

RNAseq coverage maps were preprocessed with the CLEAR v1.0, RRID:SCR_027171 (40) workflow to determine, which genes were reliably quantified. In brief, read coverage distributions were profiled by CLEAR in a gene-by-gene manner. Per sample read coverage with a profile deviating from a Double Beta Distribution were deemed “non-CLEAR passed” and therefore not used in downstream timepoint analysis. The final counts table consisted of raw gene read counts for all RefSeq genes and NAs were added to replace read counts for “non-CLEAR passed” genes.

The following genes have been assessed for NK and T-cell activation: ETS1, KLRD1, ISG20, and PRF1 in NK cells; ETS1, CD53, ISG20, and PRF1 in CD8+ T cells; ADAM10, BIRC3, ITK, and CASP6 in CD4+ T cells. Apoptosis-related gene signature included BCL2, BIRC2, BIRC3, CASP3, CASP8, CFLAR, CYCS, NFKBIA, RELA, and TNFRSF10A genes.

Gene expression analysis was completed for each cell type population separately. The raw read counts were trimmed mean of M-values (TMM) normalized using the R package edgeR (v3.26.8, RRID:SCR_012802). Batch 1 and batch 2 datasets were combined, but colored differently, to increase sample numbers. Only patients with paired 0-hour pre-dose and 1-hour post-dose CLEAR-passed samples were included for each gene.

PK and BAFF-receptor occupancy

Pre-dose and post-dose blood samples for ianalumab PK, immunogenicity (anti-ianalumab antibodies), soluble BAFF, and BAFF-receptor occupancy (RO) were collected during ianalumab treatment. A detailed sampling schedule is provided in Supplement Table S5. These included full PK profile for ianalumab at C1 and C3D1 to D15 and immunogenicity at C1D1, C1D15, and D1 of C2 to C8 (C7 and C8 if applicable). Blood samples were also collected for PK analysis of ibrutinib at C1D1 and C1D8 at the following timepoints: pre-dose; 0.5, 2, 6, and 24 hours (C1 only) post-dose; and pre-dose on D1 of C2 to C8. Ianalumab and soluble BAFF concentrations in serum were measured using a validated sandwich ELISA assay, with lower limit of quantification of 0.025 μg/mL and 62.5 pg/mL, respectively. BAFF-R occupancy (RO) assay was a validated flow cytometry assay in whole blood. Ibrutinib concentrations in plasma were derived using a validated liquid chromatography mass spectrometry assay, with a lower limit of quantification of 0.5 ng/mL. The presence of antidrug antibodies in clinical study samples was evaluated using a validated assay in a 3-tiered approach using a bridging format electrochemiluminescence assay.

Statistical and pharmacokinetic analyses

In the escalation phase, the protocol-specified maximum tolerated dose (MTD) was defined as the highest combination dose with an estimated DLT rate below 33% in the first cycle of treatment. The Bayesian logistic regression model provided an estimate of ianalumab and ibrutinib combination not exceeding the MTD, which is typically the highest combination with less than 25% risk of excessive toxicity. The recommended dose for expansion (RDE) was defined as a dose less than or equal to MTD, with the most appropriate benefit-risk assessment according to the investigators/sponsor, based on the safety and tolerability, PK, and pharmacodynamics data. The full analysis set consisted of all patients who received at least 1 dose of study treatment.

The dose-determining set included all patients from the full analysis set who met the minimum exposure criterion of taking the planned doses of ianalumab for the first 28 days and 75% of the planned ibrutinib doses, had sufficient safety evaluations, or experienced a DLT during C1 (in the first 28 days of dosing). PK analysis set included all patients who provided an evaluable PK profile, which was considered as the following conditions being satisfied: the patient had received one of the planned treatments, provided at least 1 primary PK parameter, and did not vomit within 4 hours after ibrutinib dosing. PK parameters of ianalumab and ibrutinib were derived with WinNonLin^® Phoenix^® Version 8.3.4 (Certara, New Jersey, US). PK parameters of ianalumab included total systemic exposure during the dosing interval (AUCtau), peak serum concentration (Cmax), time of maximum concentration observed (Tmax), and elimination half-life (T1/2).

The dose proportionality of ianalumab over the dose range of 0.3 mg/kg to 9.0 mg/kg at Cycle 1 and Cycle 3 was evaluated by fitting a power model.

Descriptive summary statistics were used for summarizing continuous data, and frequencies and percentages were used for categorical data. Overall response rates and their exact 90% confidence intervals (CIs) were calculated based on a binomial distribution. Time to progression (TTP) is the time from start of study treatment to the date of event which is defined as the first documented progression (defined as overall disease response assessment of progressive disease [PD]) or death due to underlying cancer. If a patient has not had an event, TTP is censored at the date of last adequate disease assessment. Assessments taken after the start of alternative cancer therapy will be excluded. TTP was estimated using the Kaplan Meier method

Ethical oversight

This study was conducted in accordance with the International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) Harmonized Tripartite Guidelines for Good Clinical Practice, with applicable local regulations, including European Directive 2001/20/EC and US Code of Federal Regulations Title 21, and the Declaration of Helsinki. The protocol and all amendments were reviewed and approved by a properly constituted Institutional Review Board/Independent Ethics Committee/Research Ethics Board (IRB/IEC/REB) before study start. Eligible patients provided written, IRB/IEC/REB-approved informed consent prior to study start.

Data availability

Data are available upon reasonable request. Novartis will not provide access to patient-level data if there is a reasonable likelihood that individual patients could be reidentified. Phase 1 studies, by their nature, present a high risk of patient re-identification; therefore, patient individual results for phase 1 studies cannot be shared. In addition, clinical data, in some cases, have been collected subject to contractual or consent provisions that prohibit transfer to third parties. Such restrictions may preclude granting access under these provisions. Where co-development agreements or other legal restrictions prevent companies from sharing particular data, companies will work with qualified requestors to provide summary information where possible.

Results

Patient characteristics and disposition

A total of 39 patients were enrolled in the escalation part (n = 15) and the expansion part (n = 24), with expansion consisting of arm A for patients without ibrutinib resistance mutations (n=19) and arm B for patients with ibrutinib resistance mutations in their CLL (n=5).

Baseline patient demographics and characteristics are shown in Table 1. The median age was 65 years and 92.3% of patients had an ECOG performance status of 0. Twelve patients (30.8%), all of whom were enrolled in the ianalumab 3.0 mg/kg + 420 mg ibrutinib expansion part, had no prior therapies aside from ibrutinib before enrollment, and the median (min-max) number of prior therapies aside from ibrutinib was 1 (0–14). Of the enrolled patients, 8 patients (20.5%) had BTK(C481S) mutations, 4 (10.3%) had PLCγ2 mutations, and 3 (7.7%) had both BTK(C481S) and PLCγ2 mutations. In terms of baseline CLL characteristics, 6 patients (15.4%) had 17p deletion, 32 patients (82.1%) had unmutated immunoglobulin heavy chain variable (IGHV) region genes, and 20 patients (51.3%) had a complex karyotype (≥5 abnormalities). At screening, as assessed by CT scans, 19 patients had one or more measurable lymph nodes present for which the sum of the product of perpendicular diameters (SPD) of up to six lymph nodes was derived. The median SPD was 464.0 mm², with a range of 0 to 4343 mm². The absolute lymphocyte count at baseline was median (range) 5.45 (0.47–202.16) 10⁹/L; 21 patients had ALC> 5 × 10⁹/L.

Table 1.

Patient baseline characteristics.

	Ianalumab 0.3 mg/kg+ Ibrutinib (420 mg)	Ianalumab 1.0 mg/kg+ Ibrutinib (420 mg)	Ianalumab 3.0 mg/kg+ Ibrutinib (280 mg)	Ianalumab 3.0 mg/kg+ Ibrutinib (420 mg)	Ianalumab 9.0 mg/kg+ Ibrutinib (420 mg)	All patients
Demographic variable	N=4	N=3	N=1	N=27	N=4	N=39
Median age, years (range)	70.5 (61–75)	60.0 (60–63)	57.0 (57–57)	66.0 (39–82)	63.5 (54–79)	65.0 (39–82)
Sex, n (%)
Male	4 (100)	2 (66.7)	1 (100)	20 (74.1)	1 (25.0)	28 (71.8)
Race, n (%)
White	4 (100)	2 (66.7)	1 (100)	26 (96.3)	3 (75.0)	36 (92.3)
Black or African American	0	1 (33.3)	0	0	1 (25.0)	2 (5.1)
Missing	0	0	0	1 (3.7)	0	1 (2.6)
ECOG performance status, n (%)
0	3 (75.0)	3 (100)	1 (100)	25 (92.6)	4 (100)	36 (92.3)
1	1 (25.0)	0	0	2 (7.4)	0	3 (7.7)
Prior antineoplastic regimens, n (%)
Yes	4 (100)	3 (100)	1 (100)	27 (100)	4 (100)	39 (100)
Prior ibrutinib status, n (%)
Only ibrutinib	0	0	0	12 (44.4)	0	12 (30.8)
At least one prior therapy and ibrutinib	4 (100)	3 (100)	1 (100)	15 (55.6)	4 (100)	27 (69.2)
Number of prior regimens excluding ibrutinib
Median (range)	2.5 (1–4)	1.0 (1–2)	3.0 (3–3)	1.0 (0–6)	4.0 (2–14)	1.0 (0–14)
Baseline ibrutinib resistance mutation status – Any Mutation,* n (%)	1 (25.0)	3 (100)	1 (100)	7 (25.9)	3 (75.0)	15 (38.5)
BTK(C481S)	1 (25.0)	2 (66.7)	0	4 (14.8)	1 (25.0)	8 (20.5)
PLCy2	0	0	1 (100)	2 (7.4)	1 (25.0)	4 (10.3)
L845	0	0	1 (100)	0	1 (25.0)	2 (5.1)
D334	0	0	0	0	0	0
D1140	0	0	0	0	0	0
R665	0	0	0	3 (11.1)	1 (25.0)	4 (10.3)
S707	0	1 (33.3)	0	1 (3.7)	0	2 (5.1)
BTK(C481S) + PLCy2	0	1 (33.3)	0	1 (3.7)	1 (25.0)	3 (7.7)
Other	0	1 (33.3)	0	2 (7.4)	0	3 (7.7)
Dohner classification#
Chromosome 17p deletion	0	2 (66.7)	0	3 (11.1)	1 (25.0)	6 (15.4)
Chromosome 11q deletion	1 (25.0)	0	0	6 (22.2)	2 (50.0)	9 (23.1)
Trisomy 12	0	1 (33.3)	0	2 (7.4)	0	3 (7.7)
Chromosome 13q deletion	3 (75.0)	0	0	7 (25.9)	0	10 (25.6)
None	0	0	1 (100)	9 (33.3)	1 (25.0)	11 (28.2)
IGHV mutant status
Mutant	1 (25.0)	0	0	5 (18.5)	0	6 (15.4)
Non-mutant	2 (50.0)	3 (100)	1 (100)	22 (81.4)	4 (100)	32 (82.1)
Missing	1 (25.0)	0	0	0	0	1 (2.6)
Complex Karyotype
Yes	3 (75.0)	3 (100)	0	11 (40.7)	3 (75.0)	20 (51.3)
Absolute lymphocyte count (10 9 /L) at screening
Median (range)	8.31 (1.56–11.3)	37.28 (1.56202.16)	0.47 (0.470.47)	5.28 (1.1–60.1)	15.02 (1.0274.65)	5.45 (0.47–202.16)
Sum of product of diameters (SPD) at screening (mm 2 ), n	2	2	1	12	2	19
Median (range)	541.5 (222–861)	500.0 (476–524)	2290.0 (2290–2290)	272.0 (0–4343)	1026.5 (4–2049)	464.0 (0–4343)
Time from start of ibrutinib therapy to the first study treatment (months) Median (range)	20.76 (2.0–63.0)	49.58 (14.9–66.3)	7.43(7.4–7.4)	24.97 (6.6–67.2)	64.44 (23.0–100)	24.97 (2.0–100)

^*The ibrutinib resistance mutations measured in BTK(C481S) and PLCy2 are not mutually exclusive. BTK(C481S) and PLCy2 (any variant) genetic mutations were assessed per patient. Patients with multiple specific classifications, or multiple PLCy2 variants, are counted in all corresponding rows. Other mutations include CD79A and M1141K. Patients with other mutations also had PLCy2.

^#The Dohner classification categories are defined as follows: patients with a 17p deletion; patients with an 11q deletion without a 17p deletion; patients with Trisomy 12 without a 17p deletion or an 11q deletion; patients with a 13q deletion without a 17p deletion, Trisomy 12, or an 11q deletion; patients with none of the four abnormalities.

BTK, Bruton’s Tyrosine Kinase; ECOG, Eastern Cooperative Oncology Group; IGHV, immunoglobulin heavy chain variable; PLCγ2, phospholipase C gamma 2; SPD, sum of product of diameters.

The median duration of exposure to ianalumab was 5.52 months (range: 1.8–7.8 months), and 53.8% of patients (21/39) had exposure between 5 and <6 months. During this trial the median duration of exposure to ibrutinib was 7.39 months (range: 1.8–8.5 months), and 74.4% of patients (29/39) had exposure between 7 and <8 months during the on-treatment period.

Completion of ianalumab treatment was defined as reaching C6D28, and completion of ibrutinib treatment was defined as reaching C8D28. Overall, 32 patients (82.0%) reached C6D28, 7 patients (18.0%) discontinued ianalumab before C6D28, and the primary reason for ianalumab discontinuation was progressive disease (10.3%, 4/39). Three other patients discontinued ianalumab early due to an AE (elevated amylase), subject decision, and physician decision. Thirty patients (76.9%) received ibrutinib until C8D28, with 9 patients (23.1%) discontinuing ibrutinib primarily due to progressive disease (12.8%, 5/39). Eleven patients (28.2%) were treated with ianalumab after C6D28 for an additional 2 cycles; 10 of these patients completed the full 2 cycles and 1 patient (2.6%, 1/39) discontinued ianalumab due to physician’s decision.

Safety

No DLTs were observed up to the highest tested ianalumab dose of 9.0 mg/kg q2w. AEs of any grade were reported in all patients (Figure 1 and Table 2). Hyperglycemia, any grade, was reported in 16 patients (41.0%), occurred exclusively in the first cycle, did not require intervention, and is considered to be due to premedication with steroid (100 mg prednisolone or equivalent was recommended per protocol based on institutional guidelines). Grade ≥3 AEs were reported in 16 patients (41.0%) with the most common being neutropenia/decrease in neutrophil count (15.4%), followed by increase in lipase (10.3%), hypophosphatemia (7.7%), decrease in lymphocyte count (5.1%), and amylase increase (5.1%). Two patients treated with ianalumab 3.0 mg/kg q2w + ibrutinib 420 mg experienced grade ≤ 2 infusion-related reactions.

Figure 1. Adverse events in the dose escalation and expansion phases of the study, regardless of cause or ianalumab dose, in ≥10% of patients.

AE, adverse event.

Table 2.

Adverse events, regardless of study drug relationship, in dose escalation and expansion phases

	Ianalumab 0.3 mg/kg Q2W + Ibrutinib 420 mg	Ianalumab 1.0 mg/kg Q2W + Ibrutinib 420 mg	Ianalumab 3.0 mg/kg Q2W + Ibrutinib 280 mg	Ianalumab 3.0 mg/kg Q2W + Ibrutinib 420 mg	Ianalumab 9.0 mg/kg Q2W + Ibrutinib 420 mg	All patients
Adverse events	N=4	N=3	N=1	N=27	N=4	N=39
Patients with at least 1 event, any grade, n (%)	4 (100)	3 (100)	1 (100)	27 (100)	4 (100)	39 (100)
Treatment-related, n (%)	2 (50.0)	2 (66.7)	1 (100)	15 (55.6)	1 (25.0)	21 (53.8)
Patients with at least 1 event, grade ≥3, n (%)	1 (25.0)	2 (66.7)	1 (100)	12 (44.4)	0	16 (41.0)
Treatment-related, n (%)	0	1 (33.3)	1 (100)	7 (25.9)	0	9 (23.1)
Most common grade ≥3 AEs, n (%) *
Neutropenia/Neutrophil count decreased	0	2 (66.7)	1(100)	3 (11.1)	0	6 (15.4)
Lymphocyte count decreased	0	0	0	2 (7.4)	0	2 (5.1)
Anemia	0	0	0	1 (3.7)	0	1 (2.6)
Hypophosphatemia	0	1 (33.3)	0	2 (7.4)	0	3 (7.7)
Lipase increased	0	0	0	4 (14.8)	0	4 (10.3)
Leukocytosis	0	1 (33.3)	0	0	0	1 (2.6)
Noncardiac chest pain	0	0	0	1 (3.7)	0	1 (2.6)
White blood cell count decreased	0	1 (33.3)	0	0	1	1 (2.6)
Lymphocyte count increased	1 (25.0)	0	0	0	0	1 (2.6)
Amylase increased	0	0	0	2 (7.4)	0	2 (5.1)
Hypokalemia	0	0	0	1 (3.7)	0	1 (2.6)
Hypermagnesemia	0	1 (33.3)	0	0	0	1 (2.6)
Hypertension	0	0	0	1 (3.7)	0	1 (2.6)
Aortic aneurysm	0	0	0	1 (3.7)	0	1 (2.6)

^*A patient may experience multiple AEs.AE, adverse event.

There was no observed relationship between the dosage of ianalumab and the occurrence of AEs. One patient in arm B of the expansion phase discontinued treatment due to a grade 4 AE, asymptomatic amylase increase (suspected to be treatment related). The amylase level normalized 15 days from the start date of the AE.

Serious AEs were reported in 2 patients. One patient in the ianalumab 3.0 mg/kg + ibrutinib 420 mg group reported grade 4 neutropenia/neutrophil count decreased (suspected to be treatment related) and grade 3 noncardiac chest pain (not suspected to be treatment related), and another patient in the ianalumab 1.0 mg/kg + ibrutinib 420 mg group reported grade 4 neutropenia/neutrophil count decreased (suspected to be treatment related). All patients with neutropenia or decreased neutrophil count were able to continue receiving treatment and did not experience neutropenia recurrence (of note, per protocol, the use of growth factors such as G-CSF is permitted according to institutional guidelines. Only three patients received this treatment while on the study). No on-treatment deaths were reported during the study. One patient died due to COVID-19 during the post-treatment period. The patient died more than a year after completing the ianalumab treatment while still on treatment with ibrutinib.

Any-grade infections and infestations (reported using Medical Dictionary for Regulatory activities version 25 system organ class) were reported in 18 patients (46.2%), and there were no grade ≥3 infections reported during the on-treatment period.

Efficacy

Efficacy was assessed in 39 evaluable patients from the dose escalation and expansion parts. Overall response rate (ORR) during dose escalation and expansion are shown in Table 3 A and B. The ORR during the dose escalation phase was 40% (90% CI: 19.1%, 64%). Among the 15 patients assessed in escalation, 6 (40%) achieved CR, 6 (40%) achieved SD, and 3 (20%) had PD. The ORR during the dose expansion phase was 70.8% (90% CI: 52.1%, 85.4%). Among the 24 patients assessed in expansion, 8 (33.3%) achieved CR, 8 (33.3%) achieved PR, and 6 (25%) achieved SD. Overall response at C9D1 (or before discontinuation) was 59% for all patients. Responses at C9D1 are shown for all patients in Figure 2A and by study part in Figure 2B and are summarized by prior therapy (ibrutinib only versus at least one therapy and ibrutinib) in Supplement Table S6.

Table 3.

Overall response rate.

(a) Best overall response in dose escalation phase and all patients
	Ianalumab 0.3 mg/kg Q2W + Ibrutinib 420 mg	Ianalumab 1.0 mg/kg Q2W + Ibrutinib 420 mg	Ianalumab 3.0 mg/kg Q2W + Ibrutinib 420 mg	Ianalumab 9.0 mg/kg Q2W + Ibrutinib 420 mg	All patients in dose escalation	All patients
	N=4	N=3	N=4	N=4	N=15	N=39
	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)
Best overall response
CR	2 (50.0)	0	2 (50.0)	2 (50.0)	6 (40.0)	14 (35.9)
CRi	0	0	0	0	0	1 (2.6)
PR	0	0	0	0	0	8 (20.5)
SD	1 (25.0)	2 (66.7)	2 (50.0)	1 (25.0)	6 (40.0)	12 (30.8)
PD	1 (25.0)	1 (33.3)	0	1 (25.0)	3 (20.0)	4 (10.3)
Overall response rate (CR, CRi, or PR)	2 (50.0)	0	2 (50.0)	2 (50.0)	6 (40.0)	23 (59.0)
90% CI	(9.8, 90.2)	(0, 63.2)	(9.8, 90.2)	(9.8, 90.2)	(19.1, 64.0)	(44.6, 72.3)
CI, confidence interval; CR, complete response; CRi, complete response with incomplete marrow recovery; PR, partial response; SD, stable disease; PD, progressive disease

(b) Best overall response in dose expansion phase and all patients
	Ianalumab 3.0 mg/kg Q2W Arm A + Ibrutinib 420 mg	Ianalumab 3.0 mg/kg Q2W Arm B + Ibrutinib 280 mg	Ianalumab 3.0mg/kg Q2W Arm B + Ibrutinib 420 mg	All patients in dose expansion	All patients
	N=19	N=1	N=4	N=24	N=39
	n (%)	n (%)	n (%)	n (%)	n (%)
Best overall response
CR	8 (42.1)	0	0	8 (33.3)	14 (35.9)
CRi	1 (5.3)	0	0	1 (4.2)	1 (2.6)
PR	7 (36.8)	0	1 (25.0)	8 (33.3)	8 (20.5)
SD	3 (15.8)	0	3 (75.0)	6 (25.0)	12 (30.8)
PD	0	1 (100)	0	1 (4.2)	4 (10.3)
PD Overall response rate (CR, CRi, or PR)	16 (84.2)	0	1 (25.0)	17 (70.8)	23 (59.0)
90% CI	(64.1, 95.6)	(0, 95.0)	(1.3, 75.1)	(52.1, 85.4)	(44.6, 72.3)
The 90% confidence interval (CI) was calculated using the exact method. CI, confidence interval; CR, complete response; CRi, complete response with incomplete marrow recovery; PR, partial response; SD, stable disease; PD, progressive disease.

Figure 2. Overall response at C9 (A) overall and (B) during dose escalation and dose expansion phases.

C, cycle; CR, complete response; CRi, complete response with incomplete marrow recovery; PR, partial response.

In the expansion (ianalumab 3.0 mg/kg + ibrutinib), the CR/CRi response rate at C9D1 was 37.5% (90% CI: 21.2, 56.3) with CR achieved in 8 patients (33.3%), CRi in 1 patient (4.2%), PR in 6 patients (25%), and SD in 3 patients (12.5%). Nine patients (47.4%, 9/19) in expansion arm A achieved CR/CRi and none of the expansion patients with a baseline ibrutinib resistance mutation (arm B) achieved a CR/CRi (Supplement Table S7).

Seventeen patients (43.6%) elected to discontinue ibrutinib at or after C9D1 following CR and/or uMRD and, during the 2-year follow-up period, were able to remain off therapy between 12.1 months and 24.5 months.

During dose escalation, 11 patients (73.3%) progressed and the median TTP was 19.4 months (95% CI: 2.8, 25.1). Four patients (26.7%) with no progression were censored at the date of the last adequate assessment. The Kaplan Meier estimated TTP rates at 6 months and 12 months were 71.8% (95% CI: 41.1, 88.4) and 64.6% (95% CI: 34.7, 83.5), respectively. During dose expansion, 4 (16.7%) TTP events were observed, and the median TTP was not evaluable (95% CI: 26.3, NE). Twenty patients (83.3%) were censored at the date of the last adequate assessment. The Kaplan Meier estimated TTP rates at 6 months and 12 months were consistent at 91.0% (95% CI: 68.6, 97.7).

At C9D1, 17 patients (43.6%) had uMRD at the threshold of 10⁻⁴ in blood or bone marrow. Of the 17 patients who attained uMRD, all of them achieved it in blood. Thirteen patients (33%) had uMRD in both their bone marrow and blood, while 4 patients achieved uMRD solely in the blood. Among these 17 patients, 7 patients had received ibrutinib only prior to study enrollment, and 10 patients received treatment prior to ibrutinib. Among the 17 patients who achieved uMRD at C9D1, 4 (23.5%) TTP events were observed. Thirteen (76.5%) patients were censored. The Kaplan-Meier estimates (%) TTP rate (95% CI) at 6 months and 12 months were 93.8% (63.2, 99.1). In contrast, for the 22 patients who did not achieve uMRD at C9D1, 11 (50%) TTP events were observed, with 11 patients (50%) censored, and the estimated TTP rate was 69.2% (43.3, 85.1) at 12 months. Out of 13 patients who achieved uMRD in bone marrow at C9D1, 2 (15.4%) TTP events were observed. Eleven (84.6%) patients were censored, and the median TTP was not evaluable (95% CI: 26.3, NE). The ORR in these 13 patients was 92.3% with 8 (61.5%) patients achieving CR and 4 (30.8%) patients achieving PR.

In the full set of 39 patients, the median best percentage change from baseline up to C9D1 in blood MRD was −99.16% (range: −100.0% to −16.7%), and median percent change from baseline to C9D1 in bone marrow was −99.98% (range: −100.0% to 1346.3%) (Figure 3). Changes in the levels of CLL cells in bone marrow (as measured by flow cytometry) from baseline to C9D1, for patients with or without baseline ibrutinib resistance mutations, are displayed in Supplement Figure S1. From baseline to C9D1 in blood, reductions in B-cell concentrations were often observed in the first few cycles of treatment and remained at low levels through C9D1 in the ianalumab 3.0 mg/kg + ibrutinib arm patients without baseline ibrutinib resistance mutations, compared to fewer reductions observed in the patients with mutations (Supplement Figure S2).

Figure 3. Best percentage change from baseline in (A) bone marrow MRD, and (B) blood MRD, for individual patients in the dose escalation and expansion phases of the study.

BTK, Bruton’s tyrosine kinase; MRD, minimal residual disease; PLCγ2, phospholipase C gamma 2.

PK and BAFF-receptor occupancy

Time course of ianalumab concentration during C1 and C3 and PK parameters for ianalumab are shown in Supplement Figure S 3 A–B and Supplement Table S8, respectively. Ianalumab PK was largely linear across the tested doses, when baseline B-cell levels were less than 10,000 cells/μL. Patients with higher baseline disease burden (baseline B-cell levels higher than 10,000 cells/μL) at lower doses had lower than expected trough concentration in the initial cycles, suggesting signs of target-mediated drug disposition.

Ianalumab total and maximal exposure (AUCtau and Cmax, respectively) did not show any major deviation from dose proportionality from 3.0 mg/kg to 9.0 mg/kg at both Cycle 1 and Cycle 3. The variability in PK exposure at both cycles was moderate, as shown by the geometric mean CV% (ranging from 13.3% to 59.8% for Cmax and from 6.8% to 31.8% for AUCtau). There was a moderate accumulation in total systemic exposure from Cycle 1 to Cycle 3, ranging from 1.52-to 1.76-fold across doses. The elimination half-life ranged from 2.31 to 5.28 days at Cycle 1 and from 5.04 to 6.74 days at Cycle 3.

Ibrutinib PK was in line with published data in patients with CLL. Soluble BAFF in circulation has been reported to increase rapidly after dosing across ianalumab clinical studies, likely as a result of displacement from its binding to BAFF-R on B cells and loss of BAFF-R expressing cells (41,42). In this study, the median soluble BAFF increased 2- to 5-fold 2 hours post-dose, similarly across doses, suggesting extensive target binding. Pooled data across all dose levels shows a median fold change of 2.67, with lower and upper quartiles of 1.64 and 5.34 respectively. Peripheral BAFF-RO measurements were between 95% and 99% in >90% of samples, indicating the extensive target saturation in circulation across the tested dose range (Supplement Figure S4 A–B).

Immunogenicity

Of the 39 patients, majority (36 patients, 92.3%) were anti-drug antibody (ADA)-negative at baseline, and only 3 patients (7.7%) were ADA-positive at baseline. None of the patients (whether ADA-negative or ADA-positive at baseline) had positive ADA post-dose. Therefore, there were neither treatment-induced nor treatment-boosted immunogenic responses to ianalumab during the study (Supplement Table S9). The ADA prevalence at baseline was 11.1% (3/27) in the ianalumab 3.0 mg/kg + ibrutinib 420 mg expansion group (and 0% in the other treatment groups).

Dose determination

No DLTs were observed at any of the dose levels investigated; therefore, the MTD was not reached. The RDE was established based on all available clinical data in addition to data from PK modeling. A population PK model was developed (in Monolix 2021R2) to describe the PK using a 2-compartment model with linear disposition. The model predicted linear PK at doses of 3.0 mg/kg q2w and above, as well as in patients with high tumor burden at baseline (e.g., 48, 000 cells/μL). Assuming the in vitro equilibrium dissociation constant of ianalumab, the serum and hypothetical disease-related tissue RO were simulated and used to estimate levels of target saturation. The serum RO was predicted to be saturated at the lowest dose of 0.3 mg/kg q2w (Supplement Figure S4 A). Assuming a 30% of tissue penetration (43,44), the predicted tissue RO at steady state was >99% at 3.0 mg/kg q2w in >95% of patients (Supplement Figure S4 B). Therefore, integrated analysis of safety, efficacy, and PK (population PK and RO) supported the selection of a RDE of 3.0 mg/kg q2w. At this dose, PK was anticipated to be linear independent of baseline disease burden, and the target to be fully saturated both in circulation and relevant tissues. This dose showed acceptable tolerability profile and signs of clinical response.

Biomarkers

At baseline, 15 patients had BTK and/or PCLγ2 mutations, with either one clone detected (n=9) or between 2 to 4 clones of either mutation detected (n=6); full details on baseline clones and VAFs are presented in Supplement Table S10.

Following treatment with the combination of ianalumab and ibrutinib, the majority of patients with a BTK mutation at baseline whose mutation status was re-evaluated at C9D1 continued to display a detectable BTK mutation at the end of treatment (n = 8/8 with a BTK[C481S] mutation; n = 3/4 with a PCLγ2 mutation); however, in 1 patient with baseline PLCγ2 mutation, detectable mutant clones were eradicated at C9D1. Nevertheless, the patient presented with the mutant clones again during evaluations at M9 and M15, coinciding with relapse of the disease.

For patients with multiple clones, all baseline dominant clones (determined by the maximum VAF among all mutations) remained dominant over time; non-dominant clones were excluded from subsequent analyses. Median baseline VAF for BTK mutations (n=11) was 26.8% (range: 1.2 to 96.9) and for PCLγ2 mutations (n=7) was 1.6% (range: 0.3–10.9). For 11 patients with at least one post-baseline VAF assessment, VAF changes for PCLγ2 mutations tended to be stable over time, whereas VAF changes for BTK mutations exhibited heterogeneity, with VAF increases in 4 out of 9 patients at or prior to time of progression (Supplement Figure S5).

PBMC analysis

PBMCs from patients treated with ianalumab (3.0 mg/kg) were analyzed by flow cytometry to determine levels of activated NK cells (NKp46-positive and NK group 2D [NKG2D]-expressing NK cells) (Supplement Figure S6). At C1D1 1 hour post-dose, the frequency of peripheral NKp46-positive NK cells increased at least 50% after ianalumab 3 mg/kg treatment in approximately half of the patients compared to C1D1 pre-dose. Patients in SD tended to show no or little change in NKp46 at 1 hour post-dose or at D2 post-dose, while patients in CR and PR tended to show increases at 1 hour post-dose or D2 post-dose. Baseline counts of NK cells and NKG2D-expressing NK cells in the blood were higher in responders than in non-responders (Supplement Figure S7). It should be noted that sample sizes are small, so only trends can be observed in these data.

RNA sequencing

Preliminary coverage-based limiting-cell experiment analysis for RNA-sequencing (RNAseq; CLEAR) data from 14 patients support peripheral NK and T-cell activation with ianalumab. At C1D1, 1 hour post-dose, RNAseq analysis showed an increase in effector-related gene expression in NK cells (ETS1, KLRD1, ISG20, and PRF1) (Supplement Figure S8 A), in CD8+ T cells (ETS1, CD53, ISG20, and PRF1) (Supplement Figure S8 B), and in CD4+ T cells (IL6ST, ADAM10, BIRC3, ITK, and CASP6) (Supplement Figure S8 C) with ianalumab treatment, suggesting increased effector function of these cells. RNAseq analysis showed evidence of apoptosis induction-related gene expression (TP53INP1 and apoptosis-related gene signature) in CLL cells (Supplement Figure S9 A–B).

Antibody levels in the blood were also measured at baseline and on treatment. Overall, IgA, IgG, and IgM levels did not significantly drop following initial treatment, and those levels were maintained over the course of treatment (Supplement Figure S10).

Discussion

The ianalumab and ibrutinib combination was well tolerated, appeared active, and had an acceptable safety profile enabling dose expansion.

Ianalumab exposure did not show any major deviation from dose proportionality from 3.0 mg/kg to 9.0 mg/kg. There was extensive target saturation in circulation across tested doses. Based on PK data and modeling, biomarker data along with clinical safety and efficacy, the RDE was established as 3.0 mg/kg q2w.

The long-term persistent detectable MRD has significant implications for patients, as they are required to continue ibrutinib treatment in the setting of persistent detectable MRD to maintain disease control. In this study, 43.6% of patients attained uMRD status in blood or bone marrow, which is a sufficiently deep response to discontinue treatment. CR with or without complete bone marrow recovery was numerically more common among patients in the expansion phase who were receiving first-line ibrutinib at the time of study enrollment. There is a potential for discontinuation of ibrutinib after uMRD has been achieved, allowing for the avoidance of toxicity that may occur due to the indefinite therapy of BTKi. The rate of achieving uMRD seen in the study is comparable to other similar studies in patients with CLL, where another therapy is added to ibrutinib (45–47). Patients receiving BTKi are immunocompromised and are at risk of infections (48). In this study, the infection rate appeared to be lower in patients treated with ianalumab and ibrutinib (46.2%, any-grade infection and no grade ≥3 infections reported) compared with those treated with single-agent BTKi (≥70% for any grade and ≥24% for grade ≥3 infections) (49, 50). The data suggests that the addition of ianalumab to ibrutinib did not increase the infection rates. Overall, the safety profile of ianalumab added to ibrutinib is well tolerated and aligns with findings from other studies where additional therapies are added to ibrutinib in CLL patients (45–47).

A major limitation of ibrutinib therapy is the potential that ibrutinib resistance mutations may occur such as BTK(C481S), which eliminates covalent binding and also greatly decreases the affinity of ibrutinib for BTK and ability to inhibit BTK activation of downstream target PLCγ2.(26) Resistance to treatment increases the risk that disease progression will occur in the future. Most patients enrolled in this study are BTK(C481S)/PLCγ2 wildtype; thus, conclusions related to how baseline BTK mutations influenced treatment outcomes or how mutation status altered the treatment cannot be drawn. Furthermore, another limitation of the study is lack of longer-term follow-up of the patients, leaving uncertainty on how viable a discontinuation strategy will be.

ADCC is primarily exerted by NK cells, and we demonstrate evidence of NK cell activation with treatment in this study. RNAseq data suggested increased activation of effector function in NK cells and T cells. The NK cell-mediated ADCC is also involved in other CD20 antibodies, including rituximab (51,52). Patients who responded to treatment tended to have acute increases in activated NK, suggesting that effective ianalumab treatment increases NK cell activation, likely due to positive feedback mechanisms through non-cell autonomous activation (e.g., cytokines). In addition, responders tended to have higher levels of NK cells and activated NK cells at baseline. These data together support the proposed mechanism of action of ADCC-mediated killing of tumor cells by NK cells. T-cell activation in the context of an ADCC antibody occurs either via cytokine secretion from activated NK cells or is directly mediated through binding to CD16 expressed on CD8+ T cell (53,54) in agreement with our RNA-seq data in pre- and post-ianalumab treatment. Therefore, the anti-leukemia activity of ianalumab likely is exerted through multiple immune-mediated effects, such as by ADCC-mediated killing from NK and T cells. However, given that this was a novel and exploratory analysis, we did not collect subsequent samples to perform further analysis. A larger study is warranted to correlate T-cell activation with a response. RNAseq data also demonstrated evidence of apoptosis induction in CLL cells following initial treatment with ianalumab. The mechanism of apoptosis induction on CLL cells cannot be confirmed with the current methods, as CLL cells could experience apoptosis either as a result of cytotoxic activity from NK or T cells, or through blockade of BAFF ligand binding to BAFF-R and subsequent decrease in NF-kB signaling (18).

Addition of ianalumab to covalent BTKi in the treatment of CLL holds potential. The combination could offer a safer therapeutic option with improved selectivity and reduced side effects and possibly allow some patients to discontinue BTKi after achieving uMRD, resulting in shorter treatment duration and lower toxicity. As combination therapy with BTKi for fixed durations is becoming increasingly utilized and data from other studies show that antibodies such as obinutuzumab can improve PFS (12), a strategy of adding ianalumab to ibrutinib could be beneficial in deepening disease responses and improving PFS or allowing patients to stop BTKi therapy with minimal toxicity. While this study was with ibrutinib, based on the mechanism we would expect the same results with other covalent BTKi that are now more recommended due to improved cardiovascular toxicities.

In this Phase Ib study, the combination of ianalumab and ibrutinib was well tolerated and resulted in almost half the patients achieving uMRD status in blood or bone marrow. The present results suggest that adding ianalumab to ibrutinib for the treatment of patients with CLL may be a viable strategy to reduce the duration of BTKi therapy. However, a larger study is necessary to confirm this therapeutic hypothesis.