Predictive biomarkers with isatuximab plus pomalidomide and dexamethasone in relapsed/refractory multiple myeloma

Dear Editor, Multiple myeloma (MM) is the second most common hematologic disease worldwide. CD38 is a type II transmembrane glycoprotein that is highly expressed on MM cells, and functions both as a receptor and as a multifunctional ectoenzyme. CD38 monoclonal antibodies (mAbs) exhibit anti-MM activity via multiple mechanisms of action and are being evaluated in all stages of therapy. However, not all patients respond to CD38 mAb therapy; variability in response may be explained by both hostrelated and tumor-related factors. The identification of biomarkers with predictive value for response could help to optimize and personalize the treatment of patients with relapsed/refractory multiple myeloma (RRMM). At present, CD38 receptor density (RD) on MM cells is the only biomarker with an association with clinical response to CD38 antibody therapies. Isatuximab, an IgG-kappa anti-CD38 mAb is effective as a single agent and is well tolerated in patients with RRMM. It targets tumor cells via multiple mechanisms including Fcdependent immune effector mechanisms such as antibodydependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis, and complement-dependent cytotoxicity, as well as direct apoptosis. Besides initial anti-MM activity through antibody-dependent mechanisms, isatuximab also has long-term immunomodulatory effects through decrease in T regulatory (Treg) cells, increase in T-cell clonality, and induction of myeloma-specific immunity. Data were derived from patients enrolled in two clinical studies that have been previously described. Study 1 (NCT02283775) was a phase-1b dose-escalation study of isatuximab in combination with pomalidomide and dexamethasone (Isa-Pd) in patients with RRMM; Study 2 (ICARIA-MM study; NCT02990338) was a randomized, active-controlled, phase-3 study evaluating Isa-Pd vs. Pd. In this exploratory analysis, we investigated whether baseline biomarkers including CD38 RD on bone marrow plasma cells, Fc immunoglobulin receptor (FCGR3A) genotype, bone marrow, and peripheral blood immunophenotypes have predictive value for treatment benefit to Isa-Pd in Studies 1 and 2. Biomarker analyses conducted for each study are summarized in Fig. 1A. Baseline peripheral blood samples were taken prior to first treatment in both studies. In addition, a bone marrow plasma cell sample was taken during screening in Study 1. CD38 RD in bone marrow plasma cells and immune cell markers in blood and bone marrow samples were analyzed by multiparametric flow cytometry. Immune cell populations including B cells (CD19 B cell), T cells (CD3 T cell, CD4 T cell, Tregs), and NK cells (CD56 bright CD16 low subset and CD56 dim CD16 bright subset) were characterized. The proportion of cells positive for a given marker or set of markers was correlated with response to isatuximab combination treatment. In addition, blood samples from both studies were analyzed for germline FCGR3A genotyping (V158 high-affinity and F158 low-affinity alleles). Biomarker levels were correlated with treatment response, defined as at least partial response according to International Myeloma Working Group criteria. Singlenucleotide substitution at amino acid position 158 in the FCGR3A gene (known as FCGR3A-V158F


Dear Editor,
Multiple myeloma (MM) is the second most common hematologic disease worldwide 1 . CD38 is a type II transmembrane glycoprotein that is highly expressed on MM cells, and functions both as a receptor and as a multifunctional ectoenzyme 2 . CD38 monoclonal antibodies (mAbs) exhibit anti-MM activity via multiple mechanisms of action and are being evaluated in all stages of therapy. However, not all patients respond to CD38 mAb therapy; variability in response may be explained by both hostrelated and tumor-related factors 3 . The identification of biomarkers with predictive value for response could help to optimize and personalize the treatment of patients with relapsed/refractory multiple myeloma (RRMM). At present, CD38 receptor density (RD) on MM cells is the only biomarker with an association with clinical response to CD38 antibody therapies 4 .
Isatuximab, an IgG-kappa anti-CD38 mAb is effective as a single agent and is well tolerated in patients with RRMM 5,6 . It targets tumor cells via multiple mechanisms including Fcdependent immune effector mechanisms such as antibodydependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis, and complement-dependent cytotoxicity, as well as direct apoptosis 7 . Besides initial anti-MM activity through antibody-dependent mechanisms, isatuximab also has long-term immunomodulatory effects through decrease in T regulatory (Treg) cells, increase in T-cell clonality, and induction of myeloma-specific immunity 8,9 . Data were derived from patients enrolled in two clinical studies that have been previously described 6,10 . Study 1 (NCT02283775) was a phase-1b dose-escalation study of isatuximab in combination with pomalidomide and dexamethasone (Isa-Pd) in patients with RRMM 6 ; Study 2 (ICARIA-MM study; NCT02990338) was a randomized, active-controlled, phase-3 study evaluating Isa-Pd vs. Pd 10 . In this exploratory analysis, we investigated whether baseline biomarkers including CD38 RD on bone marrow plasma cells, Fc immunoglobulin receptor (FCGR3A) genotype, bone marrow, and peripheral blood immunophenotypes have predictive value for treatment benefit to Isa-Pd in Studies 1 and 2.
Biomarker analyses conducted for each study are summarized in Fig. 1A. Baseline peripheral blood samples were taken prior to first treatment in both studies. In addition, a bone marrow plasma cell sample was taken during screening in Study 1. CD38 RD in bone marrow plasma cells and immune cell markers in blood and bone marrow samples were analyzed by multiparametric flow cytometry. Immune cell populations including B cells (CD19 + B cell), T cells (CD3 + T cell, CD4 + T cell, Tregs), and NK cells (CD56 + bright CD16 + low subset and CD56 + dim CD16 + bright subset) were characterized. The proportion of cells positive for a given marker or set of markers was correlated with response to isatuximab combination treatment. In addition, blood samples from both studies were analyzed for germline FCGR3A genotyping (V158 high-affinity and F158 low-affinity alleles). Biomarker levels were correlated with treatment response, defined as at least partial response according to International Myeloma Working Group criteria. Singlenucleotide substitution at amino acid position 158 in the FCGR3A gene (known as FCGR3A-V158F polymorphism) generates allotypes with different binding affinities of NK cells to tumor-bound IgG antibody. The binding of NK cells (via ligation of their low-affinity Fc receptor [CD16a] to an IgG antibody) is enhanced by the presence of a valine at position 158 (V/V or V/F) as compared with homozygous phenylalanine genotype (F/ F), thereby leading to a higher level of NK-cell-mediated ADCC 11 . Patients were categorized as responders (patients with a best overall response of at least partial response) or non-responders.
Statistical analyses for Study 1 and Study 2 are detailed elsewhere 6,10 . Progression-free survival (PFS) and overall survival (OS) were analyzed by the Kaplan-Meier method. Other secondary endpoints were summarized using descriptive statistics. In the current analysis, Pvalues between groups were calculated using Wilcoxon rank-sum test.
FCGR3A genotypes were available for 44/45 patients in Study 1, 138/154 patients in the Isa-Pd arm, and 139/ 153 patients in the Pd arm of Study 2. Across both studies, the frequency of the F158F and F158V genotypes of the FCGR3A gene was equal at 42% each, whereas the V158V genotype occurred at 16% frequency (Fig. 1B) To test whether CD38 RD has predictive value for response to Isa-Pd treatment, we evaluated CD38 RD for responders and non-responders in Study 1. Baseline CD38 RD was measured by quantitative flow cytometry in bone marrow samples from 31 out of 45 patients. The median CD38 RD was 108,172 (range 12,950-337,335) receptors/ cancer cell (Fig. 1D). There was a trend for the median CD38 RD value to be higher in patients responding to Isa-Pd (120,931 receptors/cancer cell, range [48,770-337,335], n = 21) vs. non-responders (85,370 receptors/cancer cell, , and Tregs) were tested to determine if a response to isatuximab treatment could be predicted (Table 1). There was no significant difference between responders and nonresponders for the immune cell markers in baseline blood samples, indicating that peripheral blood baseline levels of B-cell, T-cell, and NK-cell subsets were not predictive of response to Isa-Pd. Furthermore, percentages of host immune cell biomarkers were similar between responders and non-responders in bone marrow samples collected at baseline (Table 1).
In the current analysis, we explored the predictive value of baseline biomarkers including bone marrow plasma cell CD38 RD, FCGR3A genotype, and immune cell markers in blood/bone marrow for response to Isa-Pd treatment. We showed that there is no clear association between Fc polymorphism and isatuximab treatment outcome. Notably, both ORR and PFS benefit with Isa-Pd vs. Pd treatment was observed across all three FCGR3A genotypes (V/V, V/F, and F/F), consistent with that in the overall population. Despite the higher median CD38 RD in patients who responded to Isa-Pd (120,931 receptors/cancer cell in responders vs. 85,370 receptors/cancer cell in non-responders), there is not enough evidence to support the predictive value of CD38 RD. While some responders to Isa-Pd had RD as low as 48,770/cancer cell, non-responders had RD as high as 309,003/cancer cell. Furthermore, CD38 RD is high in MM; only 2% of the patient samples had a CD38 RD of <48,770/cancer cell, with the lowest being 12,950 receptors/cancer cell. These data indicate that CD38 RD is not a predictive biomarker and cannot be used to guide treatment decisions with Isa-Pd.
Preclinical studies showed that isatuximab eliminates CD38 + Tregs and restores T-cell and NK-cell-mediated antitumor immune responses 9 . In the current analysis, we explored whether baseline levels of immune cell subsets including NK cells, T cells, and B cells could predict clinical outcome with isatuximab treatment. No significant difference was observed between responders and non-responders to isatuximab treatment for the tested immune biomarkers both in peripheral blood as well as in bone marrow plasma cell samples, indicating that immunophenotyping does not predict response to isatuximab treatment.
Limitations of the current study include the small subsets of patients evaluated for some of the included analyses.
In conclusion, we did not find a significant association between tumor response and the evaluated baseline biomarkers (bone marrow plasma cell CD38 RD, peripheral blood cell FCGR3A genotypes, and peripheral blood or bone marrow immune cell populations) in patients with RRMM treated with Isa-Pd. Isa-Pd provides a consistent benefit to patients with RRMM, including difficult to treat subgroups. However, the results of this analysis indicate that prescreening patients for these clinical and biomarker characteristics may not be warranted for Isa-Pd.