Humoral immune reconstitution following therapy with daratumumab, carfilzomib, lenalidomide, and dexamethasone (Dara‐KRd), autologous hematopoietic cell transplantation, and measurable residual disease‐response‐adapted treatment cessation

Abstract Quadruplet induction, autologous hematopoietic cell transplant (AHCT), and measurable residual disease (MRD) response‐adapted consolidation yield an unprecedented depth of response in newly diagnosed multiple myeloma. Patients treated on MASTER (NCT03224507) ceased therapy and entered active surveillance (MRD‐SURE) after achieving MRD negativity. This study characterizes quantitative changes in the immunoglobulin (Ig) gene repertoire by next‐generation sequencing and serum gamma globulin levels. Quadruplet therapy leads to profound hypogammaglobulinemia and reduction in the Ig gene repertoire. Immune reconstitution (IR) is delayed in patients who received post‐AHCT consolidation compared to those who do not. Eighteen months after treatment cessation, there was no statistically significant difference between the groups.

Multiple myeloma (MM), the second most common hematologic malignancy, is routinely managed with continuous therapy and is considered an incurable disease. Rapid drug development has led to a dramatic improvement in the overall survival of most MM patients over the past two decades. Modern regimens combining immunomodulatory (IMiD) drugs, proteasome inhibitors (PI), and monoclonal antibodies (mAb) are generally well tolerated and allow patients to achieve deep and durable responses associated with improved survival, particularly with the achievement of undetectable measurable residual disease (MRD) [1]. Quadruplet induction therapies, including a CD38 mAb, IMiD, PI, and steroid followed by autologous hematopoietic cell transplant (AHCT), yield an unprecedented depth of response and MRD-negativity among patients with newly diagnosed MM [2][3][4], and patients who achieve sustained MRD-negativity after initial therapy have a very low rate of disease progression [5,6]. While the positive impact of quadruplet therapy on the depth of disease response is clear, the effect of intensive therapy on immune function and immunoparesis recovery, a proposed positive predictor of long-term progression and survival [7,8], has not been described.
The MASTER trial (NCT03224507) evaluated the feasibility of intensive therapy followed by MRD-adapted consolidation and/or treatment cessation with MRD surveillance. This allowed some patients to enjoy a lengthy treatment-free interval and provides a unique opportunity to evaluate immune reconstitution in patients with a deep disease response who are observed off of therapy. The study procedures and results of the primary analysis have been previously reported in detail [3]. Briefly, patients received induction with daratumumab, carfilzomib, lenalidomide, and dexamethasone In this study, we retrospectively analyzed the impact of DaraKRd and AHCT on biomarkers of humoral immunity among patients in MASTER for whom complete data were available at key time points.
The primary objective of this study was to characterize quantitative changes in the repertoire of immunoglobulin (Ig) genes (IgH, IgK, and  Figure 1A. We also individually reviewed charts for evidence of grade ≥3 infections and evaluated routinely measured parameters that are associated with immune reconstitution. Due to differences in the standard laboratory assessments of patients not on therapy, this analysis was limited to laboratory parameters specified by the protocol and consisted of serum gamma globulin levels (as measured on serum protein electrophoresis [SPEP]) and leukocyte subset populations (as assessed by routine complete blood count) at key time points.
Bone marrow aspirate samples were evaluated by NGS by Adaptive Biotechnologies. Unique IgH, IgK, or IgL sequences in each patient sample were identified and quantified per standard ClonoSeq® methodology [9], and these data were extracted from the report appendix. Data are expressed as median unique sequences (mus)/10 6 .
We first evaluated the impact of Dara-KRd on humoral immunity in the peri-transplant time period. Sixty-three patients who underwent induction, AHCT, and at least 4 cycles of Dara-KRd consolidation had evaluable data sets. We found that the Ig gene repertoire expands following AHCT and contracts following post-AHCT consolidation, as shown in Figure 1B Patients entering MRD-SURE after AHCT had significantly higher gamma globulin levels as measured by SPEP at each measured time point compared to those receiving Dara-KRd consolidation. Notably, a rise in the mean gamma fraction level was seen immediately for patients entering MRD-SURE after AHCT, whereas there was a 6month delay in Ig recovery for patients who received post-AHCT consolidation ( Figure 1D). These data, taken together with the Ig sequence repertoire, suggest that Dara-KRd consolidation delays immune reconstitution and that the immune repertoire expands prior to gamma globulin recovery. We also evaluated trends in leukocyte subsets, including neutrophils, lymphocytes, and monocytes. No significant trends were noted in these analyses (data not shown).
Finally, we extracted clinically significant infections from the clinical trial database, supplementing the available data with a chart review.
We focused on grade ≥3 infections, due to the low capture rate of grade 1-2 infections for patients who were off therapy and, there-  Figure 1E.
To our knowledge, these data are the first description of patterns of humoral immune reconstitution among patients with MM who discontinue therapy after achieving sustained MRD-negativity. Our use of NGS to describe the recovery of the immune microenvironment following intensive therapy is also novel. We have shown that quadruplet, anti-CD38 mAb-containing therapy leads to profound hypogammaglobulinemia and reduction in the Ig gene repertoire, which promptly recovers after AHCT. Patients who received post-AHCT consolida-tion had delayed immune reconstitution compared to those who did not; however, the Ig repertoire steadily recovered with no statistically significant difference by 18 months after treatment cessation. Our study has several limitations, including a small sample size, incomplete assessment of total Ig levels and immune subsets, and relatively short follow-up of patients on MRD-SURE. We are unable to compare immune reconstitution by NGS to patients on lenalidomide maintenance as patients who did not enter MRD-SURE did not have ongoing MRD surveillance testing. Furthermore, we are