AQUILA: Dousing the Embers of Smoldering Multiple Myeloma With Daratumumab Monotherapy

Smoldering multiple myeloma (SMM), an asymptomatic plasma cell neoplasm that precedes the development of symptomatic multiple myeloma (MM), is characterized by variable progression to myeloma-defining events.¹  Historically, the standard of care for management of SMM has been observation, despite evidence of benefit in trials using lenalidomide monotherapy (ECOG E3A06) or a combination of lenalidomide and dexamethasone (QuiRedex).^2,3  The reasons for a lack of widespread adoption of such treatments include suboptimal (single-time point vs. evolving) SMM risk stratification, resulting in not all patients benefitting equally; the need for fixed duration of very well-tolerated therapy given that these are asymptomatic individuals; and questions of cost-effectiveness. From a design perspective, a randomized study should ideally afford patients assigned to the observation arm with access to the best possible agents at the time of progression (i.e., early vs. late treatment). In addition to delaying progression to hypercalcemia, renal insufficiency, anemia, and bone disease (CRAB) symptoms (vs. biomarkers only) and improving progression-free survival (PFS), an overall survival (OS) benefit is also ideal, as active anti-myeloma agents are inherently likely to delay progression.

While treatment of patients with SMM is controversial, most would agree that a high-risk subset of these patients is likely to progress within the first two to five years of diagnosis and may benefit from early intervention to either delay or prevent end-organ damage from MM.⁴  In recent years, daratumumab (DARA), a human immunoglobulin (Ig) G kappa anti-CD38 monoclonal antibody with a favorable tolerability profile, has been approved in the relapsed and frontline settings for the treatment of MM.^5,6  The phase II CENTAURUS study, which investigated the use of single-agent DARA with three different dosing schedules in patients with intermediate- and high-risk SMM, showed favorable responses without any new safety signals, setting the stage for the phase III AQUILA trial.⁷ 

In AQUILA, patients diagnosed with high-risk SMM within the past five years were randomized to either subcutaneous DARA monotherapy (28-day cycles) or active monitoring for 39 cycles, 36 months, or disease progression (either biochemical or CRAB symptoms plus MM-defining biomarkers). Patients were considered to have high-risk disease based on bone marrow plasma cell (BMPC) involvement of at least 10% and one high-risk factor for progression (M protein >3.0g/dL, involved to uninvolved free light chain ratio of >8, immunoparesis of uninvolved Ig isotypes, IgA subtype, or BMPC of 50% to <60%).⁸  These criteria were predominantly used to identify patients at high risk of progression at the time of trial development, which preceded publication of the 2018 Mayo Clinic criteria (BMPC ≥20%, M protein ≥2.0 g/dL, and involved to uninvolved free light chain ratio of ≥20).⁴  Unlike prior SMM studies, the AQUILA trial appropriately required advanced imaging with low-dose, whole-body computed tomography (CT), positron emission tomography-CT, or whole-body magnetic resonance imaging to rule out osseous lesions.

A total of 390 patients were randomized in a 1:1 fashion to DARA (n=194) or active monitoring (n=196), with a median time from diagnosis of 0.72 years (range, 0 to 5.0). After a median follow-up of 65.2 months, the study met its primary endpoint, with DARA monotherapy achieving superior PFS (hazard ratio (HR) = 0.49, 95% CI 0.36-0.67; p<0.0001). Five-year PFS in the DARA and active monitoring groups was 63.1% and 40.8%, respectively, with a median time to disease progression of 44.1 months versus 17.8 months (HR=0.51, 95% CI 0.40 to 0.66). Overall response rate (ORR) was 63.4% for the DARA group (the highest figure reported with DARA monotherapy in any study) versus 2.0% for the active monitoring group. Fewer patients in the DARA arm ultimately required initiation of first-line therapy for multiple myeloma by data cutoff (33.2% vs. 53.6%, HR=0.46, 95% CI 0.33-0.62). The clinical benefit appeared to be consistent across prespecified subgroups, despite insufficient power to demonstrate this in low numbers. Most importantly, there was an improvement in OS at five years of 93.0% versus 86.9% in favor of the DARA arm, but the drivers of this OS benefit are complex and require a deep dive into the details of those patients who progressed or died during the study. Health-related quality of life (QoL), as measured by both the European Organisation for Research and Treatment of Cancer’s multiple myeloma QoL questionnaire and EuroQol’s five-dimension questionnaire, appeared to be similar between arms and consistent over the course of the study.

From a safety standpoint, there were significantly higher rates of grade 3 or 4 infections in the DARA arm (16.1% vs. 4.6%), which is important to consider when intervening on an asymptomatic population. The most common reason for treatment discontinuation was progressive disease in both groups (21.8% in the DARA arm and 41.8% in the active monitoring arm). In patients who discontinued the trial, 7.7% and 13.3% of these were due to death in the DARA and active monitoring arms, respectively. There were more deaths due to progressive disease (PD) in the active monitoring group, and these deaths occurred sooner after treatment initiation than in the DARA group, suggesting a delay in end-organ damage and myeloma-related complications with early intervention. Of the deaths listed due to PD (three in the DARA arm vs. nine in the active monitoring arm), the median time of death relative to randomization was 15.7 months versus 10.2 months, respectively. At the time of progression, few of those receiving first-line chemotherapy in the study received a regimen containing an anti-CD38 monoclonal antibody (17.9% in the active monitoring arm vs. 8.3% in the DARA arm). While the majority of patients in both groups received triplet or quadruplet regimens, 10 patients across arms received suboptimal therapy (i.e., doublets or monotherapy). When the Mayo 2018 criteria were applied to the intention-to-treat (ITT) population, it appeared that the high-risk subgroup benefitted the most with regard to PFS (HR=0.36, 95% CI 0.23-0.58), raising the question of whether the decrease in progression to CRAB symptoms and OS benefits might also be skewed to high-risk patients. Equally important to determine would be whether the benefits of DARA actually applied to patients who were in the smoldering phase for more than two years, as the lack of evolution would qualify them as having non-high-risk disease. It would also be helpful to know what frontline therapy was administered to those patients in both arms who died of PD to assure that the OS benefit seen in the study was attributable to DARA rather than suboptimal induction therapy at the time of progression.

When taking a more granular look at the subgroup analyses and progression data, we have identified some important points to consider in applying these findings to real-world patients. The discontinuation rates in the intervention arm due to adverse effects were lower in AQUILA compared to prior trials investigating lenalidomide in SMM, once again highlighting the high degree of tolerability for DARA monotherapy. With regard to PFS, it is important to assess whether improvements are due to delay in end-organ damage or progression by MM-defining biomarkers, which is difficult to ascertain as the events reported by Meletios A. Dimopoulos, MD, and colleagues are based on the denominator being the PD population rather than the ITT population. Based on the ITT population, the rates of CRAB progression in DARA compared with observation were 12 (6.2%) versus 34 (17.3%) for any CRAB symptom (hypercalcemia, 0% vs 1%; renal insufficiency, 0% vs. 0%; anemia, 1% vs. 7.1%; and bone disease, 5.2% vs. 9.2%).⁹  Of note, all CRAB criteria are not created equally, and we must prioritize reducing rates of bone disease and renal insufficiency, while progression to anemia, which was most pronounced in this study, is much less morbid and can be safely monitored. Taking all of this into consideration, we must be mindful that early intervention with immune-based therapies can be quite costly, so quality-adjusted life-years would be an important metric for evaluating the cost-effectiveness and overall benefit of early therapeutic intervention with DARA compared to a watch-and-wait approach.