Seven-year median time to progression with thalidomide for smoldering myeloma: partial response identifies subset requiring earlier salvage therapy for symptomatic disease

Multiple myeloma (MM) is thought to progress, in the majority of cases, from a clinically benign precursor condition, monoclonal gammopathy of undetermined significance (MGUS).¹  A more advanced form of MGUS is referred to as smoldering MM (SMM),^2,3  which is characterized by a greater tumor burden but, like MGUS, by the absence in general of cytogenetic abnormalities (CAs) and magnetic resonance imaging (MRI)– or metastatic bone survey (MBS)–defined focal lesions.⁴  The clinical course of patients afflicted with SMM varies considerably, with a median time to progression to symptomatic disease reportedly of 1 to 2 years.⁵  According to a recent comprehensive review of the Mayo Clinic (Rochester, MN) records covering a 26-year interval from 1970 to 1995, 276 (8%) among 3549 patients with MM fulfilled the criteria of SMM⁶ ; 59% of the 276 patients developed either symptomatic MM or primary amyloidosis. Based on the proportion of bone marrow plasma cells and serum monoclonal protein levels, 3 groups with vastly differing 10-year SMM progression estimates were identified (77% vs 64% vs 33%). The recent availability of the serum-free light chain assay also aided in the prognostication of subjects with SMM, in that the free light chain ratio helped distinguish 3 risk groups with 5-year progression rates of 76%, 51%, and 25%.⁷ 

An angiogenic switch has been postulated as a pivotal event in the progression from MGUS to MM.⁸  In fact, our first trial with thalidomide (THAL) for advanced and refractory MM⁹  was based mainly on the hypothesis that THAL's antiangiogenic properties¹⁰  could be effectively exploited as a means of overcoming tumor cell resistance to traditional cytotoxic agents.¹¹  As in the Mayo Clinic phase 2 trial reported by Rajkumar et al,¹²  both antiangiogenic and immunomodulatory properties of THAL¹³  provided a strong rationale for our group in 1998 to evaluate this noncytotoxic compound in a preemptive clinical trial aimed at delaying or even preventing the progression from SMM to overt disease. We added pamidronate (PAM), which, in addition to reducing the frequency of skeletal events in overt MM,^14,15  could shrink established tumors in a severe combined immunodeficient mouse–human (SCID-hu) model system¹⁶  and induce clinical responses in patients with SMM.¹⁷  Here we report on the response rates and clinical outcomes of 76 eligible patients with SMM enrolled into protocol UARK 98-036 at the Myeloma Institute for Research and Therapy of the University of Arkansas for Medical Sciences.

The protocol was designed to enroll previously untreated patients with SMM. Diagnostic criteria included bone marrow plasmacytosis (BMPC) more than 10% and measurable disease defined as serum M-protein levels of 2 g/dL or more or urine M-protein excretion of 400 mg/d or more; osteolytic bone lesions had to be absent.³  A median time of 6 months (range, 0 to 80 months) had elapsed from initial diagnosis of SMM to protocol therapy. The treatment plan called for daily administration of THAL at a dose of 200 mg, allowing for dose reductions to as low as 50 mg applied on alternating days in the event of severe toxicities. PAM was administered monthly at a dose of 90 mg intravenously.

Response criteria of the European Bone Marrow Transplant group and those set forth by a recent international panel were used.^18,19  Improvement (IMP) required M-protein reduction by at least 25% and partial response (PR) by 50%; near-complete remission (n-CR) required absence of M-band on standard electrophoresis and complete remission (CR), absence of M-protein band on immunofixation analysis.

Kaplan-Meier methods were used to generate survival distribution graphs,²⁰  and comparisons were made via the log-rank test. Running-log-rank tests were used to determine optimal cutoff values for baseline laboratory values in the multivariate analyses, which applied stepwise selection and Cox proportional hazard regression modeling.²¹  The institutional review board of the University of Arkansas for Medical Sciences approved the studies. Informed consent was obtained in accordance with the Declaration of Helsinki.

The 76 eligible subjects collectively had features of low-stage disease by International Staging System (ISS; stage I; 90%), beta-2-microglobulin (B2M < 3.5 mg/L; 92%), and BMPC (< 20%; 62%). None had hypercalcemia more than 2.625 mM (10.5 mg/dL) or bone lesions, 1 patient had renal insufficiency (creatinine 176.8 μM [> 2 mg/dL]), and 6 others had anemia (hemoglobin < 105 g/L [10.5 g/dL] or 20 g/L [2 g/dL] less than normal); only 5% had CAs. Applying prognostic groups defined by Kyle et al,⁶  18% qualified for group 1 (BMPC ≥ 10%, serum M-protein ≥ 3 g/dL) and 82%, for group 2 designation (BMPC ≥ 10%, M-protein < 3 g/dL); none was in group 3 (BMPC < 10%, M-protein ≥ 3 g/dL). With a median follow-up of 6 years, 34 have experienced disease progression, 17 of whom required salvage therapy after a median of 25 months (Figure 1A). Disease progression and need for salvage therapy were not affected by Kyle et al's prognostic groups. The median postsalvage therapy survival is 66 months. Seventy-six percent of patients remain alive; 8 patients died after salvage therapy for progressive MM and 10 died of unrelated causes.

Two years after entry on study, THAL had been discontinued in 50% and its dosing reduced in another 36%, due in part to peripheral neuropathy (grade 3, 8%; grade 2, 18%; grade 1, 12%). Other grade 2 or higher toxicities included neutropenia and dizziness in 8%; fatigue, thrombocytopenia and cardiovascular events were each reported in 7%. PAM was discontinued in 54% within 2 years, mainly on account of patients' and physicians' choices.

At 4 years from study entry, IMP, PR, n-CR, and CR were achieved in 63%, 25%, 12%, and 5% of patients, respectively (Figure 1B), with median times to response of 1 to 2 years as opposed to 1 to 2 months in the therapeutic study of THAL for advanced and refractory MM.⁹  We next examined the impact of the level of response to THAL within 9 months of starting therapy, when 64 subjects had not yet suffered an event, on overall survival (OS), event-free survival (EFS), and time to MM therapy (TTMT). Although OS was similar in the 3 response categories (70% estimate at 6 years), EFS was significantly shorter in patients who had achieved PR or better on THAL (Figure 1C), as was TTMT (Figure 1D). B2M elevation was significantly associated with shorter OS and EFS, and development of CAs imparted significantly shorter EFS (Table 1). Serum-free light chain levels were available in only a minority of subjects, so this important variable could not be evaluated for its prognostic implications.⁷  Achieving PR or better status on THAL therapy was the only parameter significantly and independently associated with short TTMT. The unexpected shorter TTMT in patients achieving PR or better status with THAL may be explained by a high sensitivity to THAL of tumors with greater proliferative potential that ultimately, however, progress sooner due to rapidly developing resistance or tumor escape after drug discontinuation. Thus, by inducing remissions in SMM, THAL therapy identified tumors with greater propensity for progression to MM.

The median postsalvage survival of more than 5 years following preemptive THAL therapy for SMM is in line with treatment results achieved in primary symptomatic MM, alleviating concerns about THAL-induced resistance. Lenalidomide, a more potent and better tolerated THAL congener,^22,23  would appear a more suitable agent to be explored for longer-term preemptive intervention in SMM. In the absence of randomized clinical trial–derived data, patients with SMM should be observed without therapy until progression or symptom development. The Southwest Oncology group is preparing a protocol that randomly assigns subjects with SMM to lenalidomide or placebo, once a smoldering course has been documented during a 3-month lead observation time. Clinical end points will be time to symptomatic disease requiring MM therapy and survival. Correlative science studies will include gene expression profiling to determine molecular subgroups²⁴  and the presence of a MGUS-like signature.²⁵ 

Celgene (Summit, NJ) provided the study drug free of cost and contributed data management support.

This work was supported by the National Cancer Institute (CA55819; Bethesda, MD).

National Institutes of Health

Contribution: B.B. and J.B.Z. designed the trial; B.B. wrote the paper; J.D.S., J.E., and S.Y. provided important insights through discussion of data; B.B., F.R., M.P.-R., K.H., Y.A., and E.A. enrolled patients and participated in clinical research; N.P., N.S.K., G.S., and B.J. collected data; and A.H., J.S., and J.C. performed statistical analyses.

Conflict-of-interest disclosure: J.B.Z. is an employee of Celgene. All other authors declare no competing financial interests.

Correspondence: Bart Barlogie, Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, 4301 West Markham, No. 816, Little Rock, AR 72205; e-mail: barlogiebart@uams.edu.