Complete response in myeloma: a Trojan horse?

Comment on Offidani et al, page 2159, and comment on Mateos et al, page 2165

Complete response has been a primary end point in many cancer therapy trials—including the two reported in this issue for elderly patients with multiple myeloma—but does not per se translate into survival benefit.

The first major advance in myeloma therapy involved recognition of a steep dose-response effect for melphalan (safely exploited with autologous transplantation support) that improved survival compared with standard-dose regimens in randomized trials.¹  Another wave of therapeutic innovation was ushered in with the recognition that immunomodulatory agents (eg, thalidomide) and proteasome inhibitors (eg, bortezomib) have marked antimyeloma activity.^2,3  When applied up front, the addition of thalidomide to standard melphalan-prednisone (MP) induced high response rates and promising event-free and overall survival rates.^4,5 

In this issue of Blood, we are learning about 2 highly effective combination regimens—bortezomib plus MP (VMP) and thalidomide-dexamethasone-liposomal doxorubicin (ThaDD)—for untreated patients older than 65 years. Complete response (CR) rates in excess of 30% were noted and occurred within a few months of treatment initiation. In a comparison with historical controls receiving MP, Mateos and colleagues show superior 16-month event-free survival (83% versus 51%) and overall survival rates (90% versus 62%) with VMP. Similarly, Offidani and colleagues report that ThaDD yielded 36-month event-free and overall survival rates of 57% and 74%, respectively. Toxicities were manageable in both trials. These promising early results in an elderly patient population undoubtedly will stimulate similar trials in younger patients, for whom autotransplantation-supported high-dose melphalan (MEL) is still the treatment of choice. Whereas 10-year survival rates of 30% have been reported for MEL,⁶  the follow-up data for any of the newer drug combination regimens is still very limited.

Importantly, high CR rates do not always translate into improved survival and are therefore not reliable primary end points for clinical trials. Thus, when myeloma evolves from monoclonal gammopathy of undetermined significance (MGUS), a lower CR rate has no adverse consequences for survival.⁷  Despite similar CR rates of 40% after MEL treatment, the one third of patients with cytogenetic abnormalities (CAs) had a short median survival of only 2 to 3 years, compared with 7 or more years in the remainder. ⁸  Higher CR frequency and longer event-free survival on the thalidomide arm of Total Therapy 2 did not result in an overall survival gain.⁹  Thus, although high CR rates may translate into extended survival for MEL treatment of patients with good-risk myeloma, this may not hold true following treatment with the newer nongenotoxic agents.

Before being indiscriminately considered as primary therapy for all patients with myeloma, new combination regimens should demonstrate survival benefits in patients with notoriously high-risk disease. This will prevent compromising 10-year survival expectations approaching 50% in good-risk disease with high-dose MEL. Metaphase cytogenetics and molecular genetic studies should be routinely performed to guide patient selection into clinical trials for high-risk patients. Current CR definitions in myeloma rely mainly on the detection of monoclonal protein and random bone marrow examination. However, focal lesions recognized on magnetic resonance imaging (MRI) and harboring viable monoclonal plasma cells can persist in clinical CR; their resolution with further therapy (MRI-CR) lags 2 years behind the onset of clinical CR and, when achieved, has favorable survival implications.¹⁰  The presence of such myeloma cells in focal lesions has only recently been appreciated as a potential source of disease recurrence. ▪