Purpose: Four Southwest Oncology Group (SWOG) standard-dose chemotherapy protocols for multiple myeloma (MM) initiated between 1982 and 1992 were evaluated. The purpose was to clarify the predictive value of specific levels of myeloma-associated monoclonal protein reduction and time to first progression using mature data sets.
Patients and methods: Study data on 1,555 eligible previously untreated patients with MM enrolled onto SWOG phase III trials 8229, 8624, 9028, and 9210 were used in these analyses. Six-month and 12-month landmark analyses were performed to evaluate the outcome for patients in each response category.
Results: The overall and event-free survivals for the four protocols combined were 33 months and 18 months, respectively. Using 6- and 12-month landmarks, the median survivals of 30 to 35 months were not different for responders (> or = 50% and > or = 75% regression) versus nonresponders in patients without disease progression before the landmarks. Conversely, at the 6- and 12-month landmarks, the median survivals for patients who had experienced disease progression were 13 and 15 months, respectively, versus a 34-month median for patients who did not experience progression. Using the Cox survival model, with response and progression considered as time-dependent covariates, survival duration was influenced more by the occurrence of progression than by the occurrence of response.
Conclusion: The magnitude of response, as a single variable, does not predict survival duration. Patients with response and stable disease have equivalent outcome. Only patients with progressive disease have a poorer outcome. The best indicator of survival is time to first progression.