Poster Session I: Myeloma and other monoclonal gammopathies - Clinical
Rapid remission has long been recognized as an important predictor for long-time survival in acute leukemia (AL). However, the effects of response kinetics on survival seems to be different from those in AL. A rapid reduction of tumor burden probably indicates a chemotherapy-sensitive population. But on the other hand, a high proliferative activity of plasma cell may be responsible for the rapid response. The effects of response kinetics on outcome for MM remain largely unexplored.
In the era of novel drugs, the proportion of patients attaining complete remission (CR) has been significant increased. However, a deep response has not always translated into an improved survival. In both transplant-eligible and ineligible patients, we identified different response kinetic patterns after induction therapy which are associated with distinct patient outcomes.
The relationship between response kinetics and outcome were prospectively assessed in 626 patients with newly diagnosed MM that were included in a prospective, non-randomized clinical trial (BDH 2008/02). Patients were assigned to either immunomodulatory drugs-based or proteasome inhibitors-based therapy. The response depth, time to best response (TBR) and duration of best response (DBR) were collected. Modified progression-free survival (mPFS) and overall survival (mOS) from time of first detectable best response were calculated.
Depth of response was associated with superior outcomes, consistent with findings from other studies. However, the early responders (defined as TBR ≤ 3 months) showed significantly worse survival compared with late responders. There was no significant difference in survival between patients with a rapid CR and those with a late partial remission (PR) (mPFS 33.4 vs. 36.2 months, P = 0.382; mOS 62.6 vs. 64.2 months, P = 0.169). TBR remained an independent predictor of outcome in multivariate analysis (HR = 1.9 and HR = 2.8 for mPFS and mOS).
Moreover, four distinct response kinetic patterns were identified: 1) late response and late relapse (“U-valley” pattern) (n = 157, 25%); 2) early response and late relapse (n = 101, 16%); 3) late response and early relapse (n = 130, 21%); 4) early response and early relapse (“Roller coaster” pattern) (n = 172, 25%), and best response less than PR (n = 66, 11%). Response kinetics based stratification resulted in remarkable differences in outcome. Patients with gradual and sustained remission identified as “U-valley” pattern experienced superior outcomes, followed by patients with “early response and late relapse” and “late response and early relapse”, while poorest outcomes were observed in rapid and transient responders (“Roller coaster” pattern) (median OS 126, 81, 44 and 30 months, respectively). The effects of response patterns on survival were confirmed in patients at different stages of disease and cytogenetic risk, including transplant-eligible patients and those attaining different extents of response depth.
Our data indicate that a slow and gradual response is a good prognostic factor for survival, arguing against premature change to more intensive regimens, especially for elderly or unfit patients. In addition to response depth, the kinetic pattern of response is a simple and powerful predictor for survival even in the era of novel agents.