Is early, deep free light chain response really an adverse prognostic factor?

To the editor:

In a cohort of 303 patients, van Rhee et al made the important observations that baseline free light chain (FLC) and early FLC reduction after 1 to 3 cycles of highly effective chemotherapy are prognostic for both overall and event-free survival.¹  First they demonstrated that patients with the highest tercile of FLC levels at baseline (in their case >75 mg/dL) had the worst overall survival, independent of high LDH and of abnormal cytogenetics. Their second observation was that patients with the deepest FLC response after 1 to 3 cycles of VDT-PACE [bortezomib, dexamethasone, thalidomide, cisplatin, doxorubicin, cyclophosphamide, and etoposide] had the worst outcomes. Immediately preceding cycle 2 of therapy, the FLC reduction terciles were less than 58%, 58% to less than 86%, and 86% to 100%, with respective 24-month estimated survival rates of 90%, 91%, and 81%. After approximately 2 to 3 cycles, the FLC reduction terciles were less than 75%, 75 to less than 96%, and 96 to 100% with respective 24-month estimated survival rates of 91%, 93%, and 79%. The negative impact of extreme drops in involved FLC on event free survival and overall survival was independent of high LDH and abnormal cytogenetics, but the authors provide no information about whether it was independent of baseline FLC. One-third of patients had baseline FLC of less than 10.7 mg/dL (lowest tercile). Given the fact that the upper limits of normal for κ and λ FLC are 1.93 and 2.64 mg/dL, respectively, a high fraction of patients in this low baseline FLC tercile group were not eligible to have a FLC reduction of 86% or 96% or more (highest tercile of FLC reduction). Therefore, this good prognostic group was destined to be a “low responder.” Depending on the distribution of patients in the middle tercile of baseline FLC, a high proportion of these patients also may not have been eligible to have FLC reduction of 86 or 96%. Van Rhee's observation would have more meaning if patients with “inevaluable” FLC (eg, <10 mg/dL, or difference between involved and uninvolved less than 5 mg/dL) had been excluded from the analyses. In short, as analyzed in the Arkansas cohort, the depth of early FLC reduction may merely be a surrogate for high baseline FLC.