Serum Immunoglobulin Free Light Chain (sFLC) Is a Sensitive Marker of Response in Waldenstrom Macroglobulinemia (WM).

Background. WM is characterized by excess secretion of IgM in the serum. Response to therapy is assessed according to the Consensus panel recommendations with the use of IgM by serum protein electrophoresis (M spike). However, there are many limitations to the use of M-spike level, and new markers are needed. We, and others, have recently demonstrated that the measurement of involved sFLC values accurately identified patients with poor prognostic features of WM. We sought to determine whether involved sFLC level can be used as a reliable and sensitive marker to study response to therapy in WM.

Methods. We prospectively studied M-spike and involved sFLC levels in 32 patients with relapse/refractory WM, treated on two phase II clinical trials, single agent perifosine (N=21; given 150mg oral daily for 28 days), or bortezomib-rituximab (N=13; given IV bortezomib 1.6mg/m² at days 1, 8, 15 q28 days x 6 cycles and rituximab 375mg/m² at days 1, 8, 15, 22 on cycles 1 and 4). Levels of sFLC and M spike were measured on day 1 of each cycle. In addition, sFLCs were measured weekly for the first month in the bortezomib-rituximab study (N=12). Responses were assessed after 2 cycles of therapy. Time to response was determined from initiation of therapy to the time of initial response.

Results. Characteristics in the overall population were not different between the 2 groups treated, and median values were: beta-2 microglobulin (B2M) 2.7mg/L, hemoglobin level 11g/dL, serum IgM 40g/L, M-spike 2.4g/dL, and involved sFLC 111mg/L. The median follow-up was 5 months, and was not significantly different between the 2 trials. The overall response rate (ORR) of the bortezomib/rituximab clinical trial was 85% and the perifosine clinical trial was 33%. The ORR as calculated by the sFLC in the bortezomib-rituximab trial was 100% and in the perifosine trial was 39%. Eleven (34%) patients achieved better responses by sFLC compared to M spike. Of these, 7 were classified as minor response (MR) by M spike but achieved a partial response by sFLC, and 4 were classified as stable disease by M-spike, but achieved MR using involved sFLC. Only 2 patients (6%) had a better response using M spike compared to sFLC. The overall median time to response was 79 and 110 days using M-spike, and 68 and 41 days using sFLC, in bortezomib-rituximab and perifosine trials, respectively. We next found that neither M-spike nor involved sFLC measured prior to therapy could predict response. We then investigated whether sFLC could predict response to therapy using weekly sFLC for the first month. Among the 14 patients studied, 7 showed a decrease in sFLC within the first month, which predicted response by using M spike. On the other hand, the other 7 patients had an sFLC “flare” within the first 4 weeks of therapy, and these patients had a delayed response to therapy using M spike criteria.

Conclusion. Involved sFLC is a sensitive marker for monitoring response to therapy in patients with WM. Using sFLC measurement, we showed a higher response rate in patients treated on 2 prospective clinical trials than using M-spike measurement. More importantly, we demonstrated that responses occurred earlier using sFLC compared to M-spike measurement. sFLC measurement could predict response to therapy within the first month of therapy.