Neutropenia and G-CSF Use In Patients with Relapsed/Refractory Multiple Myeloma Treated with Lenalidomide Plus Dexamethasone

Lenalidomide is approved in relapsed/refractory (rel/ref) MM based on 2 large phase III trials (Dimopoulos NEJM 2007; Weber NEJM 2007). In these trials, grade 3–4 neutropenia was common (29-41%) and was a main cause for lenalidomide (len) dose reductions despite mandated use of daily GCSF as first-step neutropenia management. Optimal management for len-induced neutropenia is unclear. At our institution, 216 patients (pts) with rel/ref MM were treated on the MM016, an Expanded Access Program (EAP), using len and dexamethasone (dex) at the same dose/schedule as that of the randomized trials. Due to limited access, GCSF was not mandated for grade 3–4 neutropenia, though we developed an intermittent schedule of GCSF (300ug SC 2–3 doses/week for weeks 3 and 4 of each 28 day cycle), typically continued into subsequent cycles to prevent recurrence. We aimed to avoid len dose reduction to maximize efficacy. In this retrospective analysis, our neutropenia management approach is evaluated, focusing upon recurrent neutropenia, infections, len dose-intensity, response, survival outcomes.

From 2005–2008, 216 rel/ref MM pts were treated on the EAP at our center. Similar to the phase III studies, len was initiated at 25mg OD × 21days with dex 40mg days 1–4, 9–12, 17–20 every 28 day cycle. Recurrent neutropenia and infectious complications were reviewed. Comparison of baseline variables and neutropenia/survival was performed between pts receiving GCSF (group 1) and those not (group 2). OS and PFS were estimated by Kaplan-Meier curves and log rank test.

Neutropenia and G-CSF use: Of 216 pts treated with len, 117 pts (54.2%) received GCSF for grade 3–4 neutropenia (group 1); 99 pts (45.8%) did not (group 2). For group 1, the first episode of grade 3–4 neutropenia occurred early (median cycle 2, range 1–20) with GCSF started at median cycle 3 (range 1–19). Although most pts continued GCSF prophylactically into subsequent cycles, almost half (58 pts; 49.6%) recurred with ≥1 episode of grade 3–4 neutropenia, 18 pts (15.4%) ≥3 episodes. Despite GCSF use, pts in group 1 had more grade 3–4 infections (47.0% vs. 25.5%, p=0.002) and hospitalization due to infection (40.2% vs. 25.3%, p=0.021). In addition, len dose reductions were frequent in group 1 (40.2% vs 16.2%; p<0.001). As a surrogate for dose-intensity, the proportion of cycles at full-dose without delay was lower in group 1 (0.67 vs. 0.96, p<0.001). However, group 1 pts received more lines of prior therapy than group 2 (median 3 vs. 2, p=0.021), had lower baseline platelets (<100/uL in 41.0% vs. 23.2%, p=0.006) and neutrophils (<2.0 bil/L in 41.0% vs. 20.2%, p<0.001). All other baseline characteristics were similar between groups.

Pts receiving GCSF remained on len for longer (median duration 10.3 vs 3.7 mos; p=0.01), with primary causes for len discontinuation in both groups due to disease progression (65.2%) and toxicity (15.2%). Increased responses (66.7% vs. 45.5%, p=0.002) and improved quality of response (CR/VGPR 25.6% vs. 14.1%, p=0.03) were seen in the GCSF group. G-CSF use was associated with significantly longer PFS (9.1 vs. 4.0 mos, p=0.048), though the OS between groups was not statistically different (20.9 vs. 13.7 mos; p=0.28). On multivariate analysis, G-CSF support was associated with decreased risk of death (HR 0.46, 95% CI 0.31–0.67, p<0.0001).

1. Not unexpectedly, pts requiring GCSF were predisposed to len-induced neutropenia due to impaired marrow reserve from heavy pretreatment. Contrary to our predictions, our intermittent dosing schedule of GCSF did not significantly reduce rates of subsequent neutropenia, severe infections, or lenalidomide dose reduction. More intensive GCSF schedules than that used at our institution may be required.

2. However, GCSF use is associated with longer duration on len therapy, likely leading to observed improvements in responses, quality of response, and PFS. This suggests that those pts with compromised marrow reserve benefit from dose reductions and striving for full dose-intensity is not necessary.

3. Alternatively, given that GCSF is an independent predictor for prolonged PFS, one may hypothesize that GCSF has effects unrelated to its putative role in reducing neutropenia and may perhaps be related to its anti-inflammatory and immunomodulatory effects. Further studies evaluating the mechanisms and interactions of GCSF with len may provide insights.

Reece:Celgene: Honoraria, Research Funding. Trudel:Celgene: Honoraria. Kukreti:Celgene: Honoraria. Chen:Celgene Corporation: Consultancy, Honoraria, Research Funding.