Influence of Renal Impairment and Genetic Risk Factors on Response to Induction Therapy in the HD4 and MM5 Trials of the GMMG

Background:

In the HOVON65/GMMG HD4 trial in patients with newly diagnosed multiple myeloma we have previously shown that patients with renal impairment (RI) (creatinine > 2 mg/dl) have higher response rates and better survival when receiving bortezomib both in the induction and maintenance therapy before and after high-dose chemotherapy (HDT) (Scheid et al. Haematologica 2014). In addition patients with RI showed a higher prevalence of genetic high-risk features such as del17p or t(4;14). The aim of this analysis was to further elucidate the interaction between renal and genetic risk factors in well defined homogeneously treated myeloma patients.

Methods:

For this study we selected 2 cohorts of patients entered into 2 consecutive prospective trials with centralised FISH-assessement on CD138-selected bone marrow cells. The first cohort (1) comprises 395 patients from the HOVON65/GMMG HD4 trial having been treated in the German centers and the second cohort (2) consisted in the 601 patients (intention-to-treat population) from the recently closed GMMG MM5 trial. Patients lacking FISH results were excluded from the analysis, which was the case for 53 (13.4%) patients in cohort 1 and 43 (7.2%) patients in cohort 2. In cohort 1 induction treatment was vincristine or bortezomib + doxorubicin and dexamethason followed by tandem HDT followed by bortezomib or thalidomide maintenance. Cohort 2 received doxorubicin or cyclophosphamide + bortezomib and dexamethason as induction followed by 1- 2 HDT and consolidation and maintenance with lenalidomide.

Results: In cohort 1 38 (10%) had RI and del 17p was found in 12/33 (36.4%) evaluable patients compared to 24/302 (7.9%) patients without RI (p<0.001). t(4;14) was present in 11/33 (33.3%) patients with RI and 38/304 (12.5%) without RI (p=0.005). Gain of 1q21 (> 2 copies) was present in 14/33 (42.4%) patients with RI and 92/298 (30.9%) without RI (n.s.). In cohort 2 68/601 (11.3%) had RI and 7/63 (11.1%) had del17p compared to 56/495 (11.3%) patients without RI (n.s.) while 29/63 (46%) patients with RI had t(4;14) versus 265/495 (53.5%) without RI (n.s.). Gain1q21 (>2 copies) was found in 36/63 (57.1%) with RI versus 209/495 (42.2%) without RI (p=0.025). In cohort 1 the response rate with at least VGPR after induction was low with and without RI in the VAD arm (15 vs 7.1 %) and reduced in patients with RI in the PAD arm compared to those without RI (22.2 vs 37.9%). This trend was not found in cohort 2: Patients with RI had VGPR or better in 42.4% in the PAD and 52.9% in the VCD arm, compared to 33.5% and 33.1% without RI respectively. Del17p and t(4;14) which were more frequent in patients with RI in cohort 1 did not have a negative impact on response rates after induction. Similarly gain1q was more frequent among patients in cohort 2 with RI but did not impact on response to induction.

Conclusions:

We analysed the effect of RI and genetic risk factors on the response to induction therapy in two different patient cohorts from two consecutive prospective trials. High-risk genetic features where found more frequently in patients with RI, but the pattern was entirely different between cohort 1 and 2 and they did not seem to influence response rates after induction. Our results confirm that bortezomib-based induction regimens achieve high response rates in myeloma patients with RI similar to those in patients without RI, independent of the presence of genetic risk factors.