Response Adapted Induction Treatment Improves Outcomes for Myeloma Patients; Results of the Phase III Myeloma XI Study

Background

The Myeloma XI study is the first randomized study to investigate a response-adapted approach to induction therapy for newly diagnosed myeloma (NDMM). The study addresses whether, for patients achieving less than optimum response to an initial immunomodulatory (IMiD) triplet combination, defined as at least VGPR, the use of a sequential proteasome inhibitor (PI) based triplet can improve outcomes. In total 581 patients were randomized into the study which confirms the clinically significant benefit of deepening responses by utilizing treatment with a different mode of action, and that this leads to better outcomes.

Methods

This phase III, UK-based, multicenter, open-label, parallel group, randomized controlled trial for NDMM patients of all ages, randomized patients initially between a thalidomide or lenalidomide triplet combination with cyclophosphamide and dexamethasone. This IMiD triplet was continued for a minimum of 4 cycles (transplant eligible, TE) or 6 cycles (transplant non-eligible, TNE) and to maximum response. At the end of this IMiD regimen response was assessed. Patients with a suboptimal response (MR/PR) were randomized between further induction therapy with bortezomib, cyclophosphamide and dexamethasone (CVD) or no further induction therapy. Patients with a good response (VGPR/CR) proceeded straight to ASCT (if TE), whilst refractory (SD/PD) patients all received the CVD regimen. For patients receiving CVD, treatment was planned to continue to maximum response, and eligible patients would proceed to ASCT.

The primary endpoints of the adapted approach randomization were progression-free survival (PFS) and overall survival. Secondary endpoints included upgrading of response compared to baseline and the impact of the PI combination in a high-risk subgroup. This abstract contains a preliminary analysis, final data will be presented at the meeting. The median follow up in this analysis is 29 months [IQR 17-44].

Results

581 patients (366 TE, 215 TNE) with initial response to IMiD of MR/PR were entered into the CVD randomization. In total 292/581 patients were randomized to receive sequential treatment with CVD. The arms were well-balanced with respect to clinical features and response (e.g. ISS stage: III 21% vs 19%, PR: 88% vs 88%).

This randomization has met its primary endpoint. Overall the sequential use of CVD significantly improved PFS from a median of 24 to 30 months (HR 0.67 [95%CI 0.53-0.85], p=0.0005). This was largely due to a significant improvement seen in the TE pathway, HR 0.56 [95%CI 0.40-0.77], median PFS no therapy 31 months vs CVD 55 months, p=0.0003. In the TNE pathway there was an early benefit with improved median PFS 14 months vs 20 months, but similar hazard after 2 years (HR 0.83 [95%CI 0.60-1.17], p=0.297).

Importantly upgrading of response was seen with 118/289 (41% [95%CI 35-47]) of evaluable patients who received CVD moving from MR/PR to VGPR/CR. 115/289 remained in the same response category MR/PR, but these patients still had a mean reduction in paraprotein during CVD of 24% [95%CI 11-17]. The upgrade in response was seen in both pathways and was not affected by the IMiD received in the initial induction randomization. The impact of CVD in cases with molecularly defined high-risk disease compared to standard and multiparameter flow cytometry assessment of minimal residual disease status will be presented at the meeting.

In the transplant eligible pathway an improved depth of response persisted in the 253 patients completing ASCT with VGPR/CR responses post ASCT of 65% for those who were randomized to CVD (VGPR n=133, CR n=86) compared to 38% for those who went straight to transplant (VGPR n=120, CR n=46).

Sequential CVD was well tolerated with patients receiving a median of 4 cycles of therapy (range 1-8). Relevant grade 3/4 toxicities were: neutropenia 7.1%, thrombocytopenia 7.5%, anaemia 3.1%, peripheral neuropathy 5.1%.

Conclusion

For the first time we have shown that the use of a response-adapted therapy based on the use of chemotherapeutic agents with a different mode of action in myeloma can improve response rates, both pre- and post-transplant and that these translate into improved PFS.

On behalf of the NCRI Haem-Onc CSG