Impact of Severe Infections during Induction Therapy on Dosage, Response and Survival in Newly Diagnosed Multiple Myeloma - a Subgroup Analysis from the Randomized Phase III Trial GMMG-HD4

Background: The well-described prognostic impact of tumor characteristics and biology in multiple myeloma (MM), such as the combination of cytogenetics, the International Staging System (ISS) and lactate dehydrogenase (LDH, Moreau et. al., JCO, 2014) as well as frailty (Palumbo et al., Blood, 2015) significantly influence patient outcomes. However, only limited data on the impact of infections during therapy exist (Rajkumar et al., Lancet Oncology, 2010). Therefore, we hypothesized that severe infections during induction therapy (IT) in transplant-eligible MM influence dosage of therapies, treatment responses after IT and survival.

Patients and Methods: From 05/2005 until 05/2008, 399 patients were randomly assigned to receive IT with either three cycles of VAD (vincristine, VIN, i.v. 0.4mg, days 1-4; doxorubicine, DOXO, i.v. 9mg/m², days 1-4; dexamethasone, DEX, p.o. 40mg, days 1-4, 9-12, 17-20; n=201, arm A) or PAD (bortezomib, BTZ, i.v. 1.3mg/m², days 1, 4, 8, 11; DOXO i.v. 9mg/m², days 1-4; DEX p.o. 40mg, days 1-4, 9-12, 17-20; n=194, arm B), followed by high-dose melphalan (HDM) and autologous stem cell transplantation (ASCT) and either thalidomide (arm A) or bortezomib (arm B) maintenance within the German part of the joint GMMG-HD4/HOVON65 trial (Sonneveld et al., JCO, 2012). After exclusion of ineligible patients, 395 patients (99.0%) were evaluable for analyses.

Any severe infection (equal or greater grade 3, according to the Common Terminology Criteria for Adverse Events, Version 4.0) during IT (at least once, defined from first until last date of application of IT medication) occurred in 105 patients (VAD n=53/198 and PAD n=52/192, 26.9% of all patients, missing data n=5).

Results: Among patients with a severe infection during IT in the VAD and PAD arms, total DEX and DOXO doses (equal dosage in VAD/PAD group) were significantly lower (median DEX dose (mg/m²): 689.0 [77.7, 1014.1] vs. 742.3 [0.0, 1324.1], p<0.001 and median DOXO dose (mg/m²): 106.9 [33.0, 115.4] vs. 107.6 [27.6, 149.5], p<0.001). Accordingly, the BTZ dose during IT in the PAD group was significantly lower in patients with severe infections (median BTZ dose (mg/m²): 15.1 [5.1, 16.6] vs. 15.5 [1.3, 16.4], p<0.001). Combined PAD and VAD very good partial response rates or better (VGPR+) after IT were 27.6% vs. 19.9% (p=0.12) for patients with or without a severe infection during IT.

Overall survival (OS) was significantly shortened in patients with at least one severe infection during IT (median OS: 81.8 months vs. not reached, p=0.04, Figure 1A). OS plots diverged in the early period of observation (< 3 months), driven by infection-related deaths (n=8). A landmark analysis 3 months after registration demonstrated approximated survival curves without significant differences in OS (median OS: 78.8 months vs. not reached, p=0.30, Figure 1B). Similarly, progression-free survival (PFS) was shortened, though not significantly (median PFS: 30.2 vs. 35.0 months, p=0.08). However, since not just death accounts as PFS event, the impact of infection-related deaths on PFS remains smaller than on OS. Accordingly, landmark analyses after 3 months from registration showed again closer survival curves (median PFS: 28.5 vs. 32.4 months, p=0.36).

Conclusions: Severe infections have a critical impact on the applied doses of IT and outcome in the early, vulnerable phases of MM therapy. OS for transplant-eligible MM patients with severe infections during IT was significantly shortened, mainly driven by early infection-related deaths (< 3 months). A reduction of DEX doses during PAD/PAd IT in the subsequent GMMG study generation (GMMG-HD4/HOVON65: 480mg/cycle to GMMG-MM5: 240mg/cycle) and the recommendation of antibiotic/antiviral prophylaxis throughout the whole IT led to a reduced rate of severe infections of 12% (PAd) in the GMMG-MM5 trial. Further analyses are needed to elucidate how severe infections can be avoided, and whether there is an overlap between the subgroup of patients with severe infections during IT and patients with known adverse prognostic factors or reduced fitness/pre-existing conditions.

Figure 1: Impact of severe infections on overall survival.

(A) Overall survival and (B) landmark analysis on overall survival 3 months after start of induction therapy of patients with or without at least one severe infection (equal or greater grade 3) during induction therapy.