Intra-Individual Changes in Immune Marker Concentrations and Multiple Myeloma Risk: A Longitudinal Prospective Study Using Repeated Pre-Diagnostic Plasma Samples

Background: We studied immune marker trajectories (i.e. intra-individual changes in immune marker concentrations over time) of ten immune markers in relation to multiple myeloma (MM) risk. Low blood concentrations of MCP-3, MIP-1α, VEGF, FGF-2, Fractalkine, and TGF-α have been associated with myeloma risk in a previous study of single blood samples per participant collected around six years before diagnosis (Vermeulen et al., submitted). Longitudinal studies regarding these immune markers may provide substantial input to understanding their natural history in myeloma development and how these markers could be used in clinical routine.

Methods: We identified 66 MM cases who had donated two blood samples before myeloma diagnosis, as well as matched cancer-free controls from the Northern Sweden Health and Disease Study. Blood levels of MCP-3, MIP-1α, MIP-1β, VEGF, FGF-2, Fractalkine, TGF-α, IL-13, TNF-α, and IL-10 were assessed in duplicate using Luminex bead-based multiplex assays (Millipore, Billerca, MA). Samples from matched cases and controls, as well as repeated samples from the same individuals, were included in random order within the same analytical batch. Linear mixed effects modeling was applied to investigate differences in immune marker trajectories between cases and controls.

Findings: Samples were donated on median 12.3 ± 4.7 and 4.7 ± 3.8 years (± SD) prior to MM diagnosis for the baseline and the repeated sample, respectively. Median age at myeloma diagnosis was 62.9 ± 8.6 years (± SD). MM cases had significantly lower plasma levels of MCP-3, VEGF, FGF-2, Fractalkine, and TGF-α compared to controls. Notably, these immune markers also decreased significantly over time, while remaining temporally stable among control individuals. Decreasing marker trajectories were especially pronounced for VEGF (β = -0.012, P= 3.450 x 10^-4), FGF-2 (β = -0.010, P= 5.700 x 10^-5), and TGF-α (β = -0.019, P= 3.640 x 10^-4) among future myeloma patients the closer to diagnosis the sample was donated. Subgrouping of samples by time between blood draw and diagnosis showed that plasma levels of MCP-3 were inversely associated with myeloma risk in samples collected as long as 12 to 24 years before diagnosis (OR = 0.23, P_trend = 0.036) (OR = 4^th vs. 1^st quartile).

Interpretation: Our data provide further support of inverse associations between MM risk and plasma levels of MCP-3, VEGF, FGF-2, Fractalkine, and TGF-α. Low MCP-3 levels might be involved in early myeloma development. Considering that monoclonal gammopathy of unclear significance (MGUS) is detectable in the majority of myeloma patients in samples collected up to 15 years before diagnosis (Landgren et. al. Blood 2009, Weiss et. al. Blood 2009), decreasing immune marker levels of VEGF, FGF-2, and TGF-α might be indicative of progression to MM among MGUS patients. Therefore, these markers could be of potential use as biomarkers to improve prediction of myeloma progression in patients followed for MGUS. However, larger studies including samples from both, myeloma and MGUS patients together with matched controls are needed to further evaluate this hypothesis.