SERUM LEVELS OF SELECTED PARAMETERS OF BONE MICROENVIRONMENT PREDICT OVERALL SURVIVAL IN MULTIPLE MYELOMA PATIENTS REGARDLESS OF TREATMENT REGIMEN USED

ABSSUB-3989

Background: Several cytokines have been identified as potent inductors of myeloma bone disease (MBD) or growth factors in multiple myeloma (MM). Their serum levels often correlate with the extent or aggressiveness of the disease, and, even with prognosis of MM.

Aims: Our aim was to assess serum levels of selected parameters of bone marrow microenvironment with respect to survival functions.

Methods: We have assessed serum levels of 6 selected parameters of bone marrow microenvironment – hepacyte growth factor (HGF), syndecan-1 (SYN), osteopontin (OPN), macrophage inhibitory protein 1α and 1β (MIP-1α, MIP-1β), osteoprotegerin (OPG) and endostatin (END) in a cohort of 52 patients with MM with regard to overall survival (OS), progression free survival (PFS), time to progression (TTP) and duration of response (DOR) after induction treatment line.

The patients were treated using biological based regimens (i.e. with thalidomide, bortezomib and lenalidomide) or with high-dosed chemotherapy with support of autologous stem cell transplantation.

For statistical estimation we used Mann-Whitney U test, ROC analysis, Kaplan-Meier analysis, Cox regression and log rank test at p < 0,05.

Results: Out of the 6 evaluated parameters, we found significant correlation with OS in HGF (p = 0,042), OPN (p = 0,006) and OPG (p = 0,029), the other parameters did not have significant differences. Based on these results we constructed ROC curves to find optimal cut-off value with the best discriminatory potential. For HGF the optimal cut-off value was 1919 ng/l with 70% sensitivity and 68% specificity. Patients with HGF < 1919ng/l had significantly longer OS than patients with HGF ≥ 1919ng/l (M 6,6 vs 3,1 years, p = 0,001). In the case of OPN, the optimal cut-off level was 97 ug/l with 68% sensitivity and 84% specificity, dividing the cohort into a group of patients with good prognosis with OPN < 97 ug/l and poor prognosis with OPN ≥ 97 ug/l (M 8,0 vs 2,9 years, p < 0,0001). The optimal cut-off level for OPG was 5,2 pmol/l with 69% sensitivity and 64% specificity. Patients with OPG < 5,2 pmol/l had a benefit in survival with respect to patients with OPG ≥ 5,2 pmol/l (M 6,1 vs 3,6 years, p = 0,015).

There were no significant differences between any of the selected parameters with respect to induction treatment line, i.e. to PFS, TTP and DOR.

Using the logistic regression and Cox regression analysis we constructed a model for short survival. Significant predictors of short survival were OPN ≥ 101 ug/l and OPG ≥ 5,2 pmol/l. OPN ≥ 101 ug/l significantly increased the risk of poor prognosis 5,3 times (95%CI: 1,4 - 19,9; p = 0,013). OPG ≥ 5,2 pmol/l significantly increased the risk of poor prognosis 4,3 times (95%CI: 1,2 - 16,2; p = 0,030).

Summary/Conclusion: Analysis of serum levels of the parameters of bone microenvironment contributes to closer understanding of biological mechanisms occuring in MM. Our analysis revealed three parameters with prognostic potential in multiple myeloma. Patients with high serum levels of hepatocyte growth factor, osteopontin and osteoprotegerin had significantly shorter overall survival regardless of treatment modality used. There were, however, no differences in PFS, TTP and DOR after the first line treatment suggesting long term outcomes of bone microenvironment processes rather than short term influence of the tumor burden.

Supported by the grant IGA MZ CR NT 14393, NT14400, NT12451-5 and NT12215-4.

Keywords: None

http://www.ehaweb.org/