Skip to main content
IMS Events Main banner

    Imaging, QoL and Patient-Reported Outcome and Supportive Care

    Category: Imaging, QoL and Patient-Reported Outcome and Supportive Care

    Performance of Deauville Score and Total Lesion Glycolysis as Response Markers for Multiple Myeloma with Extramedullary Disease

    (PA-159) Performance of Deauville Score and Total Lesion Glycolysis as Response Markers for Multiple Myeloma with Extramedullary Disease

    Wednesday, September 17, 2025
    Introduction: Extramedullary disease (EMD) in multiple myeloma (MM) is associated with an aggressive clinical course and poor response to current therapies. The prognostic utility of the Deauville Score (DS) in EMD has not been well characterized, and it remains unclear whether total lesion glycolysis (TLG) provides additional value beyond DS in assessing treatment response.

    Methods: We included patients with EMD who underwent fluorodeoxyglucose (FDG) PET-CT for treatment response assessment between 01/01/2009 and 12/30/2022. DS was assigned per standard criteria: complete response (CR) as DS ≤3, partial response (PR) as DS = 4, and no response/progression (NR/PD) as DS = 5. TLG was calculated as the product of mean standardized uptake value (SUVmean) and the total volume of FDG-avid disease. Metabolic response categories based on TLG reduction were defined as: Deep Metabolic Response (DMR, >90% reduction), Major Metabolic Response (MMR, 50–90%), and Non-significant Metabolic Response (NMR, < 50%). Progression-free survival (PFS) was determined using the 2016 Consensus International Myeloma Working Group criteria.

    Results: Ninety-five patients with response PET scans were included (73 relapsed and 22 newly diagnosed MM). The median interval between response PET-CTs was 3.1 months (IQR: 1.5–4.7) and the median follow-up from the first PET was 78.4 months (95% CI: 55-93). Median PFS for the overall cohort was 4.2 months. By TLG response, median PFS was 16.5 months [5.9–NR] for DMR, 4.8 months [3.6–9.6] for MMR, and 2.2 months [1.7–3.4] for NMR (p< 0.0001). Stratified by DS, the median PFS was 22.4 months [95% CI: 11.5–NR] for DS ≤3, 4.3 months [3.6–9.6] for DS = 4, and 2.8 months [2.1–4.0] for DS = 5 (p< 0.0001). The difference between DS 4 and DS 5 groups was not statistically significant (p=0.3). In contrary, the difference between MMR and NMR was statistically significant (p=0.035). Among patients with DS ≥4, 23% (n=16) had discordant responses—worse DS but improved TLG; their median PFS was 4.8 months [3.5–8.2], compared to 2.9 months [1.9–3.8] in the concordant group (p=0.057).

    In a multivariable analysis, a lower reduction in TLG was independently associated with an inferior PFS [HR 2.3 (95% CI: 1.6-3.1) after adjusting for disease status (de novo versus secondary), triple-class refractoriness (TCR) and high-risk cytogenetics (HRCA). When both DS and TLG were added to the MVA along with TCR status, HRCA and de novo-secondary disease, TLG [HR 1.6 (95% CI: 1.1-2.1), p=0.008] and DS [HR 2.2 (95% CI: 1.5-3.1), p< 0.0001] were independently prognostic for PFS.

    The 18-month OS for patients achieving a DMR was 71% (95% CI: 57-89) vs. 38% (95% CI: 23-65) for MMR and 18% (95% CI: 9-35) for NMR (p< 0.0001).


    Conclusions: Both DS-defined complete response and DMR by TLG are associated with significantly improved PFS in EMD. TLG may provide additional discriminatory power for response assessment and should be considered in future response evaluation strategies for EMD.