Circulating tumor cells (CTCs) improve risk stratification according to the second revision of the international staging system (R2-ISS) and the ims-IMWG consensus genomic staging (CGS) in newly diagnosed multiple myeloma (NDMM): An analysis from the european CTC consortium.

Abstract

Abstract Background. CTCs are independent prognostic biomarkers in NDMM patients (pts) stratified by International Staging System (ISS) and Revised ISS. However, the relevance of CTCs in the context of R2-ISS and the new IMS-IMWG definition of high-risk myeloma (Consensus Genomic Staging, CGS) has not been investigated. The aim of this analysis was to evaluate the clinical significance of CTCs in combination with R2-ISS and the IMS-IMWG CGS in a large cohort of NDMM pts. Methods. The European CTC Consortium collected patient-level data of NDMM pts from 5 collaborative groups (Czech Republic, Greece, Italy, HOVON/Netherlands-Belgium, PETHEMA/Spain). All included pts had CTC enumeration by flow cytometry prior to treatment. To investigate the role of CTCs in combination with R2-ISS we analyzed 1657 NDMM pts (R2-ISS cohort) with complete data on the R2-ISS defining variables [ISS, del(17p), t(4;14), LDH levels, 1q+]. A value was assigned to each risk feature and pts were stratified into 4 risk groups according to the total additive score (D'Agostino et al JCO 2022). To investigate the role of CTCs in combination with IMS-IMWG CGS we analyzed 190 NDMM pts (CGS cohort) with molecular data available. TP53 mutations, biallelic del(1p32) (CDKN2C log2 < -1.3219) and t(14;20) were analyzed by whole genome/exome sequencing of CD138+ tumor cells . CGS high risk was defined as per IMS/IMWG recommendations (Avet-Loiseau et al JCO 2025), In both cohorts CTCs were analyzed as log10 increase (as a continuous variable and as discrete intervals from ≤0.001% to >1%) and using a binary cutoff (£ 0.02% vs >0.02%, Bertamini et al EHA 2025) Results. In the R2-ISS cohort, 17% of pts were R2-ISS I, 29% R2-ISS II, 46% R2-ISS III and 9% R2-ISS IV. The distribution of pts with ≤0.001%, ≤0.01%, ≤0.1%, ≤1% and >1% CTCs was 20%, 20%, 31%, 18% and 10%. After a median follow-up of 50 months, both CTCs log10 increments (HR 1.15 p<0.001 for PFS; HR 1.13 p<0.001 for OS) and R2-ISS (R2-ISS II vs I HR 1.20 for PFS 1.16 for OS; R2-ISS III vs I HR 1.67 for PFS 1.90 for OS; R2-ISS IV vs I HR 2.66 for PFS HR 3.77 for OS; p<0.001) were independent predictors of outcome in a multivariable model including site, transplant eligibility and age. A prognostic model including CTCs as log10 intervals + R2-ISS (c-index 0.703 for PFS and 0.743 for OS) or CTCs with a binary cut-off + R2-ISS (c-index 0.698 for PFS and 0.736 for OS) performed better than R2-ISS alone (likelihood ratio test of both models vs R2-ISS p<0.001). Using a 0.02% cut-off, in R2-ISS I 77% vs 33% of pts have low vs high CTCs (median PFS 99 vs 72 months); in R2-ISS II 60% vs 40% of pts have low vs high CTCs (median PFS 83 vs 48 months); in R2-ISS III 41% vs 59% of pts have low vs high CTCs (median PFS 40 vs 31 months); in R2-ISS IV 22% vs 78% have low vs high CTCs (median PFS 29 vs 15 months) (p<0.001). In the IMS-IMWG CGS cohort, 45% of pts were high-risk possibly due to greater availability of molecular data in pts with high tumor burden. The distribution of pts with ≤0.001%, ≤0.01%, ≤0.1%, ≤1% and >1% CTCs was 17%, 9%, 31%, 29% and 13%. After a median follow-up of 91 months, both CTC log10 increments (HR 1.31 p<0.001 for PFS; HR 1.24 p=0.036 for OS) and IMS-IMWG CGS high risk (HR 1.55 p=0.047 for PFS; HR 2.51 p=0.003 for OS) were independent predictors of outcome in a multivariable model including site, transplant eligibility and age. A prognostic model including CTCs as log10 intervals + CGS (c-index 0.669 for PFS and 0.730 for OS) or CTCs with a binary cut-off + CGS (c-index 0.650 for PFS and 0.719 for OS) performed better than CGS alone (likelihood ratio test of both models vs CGS p<0.05) Using a 0.02% cut-off, 29% of pts were classified as CGS standard risk + low CTCs, 25% as CGS standard risk + high CTCs, 7% as CGS high risk + low CTCs and 38% as CGS high risk + high CTCs. Median PFS was not reached in CGS standard risk + low CTCs, 72 months in CGS standard risk + high CTCs, 64 months in CGS high risk + low CTCs and 33 months in CGS high risk + high CTCs (p<0.001). Overall CGS standard risk + low CTCs delineates a favorable prognosis group, CGS high risk + high CTCs delineates a poor prognosis group, while all other combinations identify pts with an intermediate prognosis. Conclusions. CTC levels have independent prognostic value in NDMM pts stratified with R2-ISS and IMS-IMWG CGS. These data support the investigation of CTCs in combination with current prognostic models and high-risk definitions.