Search

Background The 2/20/20 model to predict risk of transformation in SMM is established at diagnosis and not periodically re-evaluated, mainly because plasma cell (PC) quantification requires invasive bone marrow (BM) aspirates. It could be hypothesized that dynamic monitoring of tumor biomarkers could improve risk-stratification of SMM patients, but this requires minimally invasive methods to replace those performed in BM samples. Such methods should also monitor immune profiles to identify patients with stable tumor burden/genetics, but at risk of progression due to lost immune surveillance. Aim Compare the prognostic value of PC quantification in BM vs the evaluation of circulating tumor cells (CTCs) in peripheral blood (PB), and to define immune signatures predictive of time-to progression (TTP) in SMM. Methods 300 SMM patients are planned to be enrolled in the iMMunocell study that includes 26 sites across 8 European countries. PB samples are collected every 6 months during three years for next-generation flow (NGF) monitoring of CTCs and immune profiling of the T, NK, B and myeloid cell compartments. Cutoffs for risk stratification according to CTCs were calculated using maxstat. A total of 245 SMM patients were enrolled, and we report here a pre-planned interim analysis on the first 150. Results Thus far, 28/150 (19%) patients progressed to MM. Presence of ≤20% vs >20% PCs in BM by morphology failed to predict different TTP(19% vs 32% progressions at 2 years, respectively; p=0.14). Patients with ≤0.73 CTCs/µL had lower risk of transformation vs those with >0.73 CTCs/µL (15% vs 67% progressions at 2 years, respectively; HR: 5.1, p=4.9e-05). Patients with >0.73 CTCs/µL had median TTP of 17.5 months. Patients with low, intermediate and high-risk SMM according to the 2/20/20 model showed significantly different TTP (p=5.14e-06), but those with high-risk had median TTP not reached. Risk stratification using a new 0.7/2/20 (where 0.7 stands for >0.73 CTCs/µL) achieved identical statistical significance (p=5.14e-06), and patients with high-risk SMM had median TTP of 21 months. By applying the gradient boosting algorithm to the T cell dataset, we identified six subsets with an exhausted phenotype CD4+CD28negTIGIT+CD127lo, CD4+CD28+TIGIT+CD127+, CD4+CD28+TIGIT+CD127+CD25+, CD8+CD28negCD127+, CD8+CD28negTIGIT+ and CD8+CD28negTIGIT+PD1+ whose increased frequency was associated with inferior TTP. Patients showing an expansion of these T cell subsets had higher risk of transformation to active MM (HR: 7.33; p=0.002). Conclusions This is the first study performing CTC and immune monitoring every 6 months in PB samples from patients with SMM. Our results suggest that CTC numbers have greater prognostic value than BM PC counts, and that a new 0.7/2/20 model could be dynamically assessed to identify SMM patients at risk of developing active MM. Beyond CTC numbers,this study is also uncovering key immune cell types associated with disease progression.