Biological and Clinical Significance of Undetectable Circulating Tumor Cells (CTCs) in Patients (Pts) with Multiple Myeloma (MM)

Background: There is evidence suggesting that CTCs are responsible for the spreading of clonal plasma cells (PC) inside and outside the bone marrow (BM). However, some pts unexpectedly have no CTCs at diagnosis. Here, we hypothesize that the absence of CTC defines a specific MM subtype with unique biological and clinical features. To our knowledge, this has never been investigated.

Aim: Define the frequency and the characteristics of MM pts with undetectable CTCs.

Methods: The presence of CTCs was assessed by NGF in 1,687 blood samples from pts with MGUS (n=467), smoldering (SMM, n=368), newly diagnosed (NDMM, n=652), and relapsed refractory MM (RRMM, n=200). All pts were enrolled in prospective PETHEMA/GEM multicenter studies. BM clonal PC from pts with (n=239) and without (n=17) detectable CTCs were isolated by FACS prior exome and RNA sequencing.

Results: The frequency of pts with undetectable CTCs progressively decreased (p <.001) from MGUS (54%) to SMM (21%) and NDMM (8%) and RRMM (11%). In those with detectable CTCs, the median percentage was 0.0003, 0.002, 0.01 and 0.008, respectively (p <.001). The 1-log increment observed in the progression from MGUS to SMM and NDMM underpins the link between CTC levels and malignant transformation. The similarity between NDMM and RRMM is probably related with the fact that the latter were analyzed in between different lines of therapy.

When compared to other SMM pts, those with undetectable CTCs were more frequently staged with low risk according to the 20/2/20 IMWG model (32% vs 49%; p <.001). Absence of CTCs (HR: 0.38; p =.042) and the 20/2/20 IMWG model (HR: 0.46; p =.011) had independent prognostic value in a multivariate analysis. SMM pts with undetectable vs detectable CTCs showed lower risk of transformation (progression rates at 2 years of 7% vs 24%, p =.004).

When compared to other NDMM pts, those with undetectable CTCs had significantly less anemia and BM infiltration. There were no differences in the incidence of lytic lesions, renal insufficiency and hypercalcemia. Only one pt without CTCs was staged with R-ISS 3. These findings, together with a higher frequency of plasmacytomas, suggest that the MM subtype with undetectable CTCs may include pts with macrofocal disease.

In transplant eligible NDMM, pts with undetectable vs detectable CTCs had PFS rates at 7 years of 90% and 44% (HR: 0.14; p <.001), and OS rates of 97% and 72% (HR: 0.10; p =.02). In transplant ineligible NDMM, pts with undetectable vs detectable CTCs had PFS rates at 2 years of 83% and 54% (HR: 0.31; p =.02), and OS rates of 94% and 76% (HR: 0.33; p =.13). In a multivariate analysis including the R-ISS and transplant eligibility, the absence of CTCs showed independent prognostic value for PFS (HR: 0.25; p <.001) and OS (HR: 0.24; p =.014). Among pts with undetectable CTCs, there were no differences in PFS and OS according to the presence of standard vs high risk cytogenetic abnormalities. In contrast with pts having detectable CTCs, the PFS and OS of those without CTCs was not significantly reduced if CR or MRD negativity were not achieved after treatment.

The cytogenetic profile of BM clonal PC from pts with undetectable (n=112) vs detectable (n=846) CTCs was characterized by a lower frequency of +1q21 (16% vs 45%, p <.001) and del(1p32) (2% vs 8%, p =.02). Other alterations were not significant. There was no correlation between the number of mutations and the percentage of CTCs. There were 66 differentially expressed genes (DEG) between BM clonal PC from pts with detectable vs undetectable CTCs. Of note, 9 of the 66 DEG were in Chr1, with up- and downregulation of genes in 1p and 1q, respectively. There was a significant correlation between CTC levels and the expression of genes associated with MM dissemination (FLNA, LGALS1 and CXCR4). These findings, together with the modest correlation between the percentage of CTCs and BM clonal PC determined by morphology (r=.45, p <.001) and NGF (r=.49, p <.001), suggest that the selective egress of tumor cells from BM into the PB might depend on specific genetic alterations and transcriptional priming more than tumor burden.

Conclusions: This study uncovered that ~8% of MM pts may progress due to mechanisms different from CTC dissemination. This subgroup showed unique features and achieved unprecedented outcomes, particularly if eligible to intensive therapy. Accordingly, our results endorse the recognition of pts with undetectable CTCs as a new biological and clinical subtype of MM.