Introduction: Immunoparesis (IP) or the suppression of polyclonal uninvolved immunoglobulins (Ig) is one of the immune system alterations present in patients with multiple myeloma (MM) at diagnosis. Several studies have shown that the lack of IP recovery after treatment is associated with worse prognosis. However, these are observational retrospective studies and patients were not homogeneously treated. On the other hand, the depth of response, specially reaching minimal residual disease (MRD) negative, is one of the most important prognostic factors in patients with MM. We decided to study, in a prospective way, the prognostic value of IP recovery in a group of elderly patients not suitable for transplantation and treated in a similar way and compare it with the depth of response to treatment, especially with MRD status. Methods: We analyze patients included and treated in the PETHEMA/GEM2010MAS65 study (registered at www.clinicaltrial.gov as #NCT01237249) who reached CR or VGPR. All patients included in this trial were older than 65 years old and transplant-ineligible and were planned to receive 9 cycles of bortezomib plus melphalan and prednisone (VMP) and 9 cycles of lenalidomide plus dexamethasone (LD). Polyclonal Ig were analyzed by nephelometry or turbidimetry at diagnosis, at the end of the treatment (after 18 cycles) and during subsequent follow up. Minimal residual disease (MRD) was analyzed at the end of the treatment by multidimensional flow cytometry (MFC) with a sensitivity level of 10-5. Results: Of the 233 patients included in the study, 161 reached CR or VGPR, but the protocol design did not included determination of uninvolved Igs at diagnosis and for this reason they were only available in 120 patients. Only 9 cases (7,5%) had normal Ig at diagnosis while 111 patients (92,5%) had IP. There were no differences in progression free survival (PFS) and overall survival (OS) between patients with or without IP. We had Ig determination of 117 cases at the end of the treatment and during follow up. Those patients who had IP at diagnosis and recover it after treatment had better PFS (p<0,001; HR 0,33) and better OS (p=0,001; HR 0,35) than those who do not recover it (Figure 1A). We found no differences between those patients who had recovery of IP at cycle 18 than those who recover IP during follow up. When we focused the analysis in CR patients and VGPR patients separately, in both groups recovery of IP was associated with a better PFS and OS. In fact, when we compared both groups, we found that those patients with VGPR and complete recovery of IP had better PFS (p=0,001; HR 0,12) and OS (p=0,015; HR 0,22) than those with CR and no recovery of IP. These data suggest that IP recovery could be a prognostic factor equal or more important than conventional response. MRD information by MFC was available in 112 cases. As previously reported, in this trial those patients with MRD negative (-ve) had longer PFS and OS than MRD positive (+ve). When we analyzed IP recovery in MRD -ve patients we found that patients with MRD -ve and IP recovery (33 cases) had better PFS (p=0,004; HR 0,30) and better OS (p=0,010; HR 0,21) than patients with MRD -ve without IP recovery (11 cases), although it should be recognized that the number of patients within this latter subgroup was rather small (Figure 1B). We performed a multivariate analysis including other prognostic factors, and we found that both MRD and IP recovery have independent prognostic value for PFS, but only MRD and no IP recovery, is an independent prognosis factor for OS. Conclusions: Recovery of IP after non-continuous treatment in elderly patients with CR or VGPR confers better prognosis (PFS and OS). Prognostic value of IP recovery could be equal or more important than conventional response to treatment. Prognostic value of MRD could be complemented when combined with IP recovery, but these data must be confirmed in other series. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal