Introduction: PCL is a rare, highly proliferative form of plasma cell dyscrasia with poor clinical outcomes. The current criteria for diagnosis requires ≥20% CPCs and/or absolute CPC count of ≥2 x109/L. Recently, the prognostic impact of CPCs below this threshold has been explored. Herein we report a single center experience of PCL, analyzing patients with primary PCL (pPCL), secondary PCL (sPCL), and those with various percentages of CPCs not meeting the current diagnostic criteria.

Methods: Patients treated between 2004-2017 with CPCs detected via a manual differential and/or flow cytometry were identified. Based on percentage of peripheral blood PCs, patients were stratified into three groups: 1-4% CPCs, 5-19% CPCs, and ≥20% CPCs. These groups were further divided into pPCL or sPCL, defined as CPCs present with de novo disease versus CPCs measured in the setting of relapsed/refractory myeloma, and spillover CPCs (fulfilling neither of the former categories). Clinical features, treatment responses, and survival outcomes were compared between the three groups. Continuous variables were compared using t-test; categorical variables were compared using Fisher's exact test. Overall survival (OS) was estimated by the Kaplan-Meier method. Risk factors for OS were assessed by Cox regression model.

Results: A total of 91 patients with CPCs were identified as Group 1: CPCs ≥20% in 45 (49%), Group 2: CPCs ≥ 5-19% in 11 (12%), and Group 3, CPCs ≤4% in 35 (38%). pPCL in 26 (29%), sPCL in 35 (38%), and spillover in 30 (33%). Patient characteristics at time CPCs were first identified are shown in the Table below. Median age was 56 years (25-84), 57% were female. Sixty six patients (73%) had high-risk cytogenetics, most common abnormalities were hypodiploidy, gain of 1q, and del 17p. There was no difference in treatment response or OS among different groups, suggesting factors other than cytogenetics may be critical in the pathogenesis of PCL. Forty four patients (48%) had extramedullary disease (EMD), of which 6 had CNS involvement. CD56 expression data was available for 68 patients, 30 had positive PC expression of CD56, of which 20 (67%) had EMD, p=0.09 is of marginal statistical significance. Immunomodulatory agents (IMiD) and proteasome inhibitors (PI) were the main treatments used in > 75% of the patients; 26% received combination of IMiD and PI with cytotoxic chemotherapy (VTD-PACE). Approximately half of the patients (48%) had an autologous SCT and 4.4% (n=4) underwent allogeneic SCT. Median OS for the whole group was 8.5 months (95% CI: 0.4-32.0). Group 1 had a median OS of 9.7 months (CI: 5.4-13.4), Group 2 had median OS of 2.6 months (CI: 0.4-NR), and Group 3 had median OS of 13.4 months (CI: 7.4-32.0). Univariable Cox regression model revealed that sPCL had a poor OS compared with pPCL (p=0.01) and spillover group (p=0.004). Inferior survival was noted for patients with low platelets (p=0.0002), high LDH (p=0.003), advanced ISS (p=0.006) and possibly in those with EMD (p=0.06). Treatment regimen had no impact on OS; VTD-PACE, the most commonly used regimen in pPCL had no impact on OS, probably reflecting the aggressive biology of the disease rather than effectiveness of any particular regimen.

Conclusions: Despite improvement in outcomes for MM, PCL patients continue to do poorly. Patients with ≥5% CPCs have similar clinical presentation and overall poor outcomes when compared to the historically defined PCL patients with ≥20% CPCs, supporting the recent recommendations to redefine PCL. Low platelet counts and high LDH are the most significant predictors of OS. Further studies with this expanded PCL patient population are warranted to better understand the significant drivers of disease, and lead to new innovative treatment approaches.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.