Spinal Disease Predicts CNS Involvement in Patients with Plasma Cell Leukemia

Background:

Plasma cell leukemia (PCL) is a rare and aggressive hematologic malignancy with an incidence of approximately 1,200 patients per year in the United States with a median disease-specific survival of 6 months. PCL can either originate de novo (primary PCL) or as a secondary leukemic transformation of multiple myeloma (secondary PCL). A feared complication is infiltration of plasma cells into the central nervous system (CNS), meninges, or cerebrospinal fluid (CSF) which is challenging to treat and most often results in early mortality. We have observed a higher incidence of CNS disease among patients with PCL compared to multiple myeloma and were motivated to conduct a retrospective analysis to identify risk factors for the occurrence of CNS involvement among patients with primary or secondary PCL.

Aim:

Identify risk factors that should prompt clinicians to obtain a lumbar puncture and neuroimaging to screen for CNS involvement in patients with PCL.

Methods:

We conducted a retrospective analysis utilizing clinical data from patients who were found to have greater than 5% abnormal plasma cells in their peripheral blood detected by flow cytometry at the Seattle Cancer Care Alliance from 1990 to the present. IRB approval was obtained before data abstraction. Features extracted included laboratory values, radiographic data, pathologic data, treatment regimens, and response outcomes to chemotherapy regimens. The Fisher test was used identify risk factors associated with CNS disease. A P value <0.05 indicated a significantly asymmetrical distribution. The Kaplan-Meier product-limit method was used to estimate survival functions for groups in each risk factor. To determine the independent prognostic ability of each risk factor, multivariate survival analysis was performed using a Cox proportional hazards model.

Results:

To date, we have studied 21 patients which include 12 males and 9 females. Among this group, 18 patients were Caucasian, 2 patients were African American, and 1 patient's ethnicity was unknown. Four patients were diagnosed with primary PCL and 17 patients had secondary PCL. PCL was detected in the CSF by lumbar puncture in 5 patients (24%). The median overall survival of patients with CNS involvement from the time of PCL diagnosis was 192 days compared to 722 days in patients without CNS involvement. We reasoned that leukemic cells may translocate to the CSF more commonly in patients with spinal disease and evaluated the prevalence of spinal abnormalities. We found that all patients with CNS involvement and only 72% without CNS involvement had one or more spinal abnormalities which included degenerative disc disease, compression fractures of the spine, or lytic lesions of the spine (P < 0.05). We also found a significant correlation between CNS involvement by PCL and Caucasian race, a diagnosis of secondary PCL, and initial treatment with cyclophosphamide and dexamethasone (P < 0.05)

Conclusions:

In our study population, we observed a high incidence of CNS involvement (24%) among patients with PCL and identified multiple risk factors for the disease including evidence of malignant and non-malignant disease of the spine. These results suggest all patients with PCL should have spinal imaging performed at diagnosis and lumbar puncture should be considered for all patients, especially those with known spinal disease.