Evidence of Pre-Diagnostic Monoclonality in Blood Obtained up to 6.4 Years Preceding Diagnosis of Chronic Lymphocytic Leukemia (CLL).

Background: Monoclonal B-cell lymphocytosis (MBL) is reported in about 3% of the general adult population 65 yrs or older. Risk of progression from MBL to CLL requiring treatment has been estimated to about 1% per yr. For the first time we established the occurrence of preceding MBL using prospectively collected pre-diagnostic samples from CLL pts. Also, we tested the hypothesis that indolent (IgVH mutated) CLL occurs after MBL while aggressive (IgVH unmutated) CLL has an alternative route that develops rapidly or bypasses MBL.

Methods: Using the population-based nationwide PLCO (Prostate, Lung, Colorectal and Ovarian) cancer screening trial (n>144,500 cancer-free adults), we identified 46 CLL pts with available baseline cryopreserved whole blood obtained 3 to 77 months prior to CLL diagnosis. Clinical information was retrieved from screening questionnaires and medical records. 10 healthy adult controls were included. 6-color flow cytometry (CD45/CD19/CD5/CD10/kappa/lambda) and reverse transcription/polymerase chain reaction (RT/PCR) for IgH gene were used to determine evidence of expression of monoclonal IgH gene family and to establish the VH gene mutation status. Direct sequencing without subcloning (for clonal cases) and with subcloning (for non-clonal cases) was used to determine mutation status. We sequenced exons 5, 6, 7, 8, and 9 of the p53 suppressor gene for mutations.

Results: After exclusion of one subject with absent viable cells, we analyzed pre-diagnostic blood specimen from 45 CLL pts. CLL diagnosis was based on clinical criteria and confirmed by review of clinical records. Immunophenotyping data were available for 39 (87%) pts. The Rai CLL stage distribution was: 0 (n=26), 1 (n=9), 2 (n=2), 3 (n=1), missing/unclear (n=7). The mean time between pre-diagnostic blood draw and subsequent CLL diagnoses was 3 yrs. Using flow cytometry, sequencing, and PCR, we demonstrated evidence of monoclonality in 43/45 (96%) pre-diagnostic samples from CLL pts. Among individuals whose blood was obtained <3 yrs vs. >3 yrs prior CLL diagnosis, the proportion of pre-diagnostic monoclonality was 28/28 (100%) vs. 15/17 (88%), respectively; the proportion of IgVH mutated pre-diagnostic clones was the same in both groups: 23/28 (82%) vs. 14/17 (82%). The most frequently expressed VH gene family was VH3 (40%) followed by VH4 (29%), VH1 (9%), VH5 (9%), VH2 (7%), and VH6 (2%), with no expression of VH7 gene families. We found 2 pts with mutations in the p53 suppressor gene; one in exon7/M237K, and one in exon 6, P22 that is silent and has been reported as in some tumors and a polymorphism. One sample with unmutated IgVH appeared to have a deletion at p53 in both alleles. Immunophenotyping analyses for pre-diagnostic clones showed kappa n=27, lambda n=8, n=4 biclonal, and missing/indeterminate n=6 (kappa:lambda ratio >3:1). Comparisons of pre-diagnostic and diagnostic kappa/lambda status in the same persons showed 100% concordance. All negative controls lacked evidence of monoclonality.

Conclusions: In this first prospective study examining pre-diagnostic blood obtained 3 to 77 months prior to CLL diagnosis, we found a monoclonal B-cell population in virtually all the persons; 82% were IgVH mutated. Our findings suggest that CLL is commonly preceded by the precursor condition MBL.