Patients With DLBCL Commonly Have B-Cell Lymphopenia and Circulating Clonal B-Cell Populations With a Phenotype Concordant With The Underlying Tumour

Clonal B-cell populations with either a CLL or a non-CLL phenotype are a common finding in normal individuals but uncertainty remains about how this relates to the development of clinically significant disease. The aim of this study was to investigate the frequency of peripheral blood clonal B-cell populations and B-cell subset abnormalities in newly presenting DLBCL patients and to determine whether the incidence of these abnormalities differed between the GCB and ABC subtypes, which are regarded as having distinct pathogenesis. The study was carried out using peripheral blood samples collected from patients entered in the UK-REMoDL-B trial. This trial is testing the hypothesis that the ABC subtype of DLBCL responds preferentially to R-CHOP- Bortezomib. Gene expression profiling is performed on the diagnostic tissue biopsy (FFPE) using the Illumina WG-DASL assay prior to randomisation classified as GCB, ABC or unclassified (UN). The availability of GEP data allows meaningful comparison with the phenotype of clonal populations detected by flow cytometry.

Peripheral blood taken prior to first treatment was analysed using multi-colour flow cytometry. Following red cell lysis with ammonium chloride, samples were incubated with a panel of antibodies comprising of a CD19 and CD20 backbone, with Kappa, Lambda, CD5, CD45, CD49d, LAIR-1, CXCR5, CD31, CD95, CD38 and CD10, supplemented in some cases by CD81, CD79b, and CD43. A minimum of 500,000 events were acquired on a FacsCanto II flow cytometer (Becton Dickinson). B-cells were enumerated and any monoclonal populations identified were classified as CLL, germinal centre (GC), non-GC or not otherwise specified (NOS) where the phenotype was indeterminate.

358 samples were eligible for inclusion from patients with an average age of 62.2years (range 22.9-86.1). Abnormalities were detected in 52% of cases (B-lymphopenia ((<0.06 x 10⁹/l) 33%, B-lymphocytosis (>1 x 10⁹/l) 2.8%, CLL clone 3.6%, GC clone 9.8%, non-GC clone 9.8%, clonal population NOS 2.2%). Gene expression profiling results were available for 278 individuals; 51% GCB, 32% ABC and 17% unclassified. The relationship between peripheral blood B-cell findings and the GEP determined phenotype of the tumour is shown in the table:

In patients where clonal populations were detected in the peripheral blood there was striking concordance between the phenotype of the clone and the GEP of the underlying tumour. Presence of a GC-population by flow was highly predictive of GCB GEP (84% GC–type populations detected were in GCB cases). The number of discordant cases and the number of CLL clones detected approximate to the numbers that would be expected in a normal population of a similar age. It is, therefore, likely that in most cases circulating tumour cells or a closely related precursor clone are being detected. The similarity between the results of the ABC and unclassified GEP groups suggest that these are biologically related. An unexpected finding in this study was the high incidence of B-lymphopenia at a level that might be expected to be associated with increased risk of infection. This may reflect suppression of normal B-cells by the neoplastic clone or be a marker of underlying immune dysfunction that may predispose to the development of the tumour. Immuosuppression has a role in the pathogenesis of DLBCL in the elderly and this study suggests that this may also be a factor in the wider patient population. These results may have implications for prognostic assessment and may offer opportunities for early diagnosis and possibly response assessment in some patients. The impact on outcome will be assessed in the course of the trial.