38P - Flow cytometry immunophenotyping of pleural fluid cytology in patients with diffuse large B-cell lymphoma (DLBCL) and malignant pleural effusion

Date 07 May 2017
Event ELCC 2017
Session Poster Display Session
Topics Thoracic Malignancies
Translational Research
Basic Principles in the Management and Treatment (of cancer)
Presenter Franco Lumachi
Citation Annals of Oncology (2017) 28 (suppl_2): ii9-ii13. 10.1093/annonc/mdx089
Authors F. Lumachi1, P. Ubiali2, R. Tozzoli3, S.C. Sulfaro4, S.M. Basso2
  • 1Department Of Surgery, Oncology & Gastroenterology, University of Padua, School of Medicine, 35128 - Padova/IT
  • 2Department Of Surgery, S. Maria degli Angeli Hospital, Pordenone/IT
  • 3Department Of Laboratory Medicine, Clinical Pathology Laboratory, S. Maria degli Angeli Hospital, Pordenone/IT
  • 4Department Of Laboratory Medicine, Pathology & Histopathology Section, S. Maria degli Angeli Hospital, Pordenone/IT



Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of NHL and comprises up to 35% of cases. In patients with DLBCL the clinical presentation varies widely. Malignant pleural effusion (MPE) represents a common occurrence in cancer patients with either thoracic and extra-thoracic malignancies, often requiring invasive diagnostic procedures, including video-assisted thoracoscopic surgery (VATS)-guided biopsy. Approximately 20% of patients with DLBCL may develop MPE. The aim of this study was to demonstrate the usefulness of flow cytometry phenotyping (FCP) of pleural cytology (PC) in patients with MPE due to DLBCL.


A retrospective chart review of 40 (15 (37.5%) males and 25 (62.5%) females, median age 69 tears, range 46-85 years) patients with histologically confirmed MPE was performed at a tertiary referral Medical Center. All patients underwent PC and FCP before VATS-guided biopsy. For the FCP, the FACSCanto II system and the FACSDiva software (BD Biosciences, Franklin Lakes, USA) were used. Compensation beads were obtained with single stains of each antibody to determine the compensation settings, and the settings were applied in FlowJo software (Tree Star, Ashland, USA) after data collection. Sample cells were stained with a panel of antibodies, including CD5, CD10, CD19, CD23, Kappa, Lambda, and CD45 (Dako, Glostrup, Denmark). In the presence of a T-cell population with aberrant phenotypes or a B-cell population with monotypic light chains, the diagnosis of DLBCL-related pleural effusion was made.


Final histology showed 5 (12.5%) patients with DLBCL and 35 (87.5%) with MPE from other malignancies (mainly breast cancer and lung cancer). The age (68.0±10.1 vs. 67.5±14.2 years; p = 0.94) and the male to female ratio (p = 0.64) did not differ between groups. The results are reported in the Table. The sensitivity and negative predictive value of FCP reached 100%, the specificity was the same as obtained with the PC alone, but the accuracy was superior (97.5% vs. 85%; p = 0.003).rnTable: 38P

Results of flow cytometry phenotyping and pleural cytology

ResultsFlow cytometry immunophenotypingPleural fluid cytology
Specificity97.1% (95% CI: 91.6-100)97.1% (95% CI: 91.6-100)
Negative predictive value100%97.2%
Clinical accuracy97.5%85.0%
False positive rate2.9%2.9%
False negative rate0%14.3%
Likelihood ratio positive350
Likelihood ratio negative01.3


FCP of PC has great value in confirming (or excluding) the relationship between pleural effusion and malignancy, including lymphoma, exhibiting a significantly higher accuracy that that of PC alone. FCP should be performed in all patients with a history of DLBCL who exhibit suspicious MPE.

Clinical trial identification

Legal entity responsible for the study

Università degli Studi di Padova


Università degli Studi di Padova


All authors have declared no conflicts of interest.