Flow Cytometry (Immunophenotyping)

Flow cytometry measures light scatter and fluorescence from single cells passing a laser in a fluid stream, labelled with fluorochrome conjugated monoclonal antibodies against surface and intracellular antigens. Forward scatter reflects cell size and side scatter reflects granularity. Populations are separated by gating on CD45 intensity against side scatter, and then characterised by their antigen expression. It identifies lineage, maturity, aberrant phenotype and clonality, and measures cell numbers.

It is the first line test for suspected acute leukaemia and chronic lymphoid neoplasms, and is also used for paroxysmal nocturnal haemoglobinuria (PNH), plasma cell neoplasms, measurable residual disease, lymphocyte subsets and stem cell enumeration. It is performed in specialist laboratories, and urgent samples are usually reported within 24 hours.

Indications.

• Blasts or abnormal cells on the film, or unexplained persistent cytopenia, lymphocytosis (over 5 x10⁹/L), monocytosis or eosinophilia.
• Lymphadenopathy or effusions where fresh material or fine needle aspirate is available, and CSF in suspected CNS involvement.
• Suspected PNH (haemolysis with haemoglobinuria, aplastic anaemia, unusual site thrombosis).
• Suspected myeloma or related plasma cell disorder.
• Measurable residual disease monitoring in leukaemia and myeloma.
• Immunodeficiency and HIV (lymphocyte subsets, CD4 count).
• CD34-positive cell enumeration for stem cell collection, foetal cell quantification and EMA binding for hereditary spherocytosis.

Specimen. EDTA or heparinised blood, bone marrow aspirate (first pull to limit haemodilution), disaggregated fresh tissue in culture medium, and body fluids including CSF. Most samples are analysed within 24 hours, and CSF should be processed within 1 hour or collected into stabilising preservative. Viability under about 80 to 85% reduces reliability.

Panel design. A screening panel identifies the lineage and is followed by a tailored panel. B-cell, T-cell, myeloid and plasma cell panels use a minimum of 8 to 10 colours in many laboratories.

Clonality. B cells are assessed by surface kappa or lambda light chain restriction. A kappa to lambda ratio over about 3 or under about 0.3 within the B-cell gate suggests a clone, with assay specific thresholds. T-cell clonality is assessed by restricted T-cell receptor variable chain or TRBC1 expression, or by loss of antigens such as CD7.

Result integration. The result is combined with morphology, cytogenetics and molecular findings, and no single marker defines a lineage.

A Matutes score of 4 to 5 supports chronic lymphocytic leukaemia, and 0 to 2 argues against it.

PNH. Loss of GPI-anchored proteins is shown by FLAER and CD24 on neutrophils, FLAER and CD14 on monocytes, and CD59 on red cells. Clone size is reported as a percentage of each lineage. Granulocyte and monocyte clones are more reliable than red cell clones after transfusion or haemolysis. High sensitivity assays detect clones to about 0.01%, and small clones are common in aplastic anaemia and myelodysplasia.

Measurable residual disease. Sensitivity is about 1 in 10 000 to 1 in 1 000 000 cells, depending on the number of cells acquired and the assay. A threshold of 0.01% is widely used for risk stratification in B-lymphoblastic leukaemia.

Lymphocyte subsets and CD4 count. The normal adult CD4 count is about 500 to 1500 cells/microL. A count under 200 cells/microL carries a risk of Pneumocystis pneumonia and other opportunistic infections. Age specific ranges apply in children.

CD34 enumeration. Apheresis is usually started when the peripheral CD34 count is over 10 to 20 cells/microL. A collection of at least 2 x10⁶ CD34-positive cells/kg is usually needed for autologous transplant, and over 5 x10⁶/kg is ideal.

Critical findings needing immediate action.

• Abnormal promyelocyte phenotype (suspected acute promyelocytic leukaemia) needs same day haematology contact because of coagulopathy risk.
• Blasts with a leucocyte count over 100 x10⁹/L (leucostasis), or CSF involvement.
• T-lymphoblasts in a patient with a mediastinal mass and airway compromise.
• Burkitt phenotype (mature B cells with high proliferation).

Pitfalls.

• Clotted, aged, low viability or haemodilute samples give unreliable or falsely low blast and plasma cell percentages.
• Flow blast counts differ from morphological counts because of haemodilution and loss of fragile cells.
• Prior corticosteroids can lyse lymphoblasts, and rituximab, daratumumab and CD19-directed cell therapy can mask CD20, CD38 and CD19, so alternative markers are needed.
• Hodgkin lymphoma and many large cell lymphomas can be missed because the neoplastic cells are rare or fragile, so tissue histology is required.
• Haematogones (normal precursor B cells with a maturation pattern) can mimic B-lymphoblastic leukaemia, particularly in children and during marrow regeneration.
• Aberrant antigen expression, such as myeloid markers in lymphoblastic leukaemia, can confuse lineage assignment.

Special groups. Monoclonal B-cell lymphocytosis (clonal B cells under 5 x10⁹/L) is found in about 5 to 10% of adults over 60 years, and in more on high sensitivity assays. Children have higher lymphocyte counts and age specific subset ranges.

Follow up tests. Bone marrow aspirate and trephine biopsy, cytogenetics and FISH, molecular testing (for example NPM1, FLT3, BCR-ABL1, IGHV status, TP53), lymph node biopsy with immunohistochemistry (cyclin D1, SOX11, BCL2, Ki-67), serum protein electrophoresis with immunofixation, serum free light chains, immunoglobulin and T-cell receptor gene rearrangement studies, CSF cytology.