Send to printer »

Feature Articles : Aug 1, 2013 (Vol. 33, No. 14)

Hematologic Diagnostics Go with the Flow

  • Thalyana Smith-Vikos

Those diagnosing hematopathological disorders have been reaping the benefits of improved flow cytometry methodologies, though standardization of some techniques is still in the works.

Flow cytometry is being used to identify and evaluate clinical cases of various blood disorders. Researchers have also adapted other types of cellular analyses, such as cytogenetics and molecular genetics, in order to monitor disease states.

Anna Porwit, M.D., Ph.D., at the University of Toronto, has been immunophenotyping lymphoproliferative disorders with more than one malignant cell population. To investigate these diseases, Prof. Porwit and her colleagues use flow cytometry, which is a common immunophenotyping method in hematopathological diagnosis, to identify aberrant populations of cells in the blood and bone marrow.

As medical director of the university’s flow cytometry lab, Prof. Porwit instituted a 10-color flow cytometry method, using the Navios™ instrument from Beckman Coulter. Navios, which contains three lasers, allows Prof. Porwit and her colleagues to analyze 10-14 different antibodies simultaneously on the same cell. By using monoclonal antibodies with different connected fluorochromes, 10-14 different antigens can be assayed in one tube.

To study lymphoid malignancies, Prof. Porwit utilized two 10-color tubes of surface markers for B- and T-lymphocytes, respectively. She then evaluated kappa/lambda ratios separately in various B-cell subsets. Prof. Porwit noticed that in some lymphoma cases, there were two different pathological populations of B-cells or one of B- and one of T-cells in one sample, which was quite unusual because typically there is only a single aberrant population of cells.

In fact, Prof. Porwit and her colleagues identified a cohort of cases where there were clearly two different malignant populations. From 2,600 samples of blood, bone marrow, fine needle aspiration, and tumor cell suspensions, the researchers identified 43 samples showing two abnormal lymphoid populations—40 of which had two aberrant B-cell populations, the remaining three showing one B- and one T-cell aberrant population.

“Previously, we were using a five-color flow cytometer, but switching to the 10-color panel has allowed for increasingly more accurate diagnostics, as these lymphoma cases with two aberrant cell populations were easier to detect by identifying many more parameters for each cell,” Prof. Porwit says. “Additionally, this improves the accuracy of our data: Instead of doing several different assays in multiple tubes and attempting to compare the tubes to each other, now we have everything in one assay.”

Prof. Porwit explains that this 10-color flow cytometry method is an important advancement in cellular analysis, not only because of the machinery, but also the types of software for analyzing the data. The Kaluza® software provides easy protocols for analyzing this complex data of 12 parameters per cell: 10 fluorochromes, cell size, and granularity.

She also notes, though, that it is critical to monitor fluorescence compensation during data analysis, by calculating the amount of interference of a fluorochrome’s emission profile in a different channel. Ideally, fluorochromes at different ends of the spectrum should be used so that there is no spillover, but this is not always practical with analysis of increasing numbers of antibodies.

Prof. Porwit will continue to use 10-color flow cytometry for hematopathological diagnosis, especially myelodysplastic syndrome. She tells GEN that a research initiative through the International Council for Standardization in Hematology is currently establishing guidelines for identifying malignancies in acute myeloid leukemia and myelodysplastic syndrome using flow cytometry methodologies.

Flow Techniques for Counting Cells

Michael Keeney from the London Health Sciences Centre is working to improve flow cytometry methods in cases where cell counts are particularly low. Keeney has collaborated with breast cancer researchers to detect circulating tumor cells in peripheral blood at 1 cell/mL. He has also worked on the detection of minimal residual disease in pediatric patients with acute lymphoblastic leukemia after induction therapy.

Keeney tells GEN that rare event detection of acute lymphoblastic leukemia is critical for re-stratifying a patient group based on their response to a particular therapy. These small populations of residual tumor cells, which may be on the order of one in 10,000 cells, establish the patient’s prognosis and will inform the clinician regarding whether or not to alter the treatment.

Leukemic patients may have up to 1012 leukemic cells, and in classical morphologic "remission" there can still be up to 1010 residual leukemic cells. In his study, Keeney examined antigen expression in the remission state—i.e., 29 days after therapy initiation—in order to quantify residual tumor cells. He noted that minimal manipulation of the sample is ideal, as certain preparation methods may in fact lead to a loss of the abnormal cell population in the sample. Samples should be prepared at a higher concentration than what is normally used for flow cytometry; if a sensitivity of 0.01% is target, at least 1 x 106 cells should be collected.

Keeney and his colleagues are working with a quality assurance group from the U.K. called NEQAS, which provides blood samples containing small numbers of tumor cells to participating labs, so that researchers can test their samples against these controls. Additionally, Keeney is collaborating with the FNIH on a recent initiative to investigate standardization methods for detecting these small numbers of residual tumor cells. As a form of quality control, he has also sent many flow cytometry reference images to new labs utilizing these procedures, so that other researchers can understand the differences between normal and abnormal cell populations.

Keeney indicates that rare event detection has its limitations—both statistical and bench work expertise are required to effectively identify these cell populations. However, flow cytometry still has many advantages, he notes. “Flow cytometry is well-suited for rare event analysis in different hematological diseases. Compared to PCR or genome sequencing, it remains the least expensive method for identifying minimal residual disease, and it is a speedy process—results are obtained in the same day as running the sample,” he says.

Rare event detection has mostly been used in acute lymphoblastic leukemias, but the methodology is now being applied to adults with acute myeloid leukemias. Similar to Prof. Porwit’s studies, Keeney has also utilized 10-color flow cytometry to study blood samples from these leukemia patients, as it is much harder to clearly identify abnormal from normal cells in these patients. The testing is also being rolled out to a broader patient set, such as a chronic lymphocytic leukemia group in Europe.

According to Keeney, “We will continue to make improvements to the standardization protocols over the next few years. In the meantime, any labs performing these methods should participate in the quality assurance program.”

Along with their colleagues at Abbott Laboratories and South Western Area Pathology Service Liverpool Hospital, Keeney and London Health Sciences Centre’s Benjamin Hedley, Ph.D., are investigating standardization and validation of flow cytometric methods for counting reticulated platelets. Reticulated blood cells and platelets are useful tools for differentiating causes of anemia, but fewer laboratories utilize a flow cytometric method for reticulated platelet enumeration.

Dr. Hedley explains that a major application for counting reticulated platelets is to monitor the efficiency of certain drugs that mimic the body’s natural ability to make platelets.

“We need to be able to identify early on which patients are responding to these drugs,” he says. “This requires determining if the patient is producing more platelets, is not producing more platelets, or is producing platelets that are somehow getting destroyed by another factor. By measuring the number of immature, reticulated platelets, this tells us if the drug is helping the patient produce platelets in the bone marrow, whether or not they may be destroyed later on.”

However, the problem remained that there was no standard flow cytometric method for enumerating immature platelets newly released from the bone marrow. To tackle this issue, Dr. Hedley and his team utilized a previously published flow cytometric method from 1990 for analyzing reticulated platelets using Thiazole Orange dye, and combined this with a CD41/CD61 (platelet glycoproteins-IX and IIIa) platelet enumeration method that was standardized by the International Society for Laboratory Hematology.

“The main advantage of combining the platelet enumeration method with Thiazole Orange is that the number of platelets and percentage of immature platelets is obtained at once from the same experimental tube,” Dr. Hedley reports. The researchers used normal donor samples to establish a normal range of reticulated platelets. The samples were analyzed using Kaluza flow cytometry software from Beckman Coulter.

Dr. Hedley explains that Thiazole Orange, which is a fluorochrome that binds nucleic acids, distinguishes reticulated platelets because mature platelets have lost their RNA content. He also notes that using an eight- or 10-color flow cytometer is not necessary for counting platelets; rather, one fluorochrome to identify platelets and another to identify if they are immature or mature platelets is sufficient. He and his collaborators will be comparing their methodology using different instruments, such as Beckman Coulter and BD flow cytometers.

Dr. Hedley and his team determined that a final concentration of 10% of the initial Thiazole Orange used was sufficient for staining reticulocytes for 30 minutes, could be quenched with formaldehyde, and was stable for one hour.

“Stability is key. Clinically, we cannot expect to run our samples within minutes of being prepped. The samples are prepared all at once but need to be stable until each has been run through the flow cytometer,” he says.

Hematology counters that can screen complete blood counts may not accurately measure low numbers of platelets, and immature platelets are also larger than mature platelets and may fall outside of the set size or volume dimensions recognized by hematology analyzers. Instead, flow cytometric methods rely on antibodies and thus can still accurately count larger, immature platelets in low quantities.

Identifying Chromosomal Abnormalities

In addition to flow cytometry, hematopathological diagnostics often involve analysis of cytogenetics—the study of chromosome structure and function. Half of all patients with myelodysplastic syndromes have clonal cytogenetic abnormalities, and the International Prognostic Scoring System includes monitoring chromosome changes in bone marrow cells to assess patient prognosis.

To investigate the effectiveness of this technique, Ayesha Vawda, M.D., at the University of British Columbia and her colleagues at Vancouver General Hospital combed through patient records from the hospital’s cytogenetic department database collected between 2011 and 2012.

Characteristic features of myelodysplastic syndromes are a reduction in the number of red blood cells, white blood cells, or platelets, called cytopenia, as well as abnormalities of cellular maturation and morphological abnormalities, called dysplasia. Dr. Vawda sought to investigate the utility of cytogenetics in cases where there was a clinical suspicion of myelodysplastic syndrome because of a reduction in one, two, or all three blood cell types, but there was no definitive evidence of dysplasia based on a bone marrow biopsy.  Is performing cytogenetics of value in these cases?

The laboratory had performed cytogenetics in cases where the patient's bone marrow biopsy report warranted this additional test. Dr. Vawda and her collaborators analyzed 100 hospital records of suspected myelodysplastic syndrome cases, of which cytogenetics was successfully performed in 43 patient cases. Karyotyping identified unique banding patterns of metaphase chromosomes.

The researchers saw that abnormalities were only identified in two of the 43 karyotypic analyses: A 5q deletion in two of 75 metaphases from a unicytopenic patient with a reduction in one blood cell type, and a 9;19 translocation in all 20 metaphases in a pancytopenic patient with a reduction in all three blood cell lines.

Although patient management was ultimately not altered by these results, fluorescence in situ hybridization (FISH) had been performed to confirm the presence of the deletion in the q arm of chromosome 5 for the unicytopenic patient. A locus-specific DNA probe for 5q31 (EGR1 gene) was used, along with a control probe specific to the p arm of chromosome 5.

FISH was not performed for the confirmation of the 9:19 translocation in the pancytopenic patient, as the abnormality was clearly evident cytogenetically. Dr. Vawda explains that FISH could also be performed based on this karyotype result if the specific probes were available. As opposed to looking for loss of a fluorescent signal in the setting of a chromosomal deletion, a translocation would be identified by a fusion signal of two probes that originated from different chromosomes.

“As only two abnormal karyotypes were reported, it appears that cytogenetics may be noncontributory to patient diagnosis and management in cases of suspected myelodysplastic syndrome with no morphological dysplasia, despite the presence of cytopenia,” Dr. Vawda concludes. The team now hopes to validate these results in a larger cohort.

Dr. Vawda explains that while routine practices for cytogenetics may differ at other hospitals, Vancouver General Hospital is part of the public healthcare system, and the findings of this study will ultimately improve workload, turn-around time, and cost efficiency.

“Ideally, we would perform cytogenetics for all the patients who come through, but in a system with finite resources, we really need to decide if there is high yield or not for these procedures,” she says.

Genetic Screening via Flow Cytometry

Lisa Filipovich, M.D., studies hemophagocytic lymphohistiocytosis (HLH) and related immune disorders at Cincinnati Children’s Hospital Medical Center. HLH is commonly identified by phagocytosis of red and white blood cells, platelets, and their precursors. Dr. Filipovich runs a diagnostic lab for primary immune deficiency disorders, and she and her colleagues have developed specific assays for screening the genetic forms of HLH diagnosis, as well as to study the functional defects of these disorders.

The five subtypes of familial HLH are each associated with a specific gene, and Dr. Filipovich’s team looks for evidence of specific intracellular, cytotoxic proteins as a screen for the genetic diagnosis of a patient. For example, one subtype of HLH is caused by a deficiency in perforin, a protein that aids other cytotoxic proteins like granzyme B in entering a target cell and setting off an apoptotic cascade.

Dr. Filipovich’s diagnostic lab developed an assay to quantitate expression of intracellular perforin. This flow cytometry test for detecting perforin proved useful not only for rapid diagnosis, but also for screening potential family members to be donors for a bone marrow transplant, by separating carriers from unaffected individuals.

The researchers have since developed similar assays for two X-linked subtypes of HLH by quantifying levels of XLP1 and XLP2, respectively.

“Rather than waiting eight weeks for a genetic test, you will know in hours what the likelihood is that the patient is affected with HLH by using this flow cytometry assay,” Dr. Filipovich says. “We have gone on to develop more tests of this kind so that we can focus on what are the likely genetic defects in our patients.”

Another flow cytometry-based assay the lab has used is a degranulation assay. Natural killer (NK) cells contain cytotoxic granules within their cytoplasm as they develop, and these granules contain “killer” proteins like perforin and granzyme B. When NK cells are prepared to kill target cells, these granules are dragged to where the immunologic synapse is forming, so that the cytotoxic cell is in direct contact with the target cell. This then stimulates the release of granular components, and researchers have identified at least four proteins that are sequentially involved in opening the granules onto the external surface of the NK cell.

The degranulation assay indicates if any of these proteins are defective, by looking on the surface of NK cells for proteins that are normally found only on the inside of the granular membrane.

When the granules open, proteins like LAMP1 (CD107a), which are normally on the inner surface of granular membranes are now visible on the outside surface of the NK cells. The degranulation assay uses a monoclonal antibody to quantitate LAMP1 via flow cytometry. If there is an abnormal result, Dr. Filipovich and her colleagues will then sequence the four known genes involved in the degranulation pathway to identify which is the culprit. Therefore, this assay can point the researchers in the right direction by identifying that the patient’s HLH disorder is related to deficiencies in the degranulation pathway, and not X-linked genes or perforin.

Dr. Filipovich explained that an alternative NK cytotoxic killing assay, which assesses the ability of cytotoxic cells to lyse chromium-labeled target cells, has many pitfalls: “Our lab has shifted to primarily using the degranulation assay because flow cytometry is much more quantitative and will lead you to the likely genetic defect. Unlike cytotoxic killing assays, the degranulation assay is not dependent on the number of NK cells in the sample,” she says. “Moreover, the cytotoxic killing assay is susceptible to immunosuppressive drugs, which many patients are already receiving by the time the testing is planned.”

Aside from using flow cytometry as a screening tool, Dr. Filipovich and her collaborators are developing an HLH-specific microarray chip to identify which of the seven known genetic defects might exist in a patient. On a global scale, gene microarray studies of HLH patients have also revealed that many genes involved with key pathways of innate, B cell and cytotoxic immunity are highly downregulated.