The Evolution of Clinical Flow Cytometry

Grant Howes is marketing director for flow cytometry at Sysmex America, Inc. He is responsible for product development, commercialization, and support. Prior to Sysmex America, Grant worked for companies including Beckman Coulter, Accuri Cytometers, and Becton Dickinson. He began his career in the United Kingdom obtaining a Fellowship of the Institute for Medical Laboratory Sciences, specializing in hematology and hemostasis while at St. Thomas’ Hospital in London.

How have the main clinical applications of flow cytometry changed over time?

Twenty-five years ago, the most common reason people would buy a flow cytometer for a clinical lab was to monitor patients with HIV and AIDS with CD4 testing. However, the most common application today is immunophenotyping of immunological malignancies—leukemia and lymphoma. So, there’s been a changed emphasis of why people are testing in the laboratory. 

What challenges do clinical flow cytometry labs currently face?

One of the biggest challenges today is getting well-trained technologists that are able to run the flow cytometer and interpret data. It’s quite a large investment, not just financial, but also in terms of training people if you don’t already have specialists in the lab to do flow cytometry. Well-trained technologists have been more difficult to find because there aren’t the set of teaching establishments there used to be for medical technology, and it’s still quite a small community. 

One of the bottlenecks to efficient workflow in flow cytometry has always been sample preparation, especially when one is looking at leukemia and lymphoma. When everything was done manually, it required technologists to sit at a bench for hours prepping samples.

Another challenge for laboratories doing leukemia and lymphoma phenotyping is the fact that flow cytometers are capable of analyzing a lot of colors at once—rather than doing 4, 5, or 6 colors, clinical laboratories want to do 8, 9, or 10 colors, or more. To prepare those samples by hand is a time-consuming process and it’s easy to make mistakes. When one is dealing with 10 reagents in one tube, it becomes expensive to make a mistake.

What technological advances have simplified the use of flow cytometers for clinical labs?

The biggest advance has been miniaturization of lasers and of the electronic boards on flow cytometers, which has made these systems smaller. Another major advance in this same time period has been the greater immunological knowledge gained by developing so many antibodies.

More recently, the biggest advance for us has been eliminating the bottlenecks of sample preparation. There are sample preparation devices that can create samples for CD4 testing, but there hasn’t been a system that’s flexible enough to allow customers to develop the sample preparation components of their assays. That’s where our PS-10 has made a big difference for labs; when bench technologists see it, they’re stunned at how much less work they have to do. 

The automation saves time and allows laboratories to keep their same staff versus having to take on new staff; so rather than automating to eliminate staff, were automating to allow the current lab staff to stay in their roles but do more tasks that they’re qualified for. In many cases, the automation also allows the lab to increase the number of tests they can do.

On top of the ability to prepare samples automatically, we can now apply automation to make up cocktails of antibodies. It saves time and also adds confidence that these cocktails will be created correctly. It’s that front end automation of sample preparation and making those reagents for sample preparation that’s really making a difference now. 

On the flow cytometric side, one of the biggest advances has been that the use of software is so much easier now. 

What’s on the horizon for clinical flow cytometry?

While we have automated methods for analysis for the single applications like CD4, we really don’t have that for leukemia and lymphoma phenotyping, so all of the analysis is done by hand. In the future, I think AI and machine learning will come into play to assist in the automatic analysis of data. We’re not there yet, but it’s going to be the next big thing.