When clinicians need diagnostic data across fields as diverse as hematology, transplantation, and tumor immunology, among others, flow cytometry is traditionally the tool they choose to identify cell markers. Investigators use flow cytometry for purposes that include analyzing cells, detecting biomarkers, and protein engineering. The technique has been essential to basic bench researchers and clinicians alike.
The methods and equipment used to probe cellular questions are rapidly advancing—including, at Yale, through the addition in 2014 of CyTOF, or Cytometry Time-Of-Flight, and this past June of the CyTOF Imaging Mass Cytometer (IMC). The latter acquisition distinguishes Yale as one of the first academic medical centers in this country to host a CyTOF IMC. Both instruments greatly expand upon the types of samples and numbers of components that can be analyzed—30 to 40 different biomarkers—where flow cytometers are more limited and can generally detect only eight. While on its own the CyTOF analyzes cells in suspension, the IMC, by contrast, can produce labeled images of tissues with spatial accuracy and provide position details of cells in an intact tissue.
The process of converting a tissue sample to a labeled, spatially accurate, microscopic image begins with a thin cross section of tissue mounted on a glass slide. Researchers add antibodies to the tissue, each antibody tagged with a different heavy metal that lets it be tracked as it binds with a specific protein. The whole assemblage goes into the IMC, where a laser vaporizes the sample from its slide, hurtling the heavy metal markers into the CyTOF instrument. Each marker, based on its mass, will have a different “time of flight” to travel through the machine. The instrument translates the times-of-flight into signals that are mapped back to where they were ablated by the laser—each one corresponding to the protein tagged with the metal—while also measuring the abundance of each protein of interest at a specific tissue location.
The IMC “means that you can detect much more in the same sample, and you can detect it with more certainty,” says Ruth R. Montgomery, Ph.D., associate professor of medicine (rheumatology), associate dean for scientific affairs, and director of Yale’s CyTOF facility. “Things we couldn’t have dreamed of measuring—now you can get them in an afternoon.”
Examples include identifying tissues that respond favorably to medication and comparing them to those that don’t, or classifying immune cell responses to illness. Montgomery, for example, has used the CyTOF to compare the makeup of cells from individuals who were unperturbed by infection with West Nile virus with those from patients who developed severe symptoms of the disease.
The nascent applications for IMC are extensive, particularly in the field of pathology. “I haven’t seen a disease with bodily fluids and tissues where CyTOF or [IMC] wouldn’t be helpful,” Montgomery says. Her team has already begun taking advantage of the instrument’s vast potential. “We’ve started by demonstrating the power of the instrument and how to use it in a reproducible, validated, quantitative manner,” Montgomery says. “And our pathology guys do that as well or better than anybody.”
Montgomery is referring to Kurt Schalper, M.D., Ph.D., assistant professor of pathology and director of the translational immune-oncology laboratory, and David L. Rimm, M.D., Ph.D., professor of pathology and of medicine (medical oncology), director of pathology tissue services, and director of translational pathology. Both have used the IMC to advance their cancer research: Schalper primarily investigates tumors in the lung, and Rimm focuses on breast cancer.
Tumor heterogeneity (the metabolic, metastatic, and morphological nuances distinguishing one patient’s tumor from another’s, and even cells within a single tumor) makes positional information a prized research commodity that traditional flow cytometry cannot match. “Having that spatial context [from IMC] turns out to be a very important point,” Schalper says. It means researchers can study not only the components of a specimen, but also their interactions. “It’s really a way of amplifying—massively—the amount of information we can obtain from samples,” says Schalper.
Much as Montgomery investigates varying responses to the West Nile virus, Schalper uses IMC to interrogate what distinguishes immune cells that recognize and attack tumor cells from those that fail to sense the threat. Rimm uses the same technology to study the range of responses breast tumors exhibit to treatment with an anti-cancer drug.
“We’ve got a drug; now we can ask a question,” Rimm says. “Can we look at relationship of expression and predict a response to therapy? That’s something that will be really valuable to patients. You can see how this gets from a very technical, highly scientific question, all the way to patient care.” Rimm and Schalper do not see the IMC arriving in clinics anytime soon. For now, the instrument is too specialized and expensive for that setting. In its academic realm, however, IMC is used to track tens of variables in tissues—identifying which ones differentiate cancer-ignoring from cancer-fighting ones, for example—and then apply that knowledge to more readily available methods. Once researchers know which variable to look for—thanks to the IMC—the more limited capabilities of standard techniques no longer pose the same restrictions.
“We use all the tools we can to get at the puzzles we’re looking for,” Montgomery says. This includes collaborating with Yale’s Department of Mathematics and other sources of quantitative expertise, to consolidate the sheer volume of data that IMC produces. The digital image of a single patient’s tissue sample in Rimm’s lab contains 400,000 to 600,000 pixels, and the IMC measures 30 to 40 selected proteins within each pixel for more than 60 patients. That adds up to 120 million data points for one small pilot study, so Rimm has recruited Yuval Kluger, Ph.D., associate professor of pathology, a bioinformatics expert affiliated with Yale’s Applied Mathematics Program, to help analyze the data. Collaboration among disciplines and departments is “a requirement, at this point,” says Montgomery. “We have the technology to create giant datasets. And, to make sense of them—to interpret them for relevance to our health questions—we need to have computational colleagues.”
Once investigators learn its capabilities, Montgomery sees CyTOF IMC moving far beyond its initial uses, and becoming an essential tool for labs across the School of Medicine. “Now that we can really make use of it, we can advance into many different areas,” she says. “A lot of people come in here to see me with their great ideas about what they want to do, and I tell them to go ahead and use the machine.”