How can measurement of interferons be improved?

We utilize a flow cytometric assay to measure markers associated with TI and TII IFN signatures [i.e. SIGLEC1 (CD169) and PD-L1 (CD274), respectively] on monocytes. The advantage of this approach is that it can be done in a small amount of whole blood (or peripheral blood mononuclear cells, if needed) collected in a regular blood collection (EDTA) tube, which does not require special processing, and results can be provided relatively rapidly. If there is an elevated TI IFN signature, it will be reflected in the increased expression of SIGLEC-1 (CD169) on monocytes. The expression of CD274 (PD-L1) on monocytes is more nuanced, as it is only elevated by a high and sustained production of IFN-γ and phosphorylated STAT1 signalling, as seen, for example, in untreated HLH. Therefore, it does not always correlate with other IFN-γ markers, such as CXCL9 because these markers are regulated differently and have different thresholds.

Further points about our flow cytometry assay

Many physicians are familiar with the ISG score, where quantitative PCR is used to measure the mRNA levels of a panel of gene transcripts where the expression levels are known to be stimulated by TI-interferons. These physicians have now discovered our clinical flow cytometry assay where they can obtain rapid results without the need for special tubes for the blood draw. Hence, it is now one of our highest volume tests, so clearly there is a need for it. The test is requested by physicians from many different subspecialities, which if I had to rank them in order of test requests would be: neurology, rheumatology, hematology, immunology and dermatology, for example. However, we sometimes still need to provide some further education about the flow cytometry assay, on request.

Other analytical and clinical issues to consider

With the ISG score, different numbers of genes can be used in the test panel. For example, the assay could involve four, five, six or even eight genes. Several assays utilize a 4-gene panel but the expression of some of those genes is also stimulated by type III (TIII) IFN. For example, if we utilize peripheral blood, normally we don’t have cells expressing receptors for TIII IFN, like interferon lambda. But, if the patient has lung disease, neutrophils can be activated that can come out of the lung tissue and into circulation that express those receptors for TIII IFN. Therefore, the ISG score no longer represents a purely TI IFN signature, but, based on the clinical context, the ISG score can be influenced by other interferons. With the assay for SIGLEC-1, it represents only TI IFN levels, with no interference from other IFN types.

Another area that often raises questions from physicians is with regard to the PD-L1 (CD274) assay that we use for TII IFN, because we also offer CXCL9 testing in serum. CXCL9 and 10 are surrogates for IFN-γ, as levels of IFN-γ in circulation are low. So, people might order the flow cytometry TI and TII IFN test and can be confused if they see normal TII IFN levels but elevated CXCL9. This can happen for a few reasons but the key point is that these are two different markers that are differently regulated, controlled and expressed, so there is not a linear correlation between the two results. Understanding the foundational biological processes is equally important in interpreting results in the context of the underlying disease.

Treatment and therapeutic drug monitoring

Another crucial point is that levels of PD-L1 (CD274) are very sensitive to treatment. It is possible for HLH patients with a notable PD-L1 signature to normalize the expression of this marker once they are treated (whether by IFN-γ neutralizing antibody, steroids or some other form of immunosuppression). Because of this, the PD-L1 assay would not indicate elevated levels of TII IFN in a treated patient.

For TI IFN, depending on the underlying disease, the signature can be very persistent. For example, we frequently do serial monitoring to follow response to therapy and if treatment is inadequate, then the signature may remain unchanged for a while, and this may prompt a physician to alter the dosing of the treatment or add another therapy to effectively control the TI IFN response. With the flow cytometry assay, for the most part, the sample is analysed on the same day; all normal results are released by the lab staff, and abnormal results, which require interpretation are released by the next day at the latest, so the turnaround time is 36–48 hours, which is ideal for monitoring treatment and adjusting the treatment regimen.

Result reporting and quantitation

The final point to make is how the results are reported. We report the frequency (percent positive) of monocytes that express the particular marker. However, we also report the amount of the marker that is expressed, which we assess by median fluorescence intensity (MFI). In addition, to control for instrument-to-instrument variation, we report the ratio of the MFI for the specific antibody to the MFI of the isotype control for that patient sample (i.e. MFI ratio).

In flow cytometry, there are different ways of reporting the results, as mentioned above: the percent positive is a relative proportion; the MFI is a surrogate for the amount of protein present. However, unlike chemistry tests and so on where we have a standard curve and we can provide an exact quantitation, for these flow cytometry tests we are using the term a little more loosely, it’s relative quantitation (or semi-quantitative at best).

It is possible to provide accurate quantitation in flow cytometry using fluorescent beads with defined amounts defined amounts of fluorochrome. This allows the creation of a standard curve, from which you can obtain the number of molecules per cell [molecular equivalent of soluble fluorochrome (MESF)], or it’s sometimes called antibodies per cell (ABC). This method would then allow us to report exactly how many molecules of SIGLEC-1 are on the surface of the monocyte. It’s a good method but, as you can imagine, is a little more involved from the clinical lab standpoint.

We have to assess our assay in the context of the assay design where we are analysing a biological change, a cellular phenomenon, where as a result of increased TI IFN, cells now express a marker that they don’t normally express. We do not need to report antigen density in our system, even though it may be needed in the context of other assays. Since we are measuring the expression of a marker that is not normally present, MFI is a good way to express the results as long as one can control for reagent and instrument variability appropriately.

Very few laboratory assays are perfect in quantitation or assessment, and the important point to remember is that no assay is independently diagnostic, especially for complex disease states, and data is always correlative, to the clinical context, other relevant laboratory findings and with serial monitoring.

What are your thoughts on future perspectives of this field?

There should be a distinction between what is achievable in a research setting versus what is amenable to the clinical diagnostic laboratory. If pragmatism was not a limitation, a combination of transcriptomic and proteomic approaches, which would also permit superimposition of cellular profiling would be ideal. Additionally, it must be done on an accessible sample, a small volume and analysis has to be relatively rapid to be clinically useful and meaningful.