by ecancer reporter Janet Fricker

The technique of immunohistochemistry (IHC) exploits the exquisite ability of antibodies to bind specifically to their molecular targets (antigens) in detecting and imaging the location of antigens in cells.

The antibodies used are tagged either to an enzyme that catalyses a colour change reaction or to a fluorophore in order to generate a coloured signal at the site of the antibody-antigen complex within the cell.

Since its discovery over half a century ago it has become an invaluable technique in basic medical research and in the clinic.

In oncology, it is often used in diagnosis to detect and characterise cancer cells by the presence of protein antigens that are specific either to these cells in general or to a particular tumour type.

One important disadvantage of immunohistochemistry as a diagnostic tool is that it is extremely difficult to use to detect multiple antigens simultaneously.

It is theoretically possible to detect up to four antigens simultaneously, but difficulties in sample preparation and imaging generally limit this to two.

A group of researchers led by Garry Nolan of Stanford University, California, USA has now developed a novel immunohistochemistry method that theoretically, at least, increases the number of antigens that can be detected simultaneously from four to 100.

This technique, known as multiplexed ion beam imaging or MIBI, involves staining cells with antibodies that have been bound to pure isotopes of metals and using time-of-flight mass spectrometry to detect and identify the antibodies by the molecular masses of these metals.

The metals used are stable isotopes of the lanthanides (rare earth metals), which are all biologically inert.

In MIBI, tissue samples such as the standard formalin-fixed, paraffin-embedded (FFPE) sections that are often used for diagnosis are first stained with antibodies that have been conjugated to these metal isotope reporters, dried and loaded into the detector in a vacuum.

The sample surface is sprayed with a beam of oxygen ions which liberates the antibody-metal adducts as secondary ions.

Up to seven different such adducts can be detected and identified simultaneously using a mass spectrometer, and that spectrometer can be re-calibrated in order to scan the same sample many times and detect the complete range of antibodies.

The results can be combined into two-dimensional coloured images that locate each lanthanide-metal bound antibody and thus each protein antigen on the original specimen.

Nolan and his co-workers first validated their technique by staining peripheral blood mononuclear cells with lanthanide-conjuated antibodies to seven cell surface proteins that are commonly used as immunological markers.

They analysed the distribution of these proteins on the blood cells using both MIBI and mass cytometry, and showed that the techniques yielded comparable patterns of protein expression.

MIBI was shown to be a particularly sensitive technique and to detect and measure protein intensities over a dynamic range spanning five orders of magnitude.

In more clinically relevant trials, the researchers used MIBI to detect tumour-related proteins in breast cancer specimens.

Serial sections were obtained from a single FFPE block of human breast tumour tissue and stained with either metal-conjugated or conventional antibodies to either the proliferation marker protein Ki-67 or oestrogen receptor alpha.

The signals obtained from both techniques were of comparable intensity, and background intensity levels were also comparable, indicating that metal conjugation made no noticeable difference to antibody-antigen binding.

Finally, the researchers assessed the ability of their technique to detect and differentiate between many antigens simultaneously in a clinically appropriate setting.

They stained FFPE sections obtained from different breast tumours with metal-conjugated antibodies to double-stranded DNA and to eight proteins associated with one or more breast tumour types.

An MIBI analysis of these samples was carried out using two scans of the mass spectrometer and images of the intensity pattern for each marker antigen obtained.

Pseudo-fluorescence images mimicking the conventional three-colour immuno-fluorescence images were generated by allocating each of the red, green and blue channels to a different signal.

The quantitative intensity of each signal was also determined.

The qualitative and quantitative results for three of the nine markers – oestrogen receptor alpha, progesterone receptor and HER2 – were validated using conventional immuno-histochemistry and found to be both precise and accurate.

These results suggest that this complex and sensitive technique might have important applications in oncology, including differential diagnosis, monitoring response to therapy and improving our understanding of the molecular basis of disease.

Reference: Angelo, M., Bendall, S.C., Finck, R. and 9 others (2014). Multiplexed ion beam imaging of human breast tumors. Nature Medicine, published online ahead of print 2 March 2014. doi:10.1038/nm.3488