6. FLUORESCENCE IN-SITU HYBRIDISATION (FISH)

Previous - Index - Next

6.1 GENERAL

Interphase and metaphase FISH, either as a single probe analysis, or using multiple chromosome probes, can give reliable results in different clinical situations.

It should be noted that there may be variation in probe signals both between slides (depending on age, quality, etc. of metaphase spreads) and within a slide. Where a deletion or a rearrangement is suspected, the signal on the normal chromosome is the best control of hybridisation efficiency and control probe also provides an internal control for the efficiency of the FISH procedure.

Depending on the sensitivity and specificity of the probe and on the number of cells scored, the possibility of mosaicism should be considered, and comments made where appropriate.

When hybridisation is not optimal, the test should be repeated. When a deletion or another rearrangement is suspected, the results must be confirmed with at least one other probe.

Results should preferably be followed up by karyotype analysis. This is essential when there are discrepancies between the expected laboratory findings, and the clinical referral.

Before introducing interphase FISH as a diagnostic technique, staff need appropriate training on the type of samples to be analysed. Laboratories should set standards for classification of observations and interpretation of results.

6.2 EQUIPMENT, FACILITIES AND SAFETY

A dedicated work area should be available for FISH work.

Specialised equipment should include facilities for incubation of tubes at varying temperatures, micro-centrifuge, fluorescent microscope with appropriate filters and camera or image analysis system.

Fume cupboards should be installed to protect staff where hazardous chemicals, such as formamide, are used.

Laboratories that are making their own probes should ensure their procedures prevent DNA contamination.

6.3 REAGENTS

Any new batch of labelled probes, whether generated in-house or purchased commercially, requires validation concerning its' performance before being used diagnostically. This validation requires testing for:

• Target specificity: To test if the probe hybridises to the correct location - preferably on both normal and abnormal chromosomes demonstrating the specific aberration
• Analytical sensitivity and specificity: These involve assessment of the proportion of targets demonstrating a signal (sensitivity), and proportion of signal at the target site compared with other chromosome regions (specificity). For most commercially available probes, the supplier has usually established these parameters. Sensitivity and specificity must be high to avoid misdiagnosis.

Any validation data should be fully documented for later internal audit.

6.4 TECHNIQUES

6.4.1 WHOLE CHROMOSOME PAINTING

Commercially available paints are generally used as they are reliable. Care should be taken in interpreting breakpoint positions from FISH results, and it should be done in conjunction with banding studies.

It should be noted that the resolution of chromosome painting may vary between different paints. Small rearrangements may not be detected since whole chromosome paints may not be uniformly dispersed across the full length of the target chromosome.

6.4.2 DETECTION OF MICRODELETION AND SUBTELOMERIC REGION ANALYSIS

Commercially available kits are generally used in diagnostic laboratories. The number of cells scored needs to be commensurate with the sensitivity and specificity of the probe on the slide. If microduplication is suspected, results should preferably be con- firmed by alternative methodologies (e.g., molecular analysis, densitometry).

6.4.3 INTERPHASE FISH

Extreme care needs to be taken in interpreting results. The signal in interphase cells can be variable, so large numbers of cells must be examined. For detection of minimal residual disease in neoplastic disorders a large number of cells must be analysed.

In prenatal diagnosis the presence of maternal contamination should be noted and recorded accordingly.

It should be noted that interphase FISH analysis could only detect a subset of chromosome abnormalities and may not provide a complete result or may be misleading in the absence of conventional banded cytogenetic analysis.

Interphase FISH may be an adjunct test to assess levels of mosaicism or chimerism of cell lines with abnormalities previously established by standard chromosome analysis.

6.5 ANALYSIS

It is not recommended that FISH be used routinely to confirm cytogenetically visible abnormalities although it should be used to check uncertain variants of diagnostic or prognostic significance. It may also be appropriate to check apparently classical abnormalities in the context of an atypical presentation.

6.5.1 CONSTITUTIONAL

Locus-specific probes – 5 cells should be scored to confirm or exclude an abnormality.

Multiprobe analysis – Three cells per probe should be scored to confirm a normal signal pattern. Where an abnormal pattern is detected, confirmation is advisable.

Prenatal interphase screening for aneuploidy
-Signals should be counted in at least 30 cells for each probe set.
Interphase screening for mosaicism
- A minimum of 100 cells should be scored.

6.5.2 HEAMATOLOGY AND ONCOLOGY

The standard number of cells recommended for a FISH study is 100. However, it is recognised that an adequate positive result can often be obtained with fewer (particularly when expecting an all-or-nothing result, e.g. in CGL), and a suspicious finding may need more.

In all diagnostic FISH studies, a positive effort should be made to examine a few metaphase cells, if present, and not depend entirely on interphase nuclei. In normal metaphases this confirms that the correct probes were used and abnormal metaphases can be invaluable in interpreting unusual signal patterns.

Laboratories should be aware of the different types of FISH probes and their normal signal pattern e.g. break-apart probes and fusion probes. The limitations of the probe set should be documented, particularly if the analysis is done solely on interphase cells.

The use of FISH on paraffin wax sections in particular is an appropriate way to investigate specific rearrangements or gene amplification and has the advantage that tumour tissue can be directly screened. Analysis can be limited to ten cells scored by analyser and checker each, when an abnormality is detected but extended to 100 cells each before exclusion of the abnormality is suggested.

6.6 CHECKING

Preferably interphase FISH results should be independently scored by an appropriately trained person each who should examine 30-70% of the total of cells used for the analysis. If the two primary scores differ significantly then a third person (if necessary from another laboratory) should be called in to provide a resolution. This person should normally be informed of the previous scores.

For metaphase FISH the same process should be used as for checking conventional banding.

Previous - Index - Next

Last changed: 2008-02-04