Tuesday, October 17, 2006


ZAP-70 is a molecule used to transmit a signal from the T-cell receptor to downstream pathways. Most B cells lack ZAP-70 and instead use a related tyrosine kinase, Syk, instead. In most cases of CLL with mutated IgVH genes, stimulation of the B cell receptor fails to cause a signal to get through, while most cases with unmutated IgVH genes signal satisfactorily. The reason for this is unknown, but may have something to do with abnormalities of the accessory signaling molecules in CLL. Somehow the presence of ZAP-70 overcomes this defect. B-cell mediated activation of ZAP-70 is negligible when compared to activation of Syk. It seems likely that the effect of ZAP-70 is to inhibit the decay of downstream factors which normally terminates the signaling response

ZAP-70 appears to be a client protein of the molecular chaperone, heat shock protein 90 (Hsp90). Whereas in normal tissue Hsp90 is in a latent, uncomplexed state, in advanced cancers it exists in an activated form complexed with other molecular chaperones and catalyzes the conformational maturation of a large range of signalling proteins. The geldanamycin derivatives, 17-AAG and 17-DAG, inhibit Hsp-90. It has been shown that ZAP-70 is an Hsp90 client protein in CLL cells with unmutated IgVH genes, but not in T cells. The inhibitors blocked BCR signalling in ZAP-70 positive cells, by causing the degradation of ZAP-70 and killing the CLL cells.

The group at NIH and their extra-mural collaborators, found that ZAP-70 expression correctly predicted the IgV mutation status in 93% of cases. Four patients with mutated IgV genes had high ZAP-70 expression and three with unmutated IgV genes had low ZAP-70 expression. Both ZAP-70 expression and IgV mutational status were equally able to predict time to requirement of treatment.. In the same paper they reported on two simplified assays, Western blotting and immunohistochemistry, that they thought might be suitable for introduction into routine laboratories. However, most laboratories regarded these techniques as too cumbersome, and require a flow cytometric assay. These have proved difficult, especially as ZAP-70 is an intracellular antigen so that the cells require permeabilization, but three different methods have been reported. The first two both used the Upstate antibody 2F3.2 in an indirect assay, differing mainly on where to set the zero, one using an isotype control and the other using the lower limit of the patient’s own T cells. Not surprisingly, the two assays give different normal ranges, the first up to 10% positive cells, the second up to 20%. Several laboratories had problems setting up these assays, probably because the antibody does not appear to be stable when stored at 4oC. The third assay used a directly labeled antibody. Directly labeled 2F3.2 seems to give too many false positives, and this group therefore used a new antibody, 1E7.2, labeled with a new fluorochrome ALEXA 488.

Whereas, the first two assays produce a concordance with IgVH gene mutations of over 90%, the concordance of the third, and easier to perform, assay was only 77%. This third paper also suggested that ZAP-70 expression added prognostic information that was independent of the effect of IgVH gene mutations. Far from being able to abandon costly and labour-intensive gene sequencing, the CLL community is faced with having to do both assays. There is clearly some discrepancy between IgVH gene mutations and ZAP-70 expression, but a lot of the discrepancy is down to technical problems.

In order to resolve the conflicting findings of the various flow cytometry assays for ZAP-70 there has been much interaction between academic laboratories, commercial companies and quality assurance laboratories. Following a satellite ZAP-70 Workshop at the 20th annual meeting of the Clinical Cytometry Society in 2005 a special issue of Cytometry Part B (Clinical Cytometry) was commissioned to deal with these issues and this has just been published.

Although the best method is still disputed there is a measure of agreement. The use of commercial fixation and permeabilization kits is satisfactory although these might yet be improved upon, but despite this ZAP-70 expression changes on storage. The type of anticoagulant affects this. Over the course of 48 hours ZAP-70 levels of cells held in preservative-free heparin fell by 13%, while that of those held in EDTA rose by 3%. However, for samples to be analysed in the first 24 hours after drawing the blood, heparin gave the less variation.

Given that brightness of fluorescence is greater using indirect methods, there is still general agreement that the reagent should be directly conjugated. All directly conjugated antibodies show a degree of non-specific binding but this is greatest when antibodies conjugated with phycoerythrin. Most agree that 1E7.2, conjugated to ALEXA 488 performs well. Setting the zero remains a problem. Many laboratories continue to prefer the left hand edge of the T and NK cells or the right hand edge of residual CD5 negative B cells, others are convinced as I am that an isotope control remains the answer. Rather than report the percentage of positive cells there is strong agreement that some measurement of the degree of positivity such as mean fluorescent intensity. However, because of intralaboratory variation this needs to be calibrated either by the ratio of geometric means of test and control populations, or as molecules equivalent fluorescence against calibrating beads.

It is not completely clear that ZAP-70 is stable over time. Levels are higher in cells from bone marrow rather than peripheral blood and it is known that it can be upregulated in response to T cell stimulation. Although most workers believe it to be stable, and we have one patient in whom the CD38 expression increased from 3% to 76% over time while showing consistent ZAP-70 levels, but one study showed changes (usually from positive to negative) in 9 out of 111 patients with at least two estimates separated by a median of 29 months.

The bottom line is that ZAP-70 is still not a reliable assay on which to base treatment decisions. IgVH gene mutations remain the gold standard.


Gabe Green said...

Hello Terry,

I didn't know how to e-mail you, but I need to consult someone such as yourself. My father has just been diagnosed with CLL, and although presumably early-stage, I would like to know some things.

Could you possibly e-mail me at osiventures@yahoo.com?

Thank you in advance,


Anonymous said...

I don't think anyone is going to base treatment decisions anytime soon on the presence or absence of ZAP-70, mutational status, CD38 or other 'non-traditional' test.

The tried and true standards for deciding when to treat remain the 'gold standard'. The doubling of absolute lymphocytes in less than one year or six months, when established as a trend, is certainly a strong indication for treatment. A high Beta-2-microglobulin level may be too.

The over-all condition of the patient must be factored in; some patients report debilitating fatigue which may relent after treatment. Marrow failure, which you have described, it certainly a clarion call for treatment. Frequent infections are another sign.

It's way too early for focussing on ZAP-70 and mutational status as determinations for beginning treatment.

We have no curative treatment for CLL, and it has never been demonstrated that early treatment gives any survival advantage over delaying treatment.

I think this is getting ahead of the game.

Terry Hamblin said...

I agree, though I am even more sure that treatment should not be started on the basis of beta 2 microglobulin.

Frequent infections are certainly not an indication either, unless their causeis neutropenia. In general immune deficiency is made worse by treatment, not better.