Which method is best used in identifying tumours

In this article we attempt to compare the two systems of EnVision + and ChemMate EnVision – both systems of immunohistochemical assessment which are commercially available (DAKO Germany). Because of the paucity of published literature which compares the two systems directly, we have had to rely on what there is and also look for corroboration from circumstantial evidence gleaned from other papers which have compared either system against common others.

One area where there has been a lot of work is in the field of intra-operative frozen section assessment and we have looked at the performance of the EnVision + system in this specific field. We also look at a number of papers that provide an overview of the field of immunohistochemical systems in analysis, in order to glean what comparative information there may be.

Introduction

The object of this piece is to assess the functional differences of the two similar staining systems ChemMate and EnVision. The field of immunohistochemical analysis has mushroomed in the recent past and a number of agents have been produced commercially. The original work in the field was done by Coons and his colleagues (in 1941) when they discovered a direct fluorescence method of demonstrating tissue antigens. Since then the procedures have improved in sophistication and ease of use, through the horseradish peroxidase, peroxidase-anti-peroxidase (PAP) the avidin-biotin complex (ABC) and the alkaline phosphatase-anti alkaline phosphatase (APAAP) systems all typically taking 2-4 hrs to obtain a result.

Because the field of immunohistochemical investigation is now considered vital to the field of tumour research and also in the day to day diagnosis and management , particularly in the field of the poorly differentiated adenocarcinoma, it would be useful to have a system that could be administered comparatively quickly - certainly in the field of intra-operative diagnosis. Some of the more modern methods could be done significantly faster particularly those that required directly labelled primary antibodies. Faster methods were developed using a microwave irradiation technique but only four antibodies could be used in this technique. The equipment was bulky and cumbersome and certainly not as convenient as either of the two systems that we have under consideration here.

Its not always easy to evaluate the relative merits and drawbacks of each system – particularly in the clinical field where in one circumstance specificity may be an overriding issue, and in another – such as frozen sections – speed may be important. It is therefore vital to compare different types of immunohistochemical agent in relation to the specific type of project that is being undertaken. In this particular piece we are considering their use in the identification of certain tumours such as adenocarcinoma of the small intestine – a field where frozen sections are rarely performed, so criteria such as speed of results are less important than if we were looking specifically at an area such as breast lumps. The field where the systems such as ChemMate EnVision and EnVision + are particularly useful is the field of micrometastases where intra-operative diagnosis will often influence the course of the operation.

Speed is obviously an overriding factor when, in the clinical circumstance, the surgeon has a patient literally waiting on the operating table, but speed is also an issue when it comes to costing the various procedures. The manpower resources necessary for the various procedures are expensive commodities and clearly will influence considerations of cost. It has been a matter of some surprise that, in all the papers that we have read on the subject of comparison only one directly confronts the issues of time and money as being part of the selection criteria .

We clearly accept that if a system is neither reliable nor gives reproducible and accurate results, then it really doesn’t matter how cheap it is, it is of no use in a clinical environment. But when many systems are excellent and differ from each other generally in minor differences of expression of the results then the cost – and therefore the time spent on each procedure - does become a significant factor in the overall comparison.

EnVision +

EnVision + is a newly introduced and highly sensitive two-step immunohistochemical technique. The basic complex is a large number of secondary antibodies combined with a horseradish peroxidase which is coupled to a dextran backbone. The application of the primary antibody is followed by the application of a polymeric conjugate which comprises a large number of secondary antibodies (typically goat anti-mouse or goat anti-rabbit) which, as is quoted above is the portion bound directly to the dextran backbone. One typical conjugate would contain up to about 100 horseradish peroxidase molecules and up to about fifteen antibodies

It is based on a dual link / HPP kit (mouse and Rabbit) and from a technical point of view it can be used in conjunction with both monoclonal and polyclonal antibodies. It has the advantage that it is free of both avidin and biotin and is therefore ideal for tissues that are rich in endogenous biotin. It is therefore particularly useful in systems that employ optical screening systems

EnVision + was introduced into the UK in 1997 and the ChemMate EnVision system was introduced later in 2001. The EnVision + system is a two step system and therefore can save a considerable amount of time by comparison with other systems.

ChemMate

ChemMate (ChemMate EnVision system) was introduced into the UK in 2001 is also based on labelled polymer technology but it differs from the EnVision + in that it contains antibodies that recognise both rabbit and mouse immunoglobulins

Comparisons

There have been few studies that directly compare both EnVision + and ChemMate EnVision. One of the most objective is probably the one by Dodson (2005). It looked primarily at the specificity of each preparation and compared it to the also commercially available product – Duet. All of the products are made by the DAKO company in Denmark.

In all comparisons in this study, both ChemMate EnVision and EnVision + performed almost identically and both were marginally superior to Duet. The statistical analysis of all three on an uncorrected basis showed a Student’s T test value of p=0.001. When the systems results were corrected for optical aberration in the reading system there was a marked improvement seen in both ChemMate EnVision and EnVision + over the Duet system. Given the fact that many labs now employ optical screening systems at some stage in the reporting process , even if only for a preliminary sift, then this difference may well be considered significant.

It appears that the ChemMate EnVision and EnVision + systems were found to be superior with antibodies that required high Temperature Antigen retrieval (HTAR) and that the differences were due to the demonstration of retrieved endogenous biotin by the avidin-biotin-based system. This seems to be a strong point in favour or both systems (ChemMate EnVision and EnVision + ) as it is frequently cited as a relevant finding in comparative papers.

The authors comment on the fact that retrieved biotin will stain intensely in avidin-biotin-based systems and it typically presents as a collection of localised granular deposits. Generally it does not cause a diagnostic problem for the experienced observer as careful examination of the negative controls will generally signal its presence. It is seen in this study that the ChemMate EnVision and EnVision + systems have an advantage over the avidin-biotin-binding system (Duet) when either significant amounts of endogenous biotin are found or automatic optical image analysis systems are used, which seems to be more and more often the case.

In another paper Ruck (2005)considered the uses of these systems in the fields of frozen sections. Such procedures are not common in the field of intestinal tumours although it is occasionally used to examine peripheral lymphnodes during an operative resection and both systems are useful in the detection of micrometastases.

Micrometastases are of great importance clinically as there is a marked correlation between survival rates and the spread (even micro-spread) of the malignant process. Clearly, the involvement of peripheral nodes is a poor prognostic sign in these circumstances and surgical decisions would have to be made depending on the involvement of nodes or otherwise. Speed is clearly important in such circumstances.

Ruck outlines some of the benefits of using the ChemMate EnVision system in these intra-operative cases. The main points that he makes are that, as a general rule the histopathologist has a window in the region of about ten minutes to produce an accurate diagnosis. The usual procedure is to use a Haematoxylin and eosin (H&E) stain. This is certainly quick and efficient and good enough for many cases where the basic diagnosis is in doubt. H&E does have significant shortcomings, particularly when the lesion is either very undifferentiated or very small (viz. micrometastases). If this is thought to be a significant possibility then immunohistochemical staining methods should be employed and these are typically time consuming - generally in excess of 20 mins. Certainly too long to expect the surgeon to wait during the operation with the patient on the table.

Most of the commercially available options are expensive and available for a limited range of antigens. Ruck observes that the ChemMate EnVision two stage system can produce a highly accurate answer within 13 minutes. This appears to be a fairly consistent finding between the various papers that have reviewed this area (see on)

It should be noted that Ruck recommends the use of antibody diluent with background reducing components for the dilution of the primary antibody. If used in conjunction with DAB+ OR ACE+ the sensitivity of the system is increased two or threefold. He also suggests that for rapid results the primary antibodies should be used in concentrations of 4-10 times normal (when compared to the standard staining procedures). He observes that this will result in slightly increased costs, but if a second operation is avoided, then clearly that it a considerable cost saving.

With regard to the antibodies that are useful in intestinal tumours smooth muscle actin and Synaptophysin are particularly relevant as antigens. Both are said to be easily evaluated without a problem.

In the clinical field this method is certainly more accurate than H&E, and is certainly more useful at detecting micrometastases in sentinel nodes and can help to distinguish between large cell lymphoma and carcinoma.

Kämmerera (2001) carried out a qualitative test on the same topic as Ruck to see if it was practical to use the EnVision + for the rapid intra-operative staining of frozen sections. They tested 45 primary antibodies on frozen sections from a number of different tissues which were incubated with EnVision + . Specific staining was seen with 38 of the 45 samples “including HMB-45 and antibodies against keratin, vimentin, leukocyte common antigen, smooth muscle actin, synaptophysin, CD34, CD3, CD20, and prostate-specific antigen” (Author’s quote)

The quality of the staining was graded from 1-3 by three independent assessors who looked for “specific, evaluable and reproducible immunostaining” The authors coped with inconsistent results by accepting the mean as the value to be recorded. The authors note that with certain antibodies (against CD14, CD45, CD56, cathepsin D, epithelial membrane antigen, and smooth muscle actin) the EnVision + system produced less background staining and therefore a better “signal to noise ratio” than either of the two other procedures that were run in parallel ( HRP and LSAB procedures). This is because of the ability of the EnVision + system to spare endogenous biotin from staining which the other two systems don’t.

For some of the receptors which failed to stain with the rapid protocol (monoclonal anti-CD3 and c-erbB2 antibodies), other antibodies were available which did produce useful staining results. The authors commented that the polyclonal antibody against CD3

Actually needed a longed incubation period of up to 5 mins to allow for adequate staining. EnVision + and the standard immunohistochemical procedures all failed to stain the S100 protein

As we have commented, the EnVision + system proved itself capable of detecting a large range of antigens in frozen section very quickly. This certainly improves the amount of information available from the standard H&E preparations. The authors also make a very valid observation that a limitation of this procedure is the number of slides that can reasonably be handled. They found that two slides each carrying two sections would allow the detection of up to four different markers which they considered to be adequate for an intra-operative assessment in the majority of cases. They observe that it (EnVision + ) has the advantages that it is timesaving and simple to use and requires no specialist equipment such as microwave generators in other methods.

Within the limitations described EnVision + has shown itself to be highly sensitive in the staining of paraffin sections and picked up even weakly expressed antigens. The phenomenon of false positives with endogenous biotin is not seen with this system. As we have commented elsewhere the sensitivity could be increased with the use of DAB+ or AEC+ substrate solution the authors also suggest that such modifications may also be helpful in the full length immunohistochemical procedures.

EnVision + compared favourably, for most antibodies, with the more laborious and time consuming LSAB method. The S100 protein that was mentioned earlier as not staining well in the frozen sections, did stain well in the paraffin preparations. The two step procedure of EnVision + has a sensitivity advantage over the one step systems and can be used with any suitable primary antibody. In common with Ruck’s findings, the authors commented upon the cost implications of the higher concentrations required but this is more than outweighed by the economies in reduction in operating time.

A more specific comparison was performed by Sabattini et al. who compared the EnVision + system with the ChemMate EnVision and also the APAAP, CSA, LABC and SABC techniques. Their method was to use a large series of differently processed normal and pathological samples were used (the paper does not say exactly how many) and 53 relevant monoclonal antibodies were chosen. The various methods were subjected to a “chessboard analysis” so that the EnVision + results could be compared directly to each of the other systems. The authors note that the samples were either hand stained or put in a TechMate 500 immunostainer.

The results are reported as showing that the EnVision + allowed a two to fivefold higher dilution than the other systems (including ChemMate EnVision ) with the staining intensity and percentage of expected positive cells being the same. They found that with some of the critical antibodies (viz anti-CD5), the EnVision + proved to be the superior method as it achieved consistently reproducible results with samples that were over-fixed or differently fixed.

The authors note that only the CSA method (which includes a tyramine based enhancement process) allowed virtually the same degree of dilution as the EnVision +. Specifically in one case (viz. anti-cyclinD1 antibody) it achieved the gold-standard status. The authors were able to conclude that the EnVision + system was easy to use and it has the advantage that it avoids disturbing the endogenous biotin which is a failing of some of the other systems. It lowers the cost per test by allowing an increase in the dilution of the primary antibodies. It also has the advantage that it reduces the assay time and the lab workload and, by inference therefore, it reduces the overall cost per test as well.

An article by Hasui et al.(2002) was looking at the possibility of improving the supersensitive immunohistochemical investigations by means of a catalysed signal amplification system.(CSA system). Although the paper is not about a direct comparison between the two systems it should be noted that the team chose the EnVision + system to replace the sABC step because of the fact that endogenous biotin is detected by duplicated biotin-streptavidin reactions in the CSA process whereas the labelled polymer secondary antibody (EnVision + system) is both simpler than the standard CSA method and at least as sensitive as the sABC method. They found that any non-specific immunoreaction that was induced by the EnVision + system could be easily masked by a blocking protein in a competitive manner. The team found that this system produced better results than the standard CSA system. They also commented on the issue of endogenous biotin being spared by the EnVision + staining system

The paper by Belling et al. (1999) is interesting in so far as it provides a direct comparison between several of the currently employed immunohistochemical techniques namely APAAP, ABC, EnVision + and Immufl p4x methods. Belling comments on the number of different techniques currently available and compares these four against each other for sensitivity and specificity. Interestingly he also brings into the comparison other, less frequently considered parameters such as time spent using the techniques and also the overall cost of each procedure. His findings bear examination. He found that the APAAP was both the most time consuming and also the most expensive. He found that the EnVision + was the most economical with time and also produced very good staining results which he found to be comparable to the ABC and ImmunoMax techniques.

Kammerer (2001) produced another study looking at EnVision + as a practical gent for intra-operative staining. As his work was very similar to other studies already discussed, we shall not examine it in detail, but content ourselves with the comment that he concludes that the EnVision + system is both the fastest and most appropriate for this type of work. His results concur with the results found in the other work on the subject that reliable results can be obtained in 13 minutes.

Mokry (1996) produces a tour de force through all of the various systems that were in use. Again, there is no merit in repeating all of his findings other than to observe that, relevant to our considerations here, ChemMate EnVision is found to be at least as good as the best methods assessed both in terms of reproducibility and specificity. There is also comment about the comparative costs involved and ChemMate EnVision compares favourably.

Hasui et al. (2002) produce yet another comparative paper on the subject of immunohistochemical analysis. This paper focuses upon the problems of endogenous biotin staining. It finds that the system, which has benefits in other areas, the ImmunoMax/CSA system, has a problem with the staining of extraneous biotin which is where it considers ChemMate EnVision a better system. This paper also comments about the non-specific immunoreaction which was induced by the ChemMate EnVision system but they comment that it was masked competitively by a blocking protein.

Discussion and Conclusions

From the evidence presented so far we can safely conclude that both ChemMate EnVision and EnVision + systems are approximately equal in terms of specificity, reproducibility and range. They appear to outperform most of the alternative methods of immunohistochemical analysis, specifically the APAAP, CSA, LABC and SABC techniques. There are certainly shortcomings of the system and these have been discussed in the text. The major advantages that the two systems have over the alternatives mentioned are the fact that there does not appear to be a problem with the endogenous biotin causing aberrant staining, and the ease of use of the systems It is worth considering the activities of the UK National External Quality Assessment scheme for immunohistochemistry (UK NEQAS-ICC). It runs assessments four times a year and UK laboratories are able to participate in a number of different modules (assessments) depending on their particular speciality. In essence, it is an external system of verification to allow for standardisation of reporting and results. This system is of huge value Nationally as it allows for, and promotes consistency in, reporting and indeed preparing slides.

 It provides published document every quarter so that not only the labs themselves, but also the Public (should they want to) can see how each lab compares with their peers. It promotes and publicly rewards excellence. The reason that this is of interest to our considerations is that some of the systems commonly reported are ChemMate EnVision and EnVision + and they are regularly assessed against other methods. We can derive a considerable amount of comparative information from their published results

There is no merit at all in reproducing all of their published results as they deal with many other quality issues that are not relevant to our deliberations here, but examination of some of their findings is certainly worthy of comment. In one of the labs assessed the ChemMate EnVision system outperformed the EnVision + by a percentage of 67% to 52%. The lab offered an explanation of difficulties in the fixing of the slides for this anomaly.

As we have remarked above, neither system functioned particularly well with either the oestrogen or the progesterone receptors in the studies that we examined. In these assays however, the best system for the oestrogen receptors was found to be the Vector Ready PK7200 and the ChemMate EnVision was the best with the progesterone receptors. In an overall assessment, the UK NEQAS-ICC rated the EnVision + as good as the top performing system in five of the seven categories of examination . neither EnVision + nor ChemMate EnVision were rated as lower than in the top two in any division.

We can conclude that there is little difference in either specificity or performance between the two systems. They both have the advantage that they do not stain for endogenous biotin and this gives them an advantage over nearly all of their competitors. Both methods are comparatively quick. The EnVision + two stage system appears to be the system of choice for intra-operative sectioning.

References

Belling O, Ottesen K, Meyer W, Feller AC, Merz H. 1999 Comparative analysis of various standard immunohistochemical procedures Pathologe. 1999 Jul;20(4):242-50. Review. German.

Dodson.A. 2005 Streptavidin Biotin Complex (Duet) and two labelled polymer (EnVision + ) and ChemMate EnVision immonoperoxidease Detection Systems: a comparison DAKO website 2005-04-12

Hasui1, Tomio Takatsuka, Ryoichi Sakamoto, Liying Su, Sachie Matsushita, Shin-ichiro Tsuyama, Shuji Izumo and Fusayoshi Murata 2002 Improvement of Supersensitive Immunohistochemistry with an Autostainer: A Simplified Catalysed Signal Amplification System The Histochemical Journal Issue: Volume 34, Number 5 pp. 215 – 222. May 2002

Kämmerera, Michaela Kappa, Andrea Maria Gasselb, Thomas Richterc, Christian Tankd, Johannes Dietla, and Peter Rucke 2001 A New Rapid Immunohistochemical Staining Technique Using the EnVision Antibody Complex Journal of Histochemistry and Cytochemistry, Vol. 49, 623-630, May 2001

Mokry J.1996 Versatility of immunohistochemical reactions: comprehensive survey of detection systems. Acta Medica (Hradec Kralove). 1996;39(4):129-40.

Pileri SA, Roncador G, Ceccarelli C, Piccioli M, Briskomatis A, Sabattini E, Ascani S, Santini D, Piccaluga PP, Leone O, Damiani S, Ercolessi C, Sandri F, Pieri F, Leoncini L, Falini B. 1997 Antigen retrieval techniques in immunohistochemistry: comparison of different methods. J Pathol. 1997 Sep;183(1):116-23.

Ruck P. 2005-04-12 EnVision + for rapid immunostaining in intraoperative frozensection diagnosis Clinical Cancer Research Vol. 8, 759-767, March 2002

Sabattini, K Bisgaard, S Ascani, S Poggi, M Piccioli, C Ceccarelli, F Pieri, G Fraternali-Orcioni and SA Pileri 1998 The EnVision++ system: a new immunohistochemical method for diagnostics and research. Critical comparison with the APAAP, ChemMate, CSA, LABC, and SABC techniques Journal of Clinical Pathology 1998;51:506-511