- Email: [email protected]
Genogenic immunohistochemistry: a new era in diagnostic immunohistochemistry
Current Diagnostic Pathology (2002) 8, 193^200
c 2002 Elsevier Science Ltd. All rights reserved. doi:10.1054/ycdip.2002.0116, available online at http://www.idealibrary.com on
REVIEW
Genogenic immunohistochemistry: a new era in diagnostic immunohistochemistry A. M. Gown PhenoPath Laboratories and IRIS, 551N 34th Street, Suite 100, Seattle, Washington, USA and University of British Columbia,Vancouver, BC, Canada
KEYWORDS immunohistochemistry, anaplastic large cell lymphoma, breast cancer, desmoplastic small round cell tumour, hereditary non-polyposis colon cancer, PNET/Ewings sarcoma
Summary Immunohistochemical studies have traditionally focused on markers of speci¢c cell and/or tumour type as aids in the diagnosis of speci¢c tumours. However, as our knowledge of the molecular basis of tumours has increased, immunohistochemistryis being used withincreasing frequency to identify underlying molecularchanges or the presence of speci¢c molecular markersintumours, both as an aid to diagnosis and as a guide to appropriate therapy.There are three major classes of genogenic immunohistochemistry: (a) identi¢cation of speci¢c mutations, bydocumentation of loss of expression ofthe corresponding protein, as exempli¢ed by E-cadherininlobular breastcancer; (b) identi¢cation of proteins expressed as a consequence of speci¢c translocations, such as the t(2;5) translocation of anaplastic large cell lymphoma leading to ALK protein expression; and (c) identi¢cation of molecular targets of noveltumour therapies, e.g. identi¢cation of overexpression of the HER-2/neu protein to validate the use of trastuzumab (HerceptinTM) inthe treatmentof breastcancer patients.Thisreview o¡ers a snapshotof the current status of genogenic immunohistochemistry, which heralds a new era in diagc 2002 Elsevier Science Ltd. All rights reserved. nostic immunohistochemistry.
INTRODUCTION
This article provides a number of examples of ‘genogenic immunohistochemistry’, a phrase I have coined to describe this novel application of immunohistochemistry, where this protein-based technique is employed to answer questions about alterations at the molecular level. There are four major categories of molecular alterations that can be assessed by immunohistochemical methods: (a) identifying the presence of genetic mutations, generally truncation mutations, but also loss secondary to processes such as methylation; (b) identifying the presence of speci¢c chromosomal translocations owing to the identi¢cation of novel proteins expressed as a direct consequence of the translocation; (c) identifying the presence of gene ampli¢cation; and (d) identifying molecular therapeutic targets for the treatment of cancer and other diseases.
Traditionally, the goals of diagnostic immunohistochemical studies have been to explore and certify diagnoses by identifying the pathway of di¡erentiation (sometimes referred to as identifying the ‘cell of origin’) of a given tumour. Immunohistochemistry, the marriage of the precision of immunochemistry with the visualization provided by a sensitive detection system, is a science that has its origins in the work of Albert Coons, 60 years ago.1 Building upon the work of Coons, who initially sought to prove the presence of immunoglobulin molecules in plasma cells, the past half century has seen tremendous re¢nement in the development of the antibody tools and the detection systems. However, the overwhelming majority of this e¡ort has been directed at further analysis of protein expression as a marker of a particular cell, tissue or tumour type. However, as our understanding of the molecular basis of tumours has increased, it is becoming clear that many tumours have alterations that are best de¢ned at the molecular level.
Correspondence to: AMG.Tel.: +1206 374 9000; Fax: +1206 374 9009; E-mail: [email protected]
IDENTIFYING THE PRESENCE OF GENETIC MUTATIONS USING IMMUNOHISTOCHEMISTRY In each of these examples, immunohistochemistry is used to detect the ‘¢nal common pathway’ of loss of
194
protein expression, which may actually result from mutations, promotor methylation or other mechanisms.
Lobular carcinoma of the breast The ‘molecular signature’ of this disease is a genetic alteration which results in loss of expression of the Ecadherin protein. E-cadherin mediates adhesion and morphogenesis, and the resulting loss of expression of this protein is, in part, responsible for the loss of cell ^cell adhesion, which characterizes this tumour and leads to its distinctive histological appearance of ‘single cell’ in¢ltration. Loss of E-cadherin protein expression was ¢rst documented almost10 years ago by Moll et al.,2 in an immunohistochemical study employing cryostat and paraf¢n sections; noteworthy in their study was the demonstration of E-cadherin expression in all cases of high-grade in¢ltrating ductal carcinoma, i.e. the loss of expression of this protein was not a function of tumour grade per se but something unique to the lobular subtype.Our more recent studies3 have shown that virtually all cases of lobular carcinoma, as determined by studies in depara⁄nized, formalin ¢xed tissues, are characterized by loss of immunostaining with antibodies to Ecadherin (Fig.1).
Figure 1 In situ and in¢ltrating lobular carcinoma of the breast: (A) haematoxylin and eosin and (B) E-cadherin immunostain, showing retention of E-cadherin expression in ductal epithelial cells and complete loss of signal on focus of in situ and surrounding in¢ltrating lobular carcinoma (original magni¢cation 200).
CURRENT DIAGNOSTIC PATHOLOGY
The molecular correlates of this were ¢rst described by Kanai and colleagues,4 who examined the exons 5^ 8 (which includes the region essential for E-cadherin speci¢c binding) of the E-cadherin gene using PCR-SSCP as well as direct sequencing; in10% of the cases of lobular carcinoma examined, a point mutation in exon 7 was found. Berx and colleagues5 extended these ¢ndings, ¢nding E-cadherin mutations in 56% of lobular carcinomas, and not in any other forms of breast cancer (e.g. ductal). These authors also documented the presence of 23 di¡erent mutations in a series of lobular breast cancers, of which seven were insertions, 11 deletions, two nonsense mutations and three splice site mutations. All the frameshift and nonsense mutations would be expected to have generated secreted E-cadherin fragments instead of a transmembrane protein with cell adhesion capabilities. In a later paper, these same authors found mutations in the extracellular domain of 55% of lobular carcinomas, with complete loss of plasma membraneassociated E-cadherin expression in 84% of cases. Very recently, Droufakou and colleagues have demonstrated the heterogeneity of genetic alterations that lead to the ¢nal common pathway of lobular carcinoma.6 In this latter study, the overwhelming majority of lobular carcinoma showed methylation of the E-cadherin promoter. In addition, ¢ve frameshift mutations, which resulted in downstream stop codons and one splice site mutation were found in a subset of the tumours. It can therefore be concluded that most ILCs show genetic or epigenetic changes, e.g. biallelic inactivation of the E-cadherin gene by promoter methylation, mutation or allelic loss in any combination thereof, a¡ecting the E-cadherin gene and that most of these tumours, as a direct consequence, lack E-cadherin expression. Interestingly, Vos and colleagues7 have demonstrated that inactivating point mutations in the E-cadherin gene were found in lobular carcinoma in situ (LCIS) but not ductal carcinoma in situ (DCIS), suggesting that Ecadherin is a very early target gene in lobular neoplasia. More recent studies we have published,8 along with those of Jacobs et al.,9 have extended these ¢ndings of E-cadherin alterations in in situ disease, con¢rming the earlier results of Vos et al. In both of these recent studies, virtually all cases of LCIS were found via immunohistochemistry to be E-cadherin negative, whereas all DCIS cases were found to be E-cadherin positive. In addition, the clinical signi¢cance of demonstrating E-cadherin expression in the context of in situ breast carcinoma has recently been demonstrated.10
Microsatellite instability, mismatch repair genes and colonic neoplasia The development of the majority of sporadic colorectal adenocarcinomas, as well as adenocarcinomas
GENOGENIC IMMUNOHISTOCHEMISTRY: A NEW ERA IN DIAGNOSTIC IMMUNOHISTOCHEMISTRY
developing in the context of familial adenomatous polyps (FAP) has been documented to be a multistep process, as demonstrated in numerous studies from the Vogelstein laboratory.11 A smaller subset of sporadic colorectal adenocarcinomas, as well as colonic adenocarcinomas arising in the context of the autosomally dominant disorder known as hereditary non-polyposis colon cancer (HNPCC), arise by a di¡erent mechanism that involves alterations in genes that participate in DNA mismatch repair, with approximately 60% of HNPCC families harbouring mutations in two of these genes, hMSH2 and hMLH1.12^14 It has been demonstrated that the tumours
195
from patients with HNPCC demonstrate a unique type of genetic instability characterized by alterations within microsatellites, referred to as microsatellite instability (MSI). MSI is now recognized as the biological hallmark of the so-called mutator phenotype.15,16 Most of the MSI-positive tumours appear to be directly related to the altered expression of these two genes and can be diagnosed by the examination of protein expression Studies such as those of Kim et al.17 and Cawkwell et al.18 demonstrate that immunohistochemical studies identifying the loss of expression of the mismatch repair gene products hMLH1and hMSH2 can be e¡ective surrogates for the more expensive and time-consuming genetic studies looking for the presence of microsatellite instability. In the earlier study of Kim et al., in a series of 32 sporadic colorectal carcinomas, MSI was found in six (18%), of these six cases, ¢ve showed immunohistochemically determined loss of expression of one or both of the mismatch repair gene products. In the larger study of Cawkell et al., 66 (of 502 colorectal adenocarcinomas investigated) showed evidence of MSI (13%). Remarkably, 66/66 (100%) of these adenocarcinomas with MSI showed complete lack of expression of either hMLH1 or hMSH2 by immunohistochemical analysis. Since patients with mismatch repair defective carcinoma showed improved survival, but a 45-fold relative risk of developing a metachronous adenocarcinoma, the authors concluded that immunohistochemistry is the recommended method for selecting patients who should be investigated for HNPCC, o¡ered long-term follow-up, and who may not respond to standard chemotherapy modalities. An example of a colonic adenocarcinoma with MSI showing loss of expression of hMLH1 is shown in Figure 2.
IDENTIFYING THE PRESENCE OF CHROMOSOMALTRANSLOCATIONS USING IMMUNOHISTOCHEMISTRY ALK overexpression as an indicator of the t(2;5) translocation in anaplastic large cell lymphoma
Figure 2 Adenocarcinoma of the colon showing loss of hMLH1 as identi¢ed via immunohistochemistry: (A) haematoxylin and eosin and (B) hMLH1 immunostain. In (B), note complete loss of signalin nuclei oftumour cells (lower right) compared with the overlying normal colonic epithelium (top) (original magni¢cation 200).
Anaplastic large cell lymphoma (ALCL) is generally a rapidly progressing disease occurring in the ¢rst three decades of life, showing male predominance, and often presenting at extranodal sites, especially skin and bone. ALCL was initially referred as ‘Ki-1 lymphoma’ and ‘CD30 -positive anaplastic large cell lymphoma,’ re£ecting the morphological and immunophenotypic underpinnings of the original identi¢cation of this clinicopathological entity. The earliest suggestions that there was a unique genetic alteration in this tumour were provided by studies such as those of Rimokh and colleagues,19 who demonstrated lymphoma cases containing a chromosomal translocation involving the long arm of
196
chromosome 5. Subsequent studies revealed that this rearrangement fuses the NPM nucleolar phosphoprotein gene on chromosome 5q35 to a previously unidenti¢ed protein tyrosine kinase gene, ALK, on chromosome 2p23.20 RT-PCR studies performed by Downing et al.21 provided evidence that there was not a1:1correlation between the ‘ALCL’ phenotype and the presence of the (2;5) translocation and the presence of the NPM/ALK chimeric protein. Indeed, the NPM/ALK-expressing cases were not con¢ned to lymphomas with ‘anaplastic’ histological appearance alone, and some of the tumours with this translocation had other morphologies, e.g. diffuse large B-cell lymphoma. Furthermore, only slightly greater than half of the cases with classic histological features of ALCL, or those expressing CD30, were NPM/ ALK positive. As subsequently demonstrated by Shiota and colleagues,22 recognition of this NPM/ALK positive subgroup is of great clinical importance: those ALCLs positive for the translocation occurred in a far younger patient age group and the patients showed a far better 5year survival rate. Most importantly, in a landmark study by Lamant and colleagues, the t(2;5) translocation was detected by using cytogenetic analysis, reverse transcription, polymerase chain reaction (RT-PCR) and immunohistochemistry with an antibody directed against the kinase domain of anaplastic lymphoma kinase (ALK) of the chimeric NPM/ALK protein. In all but three cases, there was concordance among the cytogenetic analyses, RT-PCR, and immunohistochemical staining; the authors concluded that immunohistochemistry is a reliable method for detecting NPM/ALK chimeric protein and hence this subset of lymphomas. A subsequent study by Pittaluga et al.23 con¢rmed that immunohistochemistry allows the identi¢cation of a distinct subgroup of lymphomas that contain the 2p23 chromosomal abnormalities but lack ‘anaplastic’ morphology typical of ALCL. Furthermore, immunostaining proved to be the most sensitive method to identify this group. Through the use of these molecular and immunohistochemical techni-
CURRENT DIAGNOSTIC PATHOLOGY
ques, there has been a complete rethinking of the de¢nition of this tumour: as the histology may not be anaplastic and the cells neither anaplastic nor large, it has been suggested that the tumour should be renamed ‘ALK lymphoma’ or ‘ALKoma’ to re£ect this.24 Finally, variants of this disease with di¡erent translocations leading to di¡erent chimeric protein formation yielded a di¡erent pattern of immunostaining using antibodies to the ALK (or NPM) protein.25 This is perhaps the most profound demonstration to date of the power of immunohistochemistry to re£ect the diversity of genetic alteration in tumours. An example of a case of ALCL with ALK expression appears in Figure 3.
FLI-1 overexpression as an indicator of the t(11;22) translocation of PNET/ES Approximately 90% of primitive neuroectodermal tumours/Ewing’s sarcoma (PNET/ES) have a speci¢c t(11;22)(q24;q12), translocation, which results in the fusion of these genes on the der(22) chromosome, resulting in the production of a novel chimeric EWS/FLI-1 message, and overexpression of FLI-1 protein. While cytogenetic studies were originally employed to identify the presence of this chromosomal translocation, more recently identi¢cation of this genetic alteration has been possible through the use of appropriate primers using the RT-PCR technique.26 However, most recently it has been found that demonstration of the ‘molecular signature’ of this disease is possible by identifying the ‘neo-expression’ of the FLI-1 gene product using immunohistochemical techniques.27,28 In the study of Folpe and colleagues,28 and con¢rmed more recently by Llombart-Bosch and Navarro,29 FLI-1 expression has been demonstrated to be present in almost threequarters of PNET/ES, but not in other small blue round cell tumours with which it can be confused histologically. Immunohistochemical detection of FLI-1 may thus be a valuable technique for identi¢cation of PNET/ES in cases in which molecular genetic evaluation is not feasible. FLI-1 protein expression is also helpful in distinguishing PNET/ES from other tumours that may be CD99-positive.
WT-1 overexpression as an indicator of the t(11;22) translocation of DSRCT
Figure 3 Anaplastic large cell lymphoma showing positive immunostaining for ALK protein. Note nuclear and cytoplasmic localization of signal (original magni¢cation 400).
Another member of the small blue round cell group of paediatric tumours is the desmoplastic small round cell tumour (DSRCT); this distinctive tumour shows a di¡erent translocation involving chromosomes 11 and 22: t(11;22)(p13;q12). This results in unique chimeric EWS/ WT1 transcripts with expression of the corresponding fusion gene product.30 Since a portion of this fusion protein corresponds to the C-terminal portion of the
GENOGENIC IMMUNOHISTOCHEMISTRY: A NEW ERA IN DIAGNOSTIC IMMUNOHISTOCHEMISTRY
WT-1 gene product, it is not surprising that studies have demonstrated that antibodies to the WT-1gene product can detect the unique fusion product that results from this translocation, and can thus be used to identify desmoplastic small round cell tumour by immunohistochemistry.31^33 Indeed, among small blue round cell tumours (with the exception of Wilms tumour, which is generally con¢ned to the kidney), the DSRCT is unique in showing positive immunostaining for the WT-1 gene product, as shown in Figure 4.
USING IMMUNOHISTOCHEMISTRY TO IDENTIFY THERAPEUTIC TARGETS FOR CANCER HerceptinTM and HER-2/neu
197
among patients with metastatic disease, those eligible for immunotherapy treatment with the new humanized anti-HER-2/neu antibody, trastuzumab (HerceptinTM, Genentech, Inc. South San Francisco, CA). Determination of HER-2/neu status has thus taken on great importance in the routine work-up of breast cancer in the pathology laboratory, complementing but not replacing the traditional diagnostic and prognostic markers such as tumour type, tumour size, tumour grade and lymph node status. Because the alterations of HER-2/neu occur simultaneously at the genomic and protein levels, a number of studies have addressed the issue of comparative sensitivity and speci¢city of the two techniques. Controversy has ensued over the reproducibility and reliability of the semiquantitative IHC scoring system (0, 1+, 2+, and 3+) employed,47 and the use of FISH, for which binary results
The HER-2/neu (c-erbB-2) oncogene protein is a transmembrane glycoprotein in the epidermal growth factor receptor family. It is expressed at low levels in a variety of normal epithelia, including breast duct epithelium, but is ‘overexpressed’ at signi¢cantly higher levels in approximately one-¢fth to one-quarter of invasive breast cancers, usually as a direct consequence of ampli¢cation of the HER-2/neu gene.34 Many studies have now demonstrated the importance of HER-2/neu overexpression as an independent prognostic marker of clinical outcome, particularly in node-positive patients;35^37 recent studies also suggest that HER-2/neu status predicts response to adriamycin-based adjuvant chemotherapy.38 ^ 40 In addition, the predictive value of HER-2/neu overexpression in node-negative patients has been demonstrated,41 with the additional ¢nding that HER-2/neu overexpression is associated with poor response to tamoxifen, even in the setting of positive oestrogen receptor (ER) expression.41^ 44 Most recently,45,46 HER-2/neu protein overexpression in breast cancer has been used to select,
Figure 4 Desmoplastic small round cell tumour of childhood showing strong, uniform nuclear immunostaining with antibodies to the C-terminal portion of the Wilms tumour gene product (original magni¢cation 200).
Figure 5 In¢ltrating ductal carcinoma of the breast showing ampli¢cation of the HER-2/neu gene with concomitant protein overexpression. (A) Fluorescence in situ hybridization, showing multiple orange signals (corresponding to HER-2/neu gene) per tumour cell nucleus, with average of two green signals (corresponding to chromosome17 centromere) per nucleus. (B) Immunohistochemical demonstration of 3+ level of immunostaining for HER-2/neu gene product [original magni¢cations 1000 (A), 400 (B)].
198
are reported (ampli¢ed/non-ampli¢ed), has been recommended by some.48 While 2+ and 3+ immunostain scores have been considered as positive results, questions have persisted, particularly with respect to the signi¢cance of the 2+ level of immunostaining.49 Recent studies attempting to compare detection of protein overexpression (measured by IHC) to gene ampli¢cation (measured by FISH or PCR) showed concordance rates ranging from 78% to 98%,48 ^57 with most studies showing over 90% concordance rates. The recent study by Pauletti et al.52 is unique in ¢nding poor concordance, but there is reason to believe that there were problems with the IHC component of that study, in which only 17% of all cases were either 2+ or 3+ positive. In a more recent study from our laboratory, involving a series of 306 invasive breast cancer cases, we performed parallel IHC and FISH studies. Among the 173 cases negative for protein overexpression via IHC, only ¢ve cases (2.9%) showed gene ampli¢cation. Among the 21 cases showing a 3+ positive overexpression score, 100% showed gene ampli¢cation. Among the112 cases showing a 2+ positive immunostaining score, 21 (18.8%) showed gene ampli¢cation. However, among the 91 (81.2%) 2+ IHC cases negative for gene ampli¢cation, 46 (41%) showed an ampli¢cation score between 1 and 2. This suggests that immunohistochemical scores (i.e. 2+) may correlate with low levels of gene ampli¢cation, and provides further evidence to suggest that IHC, when employed with optimal antibodies and a scoring system, provides a highly accurate assessment of HER-2/neu gene status. Figure 5 compares HER-2/neu assessment by FISH and immunohistochemistry.
STI571 (GleevecTM; imatinib mesylate, formerly known as STI571) and gastrointestinal stromal tumours In the early 1990s, a drug, eventually called STI571, was developed that targeted the ‘kinase pocket’ of the BCRABL protein, the fusion gene product characteristic of chronic myeloid leukaemia (CML) cells.58,59 The BCRABL gene product manifests deregulated tyrosine kinase activity, which is instrumental in the pathogenesis of CML, and GleevecTM acts by blocking this tyrosine kinase activity. GleevecTM targets not only the BCR-ABL gene product, however, but also is e¡ective in blocking the ‘kinase pocket’of the platelet-derived growth factor receptor60 and c-kit.61 The latter is expressed in (and highly characteristic of) gastrointestinal stromal tumours (GIST), and recent published and presented data suggest that GleevecTM is a highly e¡ective reagent for the treatment of GISTs, a tumour heretofore without e¡ective therapy.61 Thus, immunohistochemical detection of the c-kit gene product can identify tumours that might be treatable with GleevecTM.62
CURRENT DIAGNOSTIC PATHOLOGY
CONCLUSION We are about to enter a new era of immunohistochemistry testing, moving from the current era in which this technique is used solely to determine the cell or tumour type. In this new era, which Albert Coons could not have foretold, immunohistochemistry will be used for the assessment of unique phenotypic aspects of the cell or tumour which directly re£ect the underlying genetic alterations of a given tumour and which will, with increasing frequency, determine the optimal treatment for the patient.
PRACTICE POINTS K
K
K
K
The range of diagnostic immunohistochemistry is being expanded to include the identi¢cation of changes at the protein level which re£ect molecular alterations thoughtto underlie speci¢c tumours Novel immunohistochemical tests now exist to identify, by virtue of the loss of immunostaining, lobular carcinoma of the breast and colonic adenocarcinomas such as those occurring in the context of HNPCC, which exhibit the molecular phenomenon of microsatellite instability Novel immunohistochemical tests now exist to identify the speci¢c chimeric proteins expressed as a consequence of chromosomal translocations characteristic of anaplastic large cell lymphoma, PNET/Ewings sarcoma, and desmoplastic small round cell tumour Novel immunohistochemical tests now exist to identify the targets of new therapies for cancer, e.g. theidenti¢cation ofthe HER-2/neu gene productin a subset of breast cancers
RESEARCH DIRECTIONS K
K
K
Developmentof antibodies to other proteins deleted as a consequence of tumour-speci¢c genetic mutations Development of antibodies to chimeric proteins expressed as a consequence of chromosomal translocations of other tumours Development of antibodies to other targets of novel cancer therapeutics
REFERENCES 1. Coons A H, Creech H J, Jones R N. Immunological properties of an antibody containing a £uorescent group. Proc Soc Exp Biol1941; 47: 200 ^202. 2. Moll R, Mitze M, Frixen U H, Birchmeier W. Di¡erential loss of E-cadherin expression in in¢ltrating ductal and lobular breast carcinomas. Am J Pathol 1993; 143: 1731^1742.
GENOGENIC IMMUNOHISTOCHEMISTRY: A NEW ERA IN DIAGNOSTIC IMMUNOHISTOCHEMISTRY
3. Lehr H A, Folpe A,Yaziji H, Kommoss F, Gown A M. Cytokeratin 8 immunostaining pattern and E-cadherin expression distinguish lobular from ductal breast carcinoma. Am J Clin Pathol 2000; 114: 190 ^196. 4. Kanai Y, Oda T, Tsuda H, Ochiai A, Hirohashi S. Point mutation of the E-cadherin gene in invasive lobular carcinoma of the breast. Jpn J Cancer Res1994; 85: 1035^1039. 5. Berx G,Cleton J-A M, Strumane K, de L-WJ, Nollet F, van R-F et al. E-cadherin is inactivated in a majority of invasive human lobular breast cancers by truncation mutations throughout its extracellular domain. Oncogene1996; 13: 1919^1925. 6. Droufakou S, DeshmaneV, Roylance R, Hanby A,Tomlinson I, Hart I R. Multiple ways of silencing E-cadherin gene expression in lobular carcinoma of the breast. Int J Cancer 2001; 92: 404 ^ 408. 7. Vos C B, Cleton J-A M, Berx G, de L-W J, ter H-N T, van R-F et al. E-cadherin inactivation in lobular carcinoma in situ of the breast: an early event in tumourigenesis. Br J Cancer 1997; 76: 1131^1133. 8. Goldstein N S, Bassi D, Watts J C, Lay¢eld L J,Yaziji H, Gown A M. E-cadherin reactivity of 95 noninvasive ductal and lobular lesions of the breast. Implications for the interpretation of problematic lesions. Am J Clin Pathol 2001; 115: 534 ^542. 9. JacobsT W, Pliss N, Kouria G, Schnitt S J. Carcinomas in situ of the breast with indeterminate features: role of E-cadherin staining in categorization. Am J Surg Pathol 2001; 25: 229^236. 10. Goldstein N S, Kestin L L,Vicini F A.Clinicopathologic implications of E-cadherin reactivity in patients with lobular carcinoma in situ of the breast. Cancer 2001; 92: 738 ^747. 11. Fearon E R, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990; 61: 759^767. 12. Marra G, Boland C R. Hereditary nonpolyposis colorectal cancer: the syndrome, the genes, and historical perspectives. J Natl Cancer Inst1995; 87: 1114 ^1125. 13. Peltomaki P, de la Chapelle A. Mutations predisposing to hereditary nonpolyposis colorectal cancer. Adv Cancer Res 1997; 71: 93^119. 14. Jass J R. Pathology of hereditary nonpolyposis colorectal cancer. Ann NYAcad Sci 2000; 910: 62^73; discussion 73^74. 15. Loeb L A. Microsatellite instability: marker of a mutator phenotype in cancer. Cancer Res 1994; 54: 5059^5063. 16. Loeb L A. A mutator phenotype in cancer. Cancer Res 2001; 61: 3230 ^3239. 17. Kim H, Piao Z, Kim J W,Choi J S, Kim N K, Lee J M et al. Expression of hMSH2 and hMLH1 in colorectal carcinomas with microsatellite instability. Pathol Res Practice 1998; 194: 3^9. 18. Cawkwell L, Gray S, Murgatroyd H, Sutherland F, Haine L, Longfellow M et al. Choice of management strategy for colorectal cancer based on a diagnostic immunohistochemical test for defective mismatch repair [see comments].Gut 1999; 45: 409^ 415. 19. Rimokh R, Magaud J P, Berger F, Samarut J, Coi⁄er B, Germain D et al. A translocation involving a speci¢c breakpoint (q35) on chromosome 5 is characteristic of anaplastic large cell lymphoma (‘Ki-1 lymphoma’). Br J Haematol 1989; 71: 31^36. 20. Morris S W, Kirstein M N, Valentine M B, Dittmer KG, Shapiro D N, Saltman D L et al. Fusion of a kinase gene, ALK, to a nucleolar protein gene, NPM, in non-Hodgkin’s lymphoma. Science1994; 263: 1281^1284. 21. Downing J R, Shurtle¡ S A, Zielenska M,Curcio-Brint A M, Behm F G, Head D R et al. Molecular detection of the (2;5) translocation of non-Hodgkin’s lymphoma by reverse transcriptase ^ polymerase chain reaction. Blood1995; 85: 3416 ^3422. 22. Shiota M, Nakamura S, Ichinohasama R, Abe M, Akagi T, Takeshita M et al. Anaplastic large cell lymphomas expressing the novel chimeric protein p80NPM/ALK: a distinct clinicopathologic entity. Blood1995; 86: 1954 ^1960. 23. Pittaluga S, Wiodarska I, Pulford K, Campo E, Morris S W, Van dB-H et al.The monoclonal antibody ALK1identi¢es a distinct mor-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
199
phological subtype of anaplastic large cell lymphoma associated with 2p23/ALK rearrangements. Am J Pathol 1997; 151: 343^351. Benharroch D, Meguerian-Bedoyan Z, Lamant L, Amin C, Brugieres L, Terrier-Lacombe M J et al. ALK-positive lymphoma: a single disease with a broad spectrum of morphology. Blood 1998; 91: 2076^2084. Falini B, Pulford K, Pucciarini A, Carbone A, De Wolf-Peeters C, Cordell J et al. Lymphomas expressing ALK fusion protein (s) other than NPM-ALK. Blood1999; 94: 3509^3515. Downing J R, Head D R, Parham D M, Douglass E C, Hulshof M G, Link M P et al. Detection of the (11;22)(q24;q12) translocation of Ewing’s sarcoma and peripheral neuroectodermal tumour by reverse transcription polymerase chain reaction. Am J Pathol 1993; 143: 1294 ^1300. Melot T,Gruel N, Doubeikovski A, Sevenet N,Teillaud JL, Delattre O. Production and characterization of mouse monoclonal antibodies to wild-type and oncogenic FLI-1proteins. Hybridoma1997; 16: 457^ 464. Folpe A L, Hill C E, Parham D M, O’Shea P A, Weiss S W. Immunohistochemical detection of FLI-1 protein expression: a study of 132 round cell tumours with emphasis on CD99-positive mimics of Ewing’s sarcoma/primitive neuroectodermal tumour. Am J Surg Pathol 2000; 24: 1657^1662. Llombart-Bosch A, Navarro S. Immunohistochemical detection of EWS and FLI-1 proteins in Ewing sarcoma and primitive neuroectodermal tumours: comparative analysis with CD99 (MIC-2) expression. Appl Immunohistochem Mol Morphol 2001; 9: 255^260. Gerald W L, Ladanyi M, de Alava E, Cuatrecasas M, Kushner B H, LaQuaglia M P et al. Clinical, pathologic, and molecular spectrum of tumours associated with t(11;22)(p13;q12): desmoplastic small round-cell tumour and its variants. Clin Oncol 1998; 16: 3028 ^3036. Barnoud R, Delattre O, Pe¤oc’h M, Pasquier D, Plantaz D, Leroux D et al. Desmoplastic small round cell tumour: RT-PCR analysis and immunohistochemical detection of the Wilm’s tumour gene WT1. Pathol Res Practice 1998; 194: 693^700. Charles A K, Moore I E, Berry P J. Immunohistochemical detection of the Wilms’ tumour gene WT1 in desmoplastic small round cell tumour. Histopathology1997; 30: 312^314. Ordonez N G. Desmoplastic small round cell tumour: II: an ultrastructural and immunohistochemical study with emphasis on new immunohistochemical markers. Am J Surg Pathol 1998; 22: 1314 ^1327. Slamon D J, Clark G M, Wong S G, Levin W J, Ullrich A, McGuire W L. Human breast cancer: correlation of relapse and survival with ampli¢cation of the HER-2/neu oncogene. Science 1987; 235: 177^182. Toikkanen S, Helin H, Isola J, Joensuu H. Prognostic signi¢cance of HER-2 oncoprotein expression in breast cancer: a 30 -year followup [see comments]. J Clin Oncol 1992; 10: 1044 ^1048. Paik S, Hazan R, Fisher E R, Sass R E, Fisher B, Redmond C et al. Pathologic ¢ndings from the National Surgical Adjuvant Breast and Bowel Project: prognostic signi¢cance of erbB-2 protein overexpression in primary breast cancer [Prior annotation incorrect]. J Clin Oncol 1990; 8:103^112. Tandon A K, Clark G M, Chamness G C, Ullrich A, McGuire W L. HER-2/neu oncogene protein and prognosis in breast cancer. J Clin Oncol 1989; 7: 1120 ^1128. Muss H B, Thor A D, Berry D A, KuteT, Liu E T, Koerner F et al. c-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994; 330: 1260 ^1266. Thor A D, Berry D A, Budman D R, Muss H B, KuteT, Henderson I C et al. erbB-2, p53, and e⁄cacy of adjuvant therapy in lymph node-positive breast cancer [see comments]. J Natl Cancer Inst 1998; 90: 1346 ^1360.
200
40. Paik S, Bryant J, Park C, Fisher B, Tan-Chiu E, Hyams D et al. erbB-2 and response to doxorubicin in patients with axillary lymph node-positive, hormone receptor-negative breast cancer [see comments]. J Natl Cancer Inst 1998; 90: 1361^1370. 41. Sjogren S, Inganas M, Lindgren A, Holmberg L, Bergh J. Prognostic and predictive value of c-erbB-2 overexpression in primary breast cancer, alone and in combination with other prognostic markers. J Clin Oncol 1998; 16: 462^ 469. 42. Wright C, Nicholson S, Angus B, Sainsbury J R, Farndon J, Cairns J et al. Relationship between c-erbB-2 protein product expression and response to endocrine therapy in advanced breast cancer. Br J Cancer 1992; 65: 118 ^121. 43. Carlomagno C, Perrone F, Gallo C, De Laurentiis M, Lauria R, Morabito A et al. c-erb B2 overexpression decreases the bene¢t of adjuvant tamoxifen in early-stage breast cancer without axillary lymph node metastases. J Clin Oncol 1996; 14: 2702^2708. 44. Te“tu B, Brisson J, PlanteV, Bernard P. p53 and c-erbB-2 as markers of resistance to adjuvant chemotherapy in breast cancer. Mod Pathol 1998; 11: 823^ 830. 45. Shak S. Overview of the trastuzumab (Herceptin) anti-HER2 monoclonal antibody clinical program in HER2-overexpressing metastatic breast cancer. Herceptin Multinational Investigator Study Group. Semin Oncol 1999; 26 (4 Suppl 12): 71^77. 46. Slamon D J, Leyland Jones B, Shak S, Fuchs H, PatonV, Bajamonde A et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344: 783^792. 47. JacobsT W, Gown A M,Yaziji H, Barnes M J, Schnitt S J. Speci¢city of HercepTest in determining HER-2/neu status of breast cancers using the United States Food and Drug Administration-Approved Scoring System. J Clin Oncol 1999; 17: 1983. 48. Tubbs R R, Pettay J D, Roche P C, Stoler M H, Jenkins R B,GroganT M. Discrepancies in clinical laboratory testing of eligibility for trastuzumab therapy: apparent immunohistochemical false-positives do not get the message. J Clin Oncol 2001; 19: 2714 ^2721. 49. Birner P, Oberhuber G, Stani J, Reithofer C, Samonigg H, Hausmaninger H et al. Evaluation of the United States food and drug administration-approved scoring and test system of her-2 protein expression in breast cancer. Clin Cancer Res 2001; 7: 1669^1675. 50. JacobsT W,Gown A M,Yaziji H, Barnes M J, Schnitt S J.Comparison of £uorescence in situ hybridization and immunohistochemistry for the evaluation of HER-2/neu in breast cancer. J Clin Oncol 1999; 17: 1974 ^1982. 51. Wang S, Saboorian M H, Frenkel E, Hynan L, Gokaslan S T, Ashfaq R. Laboratory assessment of the status of Her-2/neu protein and
CURRENT DIAGNOSTIC PATHOLOGY
52.
53.
54.
55.
56.
57.
58.
59. 60.
61.
62.
oncogene in breast cancer specimens: comparison of immunohistochemistry assay with £uorescence in situ hybridisation assays. J Clin Pathol 2000; 53: 374 ^381. Pauletti G, Dandekar S, Rong H, Ramos L, Peng H et al. Assessment of methods for tissue-based detection of the HER-2/neu alteration in human breast cancer: a direct comparison of £uorescence in situ hybridization and immunohistochemistry. J Clin Oncol 2000; 18: 3651^3664. Kakar S, Puangsuvan N, Stevens J M, Serenas R, Mangan G, Sahai S et al. HER-2/neu assessment in breast cancer by immunohistochemistry and £uorescence in situ hybridization: comparison of results and correlation with survival. Mol Diagn 2000; 5: 199^207. Couturier J,Vincent-Salomon A, Nicolas A, Beuzeboc P, Mouret E, Zafrani B et al. Strong correlation between results of £uoresence in situ hybridization and immunohistochemistry for the assessment of ERBB2 (HER-2/neu) gene status in breast carcinoma. Mod Pathol 2000; 13: 1238 ^1243. Hoang M P, Sahin A A, Ordonez N G, Sneige N. HER-2/neu gene ampli¢cation compared with HER-2/neu protein overexpression and interobserver reproducibility in invasive breast carcinoma. Am J Clin Pathol 2000; 113: 852^ 859. Ridol¢ R L, Jamehdor M R, Arber J M. HER-2/neu testing in breast carcinoma: a combined immunohistochemical and £uorescent in situ hybridization approach. Mod Pathol 2000; 13: 866^ 873. Lebeau A, Deimling D, Kaltz C, Sendelhofert A, I¡ A, Luthardt B et al. HER-2/neu analysis in archival tissue samples of human breast cancer: comparison of immunohistochemistry and £uorescence in situ hybridization. J Clin Oncol 2001; 19: 354 ^363. Buchdunger E, Matter A, Druker B J. Bcr-Abl inhibition as a modality of CML therapeutics. Biochim Biophys Acta 2001; 1551: M11^M18. O’Dwyer M E, Druker B J. Chronic myelogenous leukaemiaFnew therapeutic principles. J Intern Med 2001; 250: 3^9. Pietras K, Ostman A, Sjoquist M, Buchdunger E, Reed R K, Heldin C H et al. Inhibition of platelet-derived growth factor receptors reduces interstitial hypertension and increases transcapillary transport in tumours. Cancer Res 2001; 61: 2929^2934. Joensuu H, Roberts P J, Sarlomo-Rikala M, Andersson L C, Tervahartiala P, Tuveson D et al. E¡ect of the tyrosine kinase inhibitor STI571in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 2001; 344: 1052^1056. Berman J, O’Leary T J. Gastrointestinal stromal tumor workshop. Hum Pathol 2001; 32: 578 ^582.
c 2002 Elsevier Science Ltd. All rights reserved. doi:10.1054/ycdip.2002.0116, available online at http://www.idealibrary.com on
REVIEW
Genogenic immunohistochemistry: a new era in diagnostic immunohistochemistry A. M. Gown PhenoPath Laboratories and IRIS, 551N 34th Street, Suite 100, Seattle, Washington, USA and University of British Columbia,Vancouver, BC, Canada
KEYWORDS immunohistochemistry, anaplastic large cell lymphoma, breast cancer, desmoplastic small round cell tumour, hereditary non-polyposis colon cancer, PNET/Ewings sarcoma
Summary Immunohistochemical studies have traditionally focused on markers of speci¢c cell and/or tumour type as aids in the diagnosis of speci¢c tumours. However, as our knowledge of the molecular basis of tumours has increased, immunohistochemistryis being used withincreasing frequency to identify underlying molecularchanges or the presence of speci¢c molecular markersintumours, both as an aid to diagnosis and as a guide to appropriate therapy.There are three major classes of genogenic immunohistochemistry: (a) identi¢cation of speci¢c mutations, bydocumentation of loss of expression ofthe corresponding protein, as exempli¢ed by E-cadherininlobular breastcancer; (b) identi¢cation of proteins expressed as a consequence of speci¢c translocations, such as the t(2;5) translocation of anaplastic large cell lymphoma leading to ALK protein expression; and (c) identi¢cation of molecular targets of noveltumour therapies, e.g. identi¢cation of overexpression of the HER-2/neu protein to validate the use of trastuzumab (HerceptinTM) inthe treatmentof breastcancer patients.Thisreview o¡ers a snapshotof the current status of genogenic immunohistochemistry, which heralds a new era in diagc 2002 Elsevier Science Ltd. All rights reserved. nostic immunohistochemistry.
INTRODUCTION
This article provides a number of examples of ‘genogenic immunohistochemistry’, a phrase I have coined to describe this novel application of immunohistochemistry, where this protein-based technique is employed to answer questions about alterations at the molecular level. There are four major categories of molecular alterations that can be assessed by immunohistochemical methods: (a) identifying the presence of genetic mutations, generally truncation mutations, but also loss secondary to processes such as methylation; (b) identifying the presence of speci¢c chromosomal translocations owing to the identi¢cation of novel proteins expressed as a direct consequence of the translocation; (c) identifying the presence of gene ampli¢cation; and (d) identifying molecular therapeutic targets for the treatment of cancer and other diseases.
Traditionally, the goals of diagnostic immunohistochemical studies have been to explore and certify diagnoses by identifying the pathway of di¡erentiation (sometimes referred to as identifying the ‘cell of origin’) of a given tumour. Immunohistochemistry, the marriage of the precision of immunochemistry with the visualization provided by a sensitive detection system, is a science that has its origins in the work of Albert Coons, 60 years ago.1 Building upon the work of Coons, who initially sought to prove the presence of immunoglobulin molecules in plasma cells, the past half century has seen tremendous re¢nement in the development of the antibody tools and the detection systems. However, the overwhelming majority of this e¡ort has been directed at further analysis of protein expression as a marker of a particular cell, tissue or tumour type. However, as our understanding of the molecular basis of tumours has increased, it is becoming clear that many tumours have alterations that are best de¢ned at the molecular level.
Correspondence to: AMG.Tel.: +1206 374 9000; Fax: +1206 374 9009; E-mail: [email protected]
IDENTIFYING THE PRESENCE OF GENETIC MUTATIONS USING IMMUNOHISTOCHEMISTRY In each of these examples, immunohistochemistry is used to detect the ‘¢nal common pathway’ of loss of
194
protein expression, which may actually result from mutations, promotor methylation or other mechanisms.
Lobular carcinoma of the breast The ‘molecular signature’ of this disease is a genetic alteration which results in loss of expression of the Ecadherin protein. E-cadherin mediates adhesion and morphogenesis, and the resulting loss of expression of this protein is, in part, responsible for the loss of cell ^cell adhesion, which characterizes this tumour and leads to its distinctive histological appearance of ‘single cell’ in¢ltration. Loss of E-cadherin protein expression was ¢rst documented almost10 years ago by Moll et al.,2 in an immunohistochemical study employing cryostat and paraf¢n sections; noteworthy in their study was the demonstration of E-cadherin expression in all cases of high-grade in¢ltrating ductal carcinoma, i.e. the loss of expression of this protein was not a function of tumour grade per se but something unique to the lobular subtype.Our more recent studies3 have shown that virtually all cases of lobular carcinoma, as determined by studies in depara⁄nized, formalin ¢xed tissues, are characterized by loss of immunostaining with antibodies to Ecadherin (Fig.1).
Figure 1 In situ and in¢ltrating lobular carcinoma of the breast: (A) haematoxylin and eosin and (B) E-cadherin immunostain, showing retention of E-cadherin expression in ductal epithelial cells and complete loss of signal on focus of in situ and surrounding in¢ltrating lobular carcinoma (original magni¢cation 200).
CURRENT DIAGNOSTIC PATHOLOGY
The molecular correlates of this were ¢rst described by Kanai and colleagues,4 who examined the exons 5^ 8 (which includes the region essential for E-cadherin speci¢c binding) of the E-cadherin gene using PCR-SSCP as well as direct sequencing; in10% of the cases of lobular carcinoma examined, a point mutation in exon 7 was found. Berx and colleagues5 extended these ¢ndings, ¢nding E-cadherin mutations in 56% of lobular carcinomas, and not in any other forms of breast cancer (e.g. ductal). These authors also documented the presence of 23 di¡erent mutations in a series of lobular breast cancers, of which seven were insertions, 11 deletions, two nonsense mutations and three splice site mutations. All the frameshift and nonsense mutations would be expected to have generated secreted E-cadherin fragments instead of a transmembrane protein with cell adhesion capabilities. In a later paper, these same authors found mutations in the extracellular domain of 55% of lobular carcinomas, with complete loss of plasma membraneassociated E-cadherin expression in 84% of cases. Very recently, Droufakou and colleagues have demonstrated the heterogeneity of genetic alterations that lead to the ¢nal common pathway of lobular carcinoma.6 In this latter study, the overwhelming majority of lobular carcinoma showed methylation of the E-cadherin promoter. In addition, ¢ve frameshift mutations, which resulted in downstream stop codons and one splice site mutation were found in a subset of the tumours. It can therefore be concluded that most ILCs show genetic or epigenetic changes, e.g. biallelic inactivation of the E-cadherin gene by promoter methylation, mutation or allelic loss in any combination thereof, a¡ecting the E-cadherin gene and that most of these tumours, as a direct consequence, lack E-cadherin expression. Interestingly, Vos and colleagues7 have demonstrated that inactivating point mutations in the E-cadherin gene were found in lobular carcinoma in situ (LCIS) but not ductal carcinoma in situ (DCIS), suggesting that Ecadherin is a very early target gene in lobular neoplasia. More recent studies we have published,8 along with those of Jacobs et al.,9 have extended these ¢ndings of E-cadherin alterations in in situ disease, con¢rming the earlier results of Vos et al. In both of these recent studies, virtually all cases of LCIS were found via immunohistochemistry to be E-cadherin negative, whereas all DCIS cases were found to be E-cadherin positive. In addition, the clinical signi¢cance of demonstrating E-cadherin expression in the context of in situ breast carcinoma has recently been demonstrated.10
Microsatellite instability, mismatch repair genes and colonic neoplasia The development of the majority of sporadic colorectal adenocarcinomas, as well as adenocarcinomas
GENOGENIC IMMUNOHISTOCHEMISTRY: A NEW ERA IN DIAGNOSTIC IMMUNOHISTOCHEMISTRY
developing in the context of familial adenomatous polyps (FAP) has been documented to be a multistep process, as demonstrated in numerous studies from the Vogelstein laboratory.11 A smaller subset of sporadic colorectal adenocarcinomas, as well as colonic adenocarcinomas arising in the context of the autosomally dominant disorder known as hereditary non-polyposis colon cancer (HNPCC), arise by a di¡erent mechanism that involves alterations in genes that participate in DNA mismatch repair, with approximately 60% of HNPCC families harbouring mutations in two of these genes, hMSH2 and hMLH1.12^14 It has been demonstrated that the tumours
195
from patients with HNPCC demonstrate a unique type of genetic instability characterized by alterations within microsatellites, referred to as microsatellite instability (MSI). MSI is now recognized as the biological hallmark of the so-called mutator phenotype.15,16 Most of the MSI-positive tumours appear to be directly related to the altered expression of these two genes and can be diagnosed by the examination of protein expression Studies such as those of Kim et al.17 and Cawkwell et al.18 demonstrate that immunohistochemical studies identifying the loss of expression of the mismatch repair gene products hMLH1and hMSH2 can be e¡ective surrogates for the more expensive and time-consuming genetic studies looking for the presence of microsatellite instability. In the earlier study of Kim et al., in a series of 32 sporadic colorectal carcinomas, MSI was found in six (18%), of these six cases, ¢ve showed immunohistochemically determined loss of expression of one or both of the mismatch repair gene products. In the larger study of Cawkell et al., 66 (of 502 colorectal adenocarcinomas investigated) showed evidence of MSI (13%). Remarkably, 66/66 (100%) of these adenocarcinomas with MSI showed complete lack of expression of either hMLH1 or hMSH2 by immunohistochemical analysis. Since patients with mismatch repair defective carcinoma showed improved survival, but a 45-fold relative risk of developing a metachronous adenocarcinoma, the authors concluded that immunohistochemistry is the recommended method for selecting patients who should be investigated for HNPCC, o¡ered long-term follow-up, and who may not respond to standard chemotherapy modalities. An example of a colonic adenocarcinoma with MSI showing loss of expression of hMLH1 is shown in Figure 2.
IDENTIFYING THE PRESENCE OF CHROMOSOMALTRANSLOCATIONS USING IMMUNOHISTOCHEMISTRY ALK overexpression as an indicator of the t(2;5) translocation in anaplastic large cell lymphoma
Figure 2 Adenocarcinoma of the colon showing loss of hMLH1 as identi¢ed via immunohistochemistry: (A) haematoxylin and eosin and (B) hMLH1 immunostain. In (B), note complete loss of signalin nuclei oftumour cells (lower right) compared with the overlying normal colonic epithelium (top) (original magni¢cation 200).
Anaplastic large cell lymphoma (ALCL) is generally a rapidly progressing disease occurring in the ¢rst three decades of life, showing male predominance, and often presenting at extranodal sites, especially skin and bone. ALCL was initially referred as ‘Ki-1 lymphoma’ and ‘CD30 -positive anaplastic large cell lymphoma,’ re£ecting the morphological and immunophenotypic underpinnings of the original identi¢cation of this clinicopathological entity. The earliest suggestions that there was a unique genetic alteration in this tumour were provided by studies such as those of Rimokh and colleagues,19 who demonstrated lymphoma cases containing a chromosomal translocation involving the long arm of
196
chromosome 5. Subsequent studies revealed that this rearrangement fuses the NPM nucleolar phosphoprotein gene on chromosome 5q35 to a previously unidenti¢ed protein tyrosine kinase gene, ALK, on chromosome 2p23.20 RT-PCR studies performed by Downing et al.21 provided evidence that there was not a1:1correlation between the ‘ALCL’ phenotype and the presence of the (2;5) translocation and the presence of the NPM/ALK chimeric protein. Indeed, the NPM/ALK-expressing cases were not con¢ned to lymphomas with ‘anaplastic’ histological appearance alone, and some of the tumours with this translocation had other morphologies, e.g. diffuse large B-cell lymphoma. Furthermore, only slightly greater than half of the cases with classic histological features of ALCL, or those expressing CD30, were NPM/ ALK positive. As subsequently demonstrated by Shiota and colleagues,22 recognition of this NPM/ALK positive subgroup is of great clinical importance: those ALCLs positive for the translocation occurred in a far younger patient age group and the patients showed a far better 5year survival rate. Most importantly, in a landmark study by Lamant and colleagues, the t(2;5) translocation was detected by using cytogenetic analysis, reverse transcription, polymerase chain reaction (RT-PCR) and immunohistochemistry with an antibody directed against the kinase domain of anaplastic lymphoma kinase (ALK) of the chimeric NPM/ALK protein. In all but three cases, there was concordance among the cytogenetic analyses, RT-PCR, and immunohistochemical staining; the authors concluded that immunohistochemistry is a reliable method for detecting NPM/ALK chimeric protein and hence this subset of lymphomas. A subsequent study by Pittaluga et al.23 con¢rmed that immunohistochemistry allows the identi¢cation of a distinct subgroup of lymphomas that contain the 2p23 chromosomal abnormalities but lack ‘anaplastic’ morphology typical of ALCL. Furthermore, immunostaining proved to be the most sensitive method to identify this group. Through the use of these molecular and immunohistochemical techni-
CURRENT DIAGNOSTIC PATHOLOGY
ques, there has been a complete rethinking of the de¢nition of this tumour: as the histology may not be anaplastic and the cells neither anaplastic nor large, it has been suggested that the tumour should be renamed ‘ALK lymphoma’ or ‘ALKoma’ to re£ect this.24 Finally, variants of this disease with di¡erent translocations leading to di¡erent chimeric protein formation yielded a di¡erent pattern of immunostaining using antibodies to the ALK (or NPM) protein.25 This is perhaps the most profound demonstration to date of the power of immunohistochemistry to re£ect the diversity of genetic alteration in tumours. An example of a case of ALCL with ALK expression appears in Figure 3.
FLI-1 overexpression as an indicator of the t(11;22) translocation of PNET/ES Approximately 90% of primitive neuroectodermal tumours/Ewing’s sarcoma (PNET/ES) have a speci¢c t(11;22)(q24;q12), translocation, which results in the fusion of these genes on the der(22) chromosome, resulting in the production of a novel chimeric EWS/FLI-1 message, and overexpression of FLI-1 protein. While cytogenetic studies were originally employed to identify the presence of this chromosomal translocation, more recently identi¢cation of this genetic alteration has been possible through the use of appropriate primers using the RT-PCR technique.26 However, most recently it has been found that demonstration of the ‘molecular signature’ of this disease is possible by identifying the ‘neo-expression’ of the FLI-1 gene product using immunohistochemical techniques.27,28 In the study of Folpe and colleagues,28 and con¢rmed more recently by Llombart-Bosch and Navarro,29 FLI-1 expression has been demonstrated to be present in almost threequarters of PNET/ES, but not in other small blue round cell tumours with which it can be confused histologically. Immunohistochemical detection of FLI-1 may thus be a valuable technique for identi¢cation of PNET/ES in cases in which molecular genetic evaluation is not feasible. FLI-1 protein expression is also helpful in distinguishing PNET/ES from other tumours that may be CD99-positive.
WT-1 overexpression as an indicator of the t(11;22) translocation of DSRCT
Figure 3 Anaplastic large cell lymphoma showing positive immunostaining for ALK protein. Note nuclear and cytoplasmic localization of signal (original magni¢cation 400).
Another member of the small blue round cell group of paediatric tumours is the desmoplastic small round cell tumour (DSRCT); this distinctive tumour shows a di¡erent translocation involving chromosomes 11 and 22: t(11;22)(p13;q12). This results in unique chimeric EWS/ WT1 transcripts with expression of the corresponding fusion gene product.30 Since a portion of this fusion protein corresponds to the C-terminal portion of the
GENOGENIC IMMUNOHISTOCHEMISTRY: A NEW ERA IN DIAGNOSTIC IMMUNOHISTOCHEMISTRY
WT-1 gene product, it is not surprising that studies have demonstrated that antibodies to the WT-1gene product can detect the unique fusion product that results from this translocation, and can thus be used to identify desmoplastic small round cell tumour by immunohistochemistry.31^33 Indeed, among small blue round cell tumours (with the exception of Wilms tumour, which is generally con¢ned to the kidney), the DSRCT is unique in showing positive immunostaining for the WT-1 gene product, as shown in Figure 4.
USING IMMUNOHISTOCHEMISTRY TO IDENTIFY THERAPEUTIC TARGETS FOR CANCER HerceptinTM and HER-2/neu
197
among patients with metastatic disease, those eligible for immunotherapy treatment with the new humanized anti-HER-2/neu antibody, trastuzumab (HerceptinTM, Genentech, Inc. South San Francisco, CA). Determination of HER-2/neu status has thus taken on great importance in the routine work-up of breast cancer in the pathology laboratory, complementing but not replacing the traditional diagnostic and prognostic markers such as tumour type, tumour size, tumour grade and lymph node status. Because the alterations of HER-2/neu occur simultaneously at the genomic and protein levels, a number of studies have addressed the issue of comparative sensitivity and speci¢city of the two techniques. Controversy has ensued over the reproducibility and reliability of the semiquantitative IHC scoring system (0, 1+, 2+, and 3+) employed,47 and the use of FISH, for which binary results
The HER-2/neu (c-erbB-2) oncogene protein is a transmembrane glycoprotein in the epidermal growth factor receptor family. It is expressed at low levels in a variety of normal epithelia, including breast duct epithelium, but is ‘overexpressed’ at signi¢cantly higher levels in approximately one-¢fth to one-quarter of invasive breast cancers, usually as a direct consequence of ampli¢cation of the HER-2/neu gene.34 Many studies have now demonstrated the importance of HER-2/neu overexpression as an independent prognostic marker of clinical outcome, particularly in node-positive patients;35^37 recent studies also suggest that HER-2/neu status predicts response to adriamycin-based adjuvant chemotherapy.38 ^ 40 In addition, the predictive value of HER-2/neu overexpression in node-negative patients has been demonstrated,41 with the additional ¢nding that HER-2/neu overexpression is associated with poor response to tamoxifen, even in the setting of positive oestrogen receptor (ER) expression.41^ 44 Most recently,45,46 HER-2/neu protein overexpression in breast cancer has been used to select,
Figure 4 Desmoplastic small round cell tumour of childhood showing strong, uniform nuclear immunostaining with antibodies to the C-terminal portion of the Wilms tumour gene product (original magni¢cation 200).
Figure 5 In¢ltrating ductal carcinoma of the breast showing ampli¢cation of the HER-2/neu gene with concomitant protein overexpression. (A) Fluorescence in situ hybridization, showing multiple orange signals (corresponding to HER-2/neu gene) per tumour cell nucleus, with average of two green signals (corresponding to chromosome17 centromere) per nucleus. (B) Immunohistochemical demonstration of 3+ level of immunostaining for HER-2/neu gene product [original magni¢cations 1000 (A), 400 (B)].
198
are reported (ampli¢ed/non-ampli¢ed), has been recommended by some.48 While 2+ and 3+ immunostain scores have been considered as positive results, questions have persisted, particularly with respect to the signi¢cance of the 2+ level of immunostaining.49 Recent studies attempting to compare detection of protein overexpression (measured by IHC) to gene ampli¢cation (measured by FISH or PCR) showed concordance rates ranging from 78% to 98%,48 ^57 with most studies showing over 90% concordance rates. The recent study by Pauletti et al.52 is unique in ¢nding poor concordance, but there is reason to believe that there were problems with the IHC component of that study, in which only 17% of all cases were either 2+ or 3+ positive. In a more recent study from our laboratory, involving a series of 306 invasive breast cancer cases, we performed parallel IHC and FISH studies. Among the 173 cases negative for protein overexpression via IHC, only ¢ve cases (2.9%) showed gene ampli¢cation. Among the 21 cases showing a 3+ positive overexpression score, 100% showed gene ampli¢cation. Among the112 cases showing a 2+ positive immunostaining score, 21 (18.8%) showed gene ampli¢cation. However, among the 91 (81.2%) 2+ IHC cases negative for gene ampli¢cation, 46 (41%) showed an ampli¢cation score between 1 and 2. This suggests that immunohistochemical scores (i.e. 2+) may correlate with low levels of gene ampli¢cation, and provides further evidence to suggest that IHC, when employed with optimal antibodies and a scoring system, provides a highly accurate assessment of HER-2/neu gene status. Figure 5 compares HER-2/neu assessment by FISH and immunohistochemistry.
STI571 (GleevecTM; imatinib mesylate, formerly known as STI571) and gastrointestinal stromal tumours In the early 1990s, a drug, eventually called STI571, was developed that targeted the ‘kinase pocket’ of the BCRABL protein, the fusion gene product characteristic of chronic myeloid leukaemia (CML) cells.58,59 The BCRABL gene product manifests deregulated tyrosine kinase activity, which is instrumental in the pathogenesis of CML, and GleevecTM acts by blocking this tyrosine kinase activity. GleevecTM targets not only the BCR-ABL gene product, however, but also is e¡ective in blocking the ‘kinase pocket’of the platelet-derived growth factor receptor60 and c-kit.61 The latter is expressed in (and highly characteristic of) gastrointestinal stromal tumours (GIST), and recent published and presented data suggest that GleevecTM is a highly e¡ective reagent for the treatment of GISTs, a tumour heretofore without e¡ective therapy.61 Thus, immunohistochemical detection of the c-kit gene product can identify tumours that might be treatable with GleevecTM.62
CURRENT DIAGNOSTIC PATHOLOGY
CONCLUSION We are about to enter a new era of immunohistochemistry testing, moving from the current era in which this technique is used solely to determine the cell or tumour type. In this new era, which Albert Coons could not have foretold, immunohistochemistry will be used for the assessment of unique phenotypic aspects of the cell or tumour which directly re£ect the underlying genetic alterations of a given tumour and which will, with increasing frequency, determine the optimal treatment for the patient.
PRACTICE POINTS K
K
K
K
The range of diagnostic immunohistochemistry is being expanded to include the identi¢cation of changes at the protein level which re£ect molecular alterations thoughtto underlie speci¢c tumours Novel immunohistochemical tests now exist to identify, by virtue of the loss of immunostaining, lobular carcinoma of the breast and colonic adenocarcinomas such as those occurring in the context of HNPCC, which exhibit the molecular phenomenon of microsatellite instability Novel immunohistochemical tests now exist to identify the speci¢c chimeric proteins expressed as a consequence of chromosomal translocations characteristic of anaplastic large cell lymphoma, PNET/Ewings sarcoma, and desmoplastic small round cell tumour Novel immunohistochemical tests now exist to identify the targets of new therapies for cancer, e.g. theidenti¢cation ofthe HER-2/neu gene productin a subset of breast cancers
RESEARCH DIRECTIONS K
K
K
Developmentof antibodies to other proteins deleted as a consequence of tumour-speci¢c genetic mutations Development of antibodies to chimeric proteins expressed as a consequence of chromosomal translocations of other tumours Development of antibodies to other targets of novel cancer therapeutics
REFERENCES 1. Coons A H, Creech H J, Jones R N. Immunological properties of an antibody containing a £uorescent group. Proc Soc Exp Biol1941; 47: 200 ^202. 2. Moll R, Mitze M, Frixen U H, Birchmeier W. Di¡erential loss of E-cadherin expression in in¢ltrating ductal and lobular breast carcinomas. Am J Pathol 1993; 143: 1731^1742.
GENOGENIC IMMUNOHISTOCHEMISTRY: A NEW ERA IN DIAGNOSTIC IMMUNOHISTOCHEMISTRY
3. Lehr H A, Folpe A,Yaziji H, Kommoss F, Gown A M. Cytokeratin 8 immunostaining pattern and E-cadherin expression distinguish lobular from ductal breast carcinoma. Am J Clin Pathol 2000; 114: 190 ^196. 4. Kanai Y, Oda T, Tsuda H, Ochiai A, Hirohashi S. Point mutation of the E-cadherin gene in invasive lobular carcinoma of the breast. Jpn J Cancer Res1994; 85: 1035^1039. 5. Berx G,Cleton J-A M, Strumane K, de L-WJ, Nollet F, van R-F et al. E-cadherin is inactivated in a majority of invasive human lobular breast cancers by truncation mutations throughout its extracellular domain. Oncogene1996; 13: 1919^1925. 6. Droufakou S, DeshmaneV, Roylance R, Hanby A,Tomlinson I, Hart I R. Multiple ways of silencing E-cadherin gene expression in lobular carcinoma of the breast. Int J Cancer 2001; 92: 404 ^ 408. 7. Vos C B, Cleton J-A M, Berx G, de L-W J, ter H-N T, van R-F et al. E-cadherin inactivation in lobular carcinoma in situ of the breast: an early event in tumourigenesis. Br J Cancer 1997; 76: 1131^1133. 8. Goldstein N S, Bassi D, Watts J C, Lay¢eld L J,Yaziji H, Gown A M. E-cadherin reactivity of 95 noninvasive ductal and lobular lesions of the breast. Implications for the interpretation of problematic lesions. Am J Clin Pathol 2001; 115: 534 ^542. 9. JacobsT W, Pliss N, Kouria G, Schnitt S J. Carcinomas in situ of the breast with indeterminate features: role of E-cadherin staining in categorization. Am J Surg Pathol 2001; 25: 229^236. 10. Goldstein N S, Kestin L L,Vicini F A.Clinicopathologic implications of E-cadherin reactivity in patients with lobular carcinoma in situ of the breast. Cancer 2001; 92: 738 ^747. 11. Fearon E R, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990; 61: 759^767. 12. Marra G, Boland C R. Hereditary nonpolyposis colorectal cancer: the syndrome, the genes, and historical perspectives. J Natl Cancer Inst1995; 87: 1114 ^1125. 13. Peltomaki P, de la Chapelle A. Mutations predisposing to hereditary nonpolyposis colorectal cancer. Adv Cancer Res 1997; 71: 93^119. 14. Jass J R. Pathology of hereditary nonpolyposis colorectal cancer. Ann NYAcad Sci 2000; 910: 62^73; discussion 73^74. 15. Loeb L A. Microsatellite instability: marker of a mutator phenotype in cancer. Cancer Res 1994; 54: 5059^5063. 16. Loeb L A. A mutator phenotype in cancer. Cancer Res 2001; 61: 3230 ^3239. 17. Kim H, Piao Z, Kim J W,Choi J S, Kim N K, Lee J M et al. Expression of hMSH2 and hMLH1 in colorectal carcinomas with microsatellite instability. Pathol Res Practice 1998; 194: 3^9. 18. Cawkwell L, Gray S, Murgatroyd H, Sutherland F, Haine L, Longfellow M et al. Choice of management strategy for colorectal cancer based on a diagnostic immunohistochemical test for defective mismatch repair [see comments].Gut 1999; 45: 409^ 415. 19. Rimokh R, Magaud J P, Berger F, Samarut J, Coi⁄er B, Germain D et al. A translocation involving a speci¢c breakpoint (q35) on chromosome 5 is characteristic of anaplastic large cell lymphoma (‘Ki-1 lymphoma’). Br J Haematol 1989; 71: 31^36. 20. Morris S W, Kirstein M N, Valentine M B, Dittmer KG, Shapiro D N, Saltman D L et al. Fusion of a kinase gene, ALK, to a nucleolar protein gene, NPM, in non-Hodgkin’s lymphoma. Science1994; 263: 1281^1284. 21. Downing J R, Shurtle¡ S A, Zielenska M,Curcio-Brint A M, Behm F G, Head D R et al. Molecular detection of the (2;5) translocation of non-Hodgkin’s lymphoma by reverse transcriptase ^ polymerase chain reaction. Blood1995; 85: 3416 ^3422. 22. Shiota M, Nakamura S, Ichinohasama R, Abe M, Akagi T, Takeshita M et al. Anaplastic large cell lymphomas expressing the novel chimeric protein p80NPM/ALK: a distinct clinicopathologic entity. Blood1995; 86: 1954 ^1960. 23. Pittaluga S, Wiodarska I, Pulford K, Campo E, Morris S W, Van dB-H et al.The monoclonal antibody ALK1identi¢es a distinct mor-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
199
phological subtype of anaplastic large cell lymphoma associated with 2p23/ALK rearrangements. Am J Pathol 1997; 151: 343^351. Benharroch D, Meguerian-Bedoyan Z, Lamant L, Amin C, Brugieres L, Terrier-Lacombe M J et al. ALK-positive lymphoma: a single disease with a broad spectrum of morphology. Blood 1998; 91: 2076^2084. Falini B, Pulford K, Pucciarini A, Carbone A, De Wolf-Peeters C, Cordell J et al. Lymphomas expressing ALK fusion protein (s) other than NPM-ALK. Blood1999; 94: 3509^3515. Downing J R, Head D R, Parham D M, Douglass E C, Hulshof M G, Link M P et al. Detection of the (11;22)(q24;q12) translocation of Ewing’s sarcoma and peripheral neuroectodermal tumour by reverse transcription polymerase chain reaction. Am J Pathol 1993; 143: 1294 ^1300. Melot T,Gruel N, Doubeikovski A, Sevenet N,Teillaud JL, Delattre O. Production and characterization of mouse monoclonal antibodies to wild-type and oncogenic FLI-1proteins. Hybridoma1997; 16: 457^ 464. Folpe A L, Hill C E, Parham D M, O’Shea P A, Weiss S W. Immunohistochemical detection of FLI-1 protein expression: a study of 132 round cell tumours with emphasis on CD99-positive mimics of Ewing’s sarcoma/primitive neuroectodermal tumour. Am J Surg Pathol 2000; 24: 1657^1662. Llombart-Bosch A, Navarro S. Immunohistochemical detection of EWS and FLI-1 proteins in Ewing sarcoma and primitive neuroectodermal tumours: comparative analysis with CD99 (MIC-2) expression. Appl Immunohistochem Mol Morphol 2001; 9: 255^260. Gerald W L, Ladanyi M, de Alava E, Cuatrecasas M, Kushner B H, LaQuaglia M P et al. Clinical, pathologic, and molecular spectrum of tumours associated with t(11;22)(p13;q12): desmoplastic small round-cell tumour and its variants. Clin Oncol 1998; 16: 3028 ^3036. Barnoud R, Delattre O, Pe¤oc’h M, Pasquier D, Plantaz D, Leroux D et al. Desmoplastic small round cell tumour: RT-PCR analysis and immunohistochemical detection of the Wilm’s tumour gene WT1. Pathol Res Practice 1998; 194: 693^700. Charles A K, Moore I E, Berry P J. Immunohistochemical detection of the Wilms’ tumour gene WT1 in desmoplastic small round cell tumour. Histopathology1997; 30: 312^314. Ordonez N G. Desmoplastic small round cell tumour: II: an ultrastructural and immunohistochemical study with emphasis on new immunohistochemical markers. Am J Surg Pathol 1998; 22: 1314 ^1327. Slamon D J, Clark G M, Wong S G, Levin W J, Ullrich A, McGuire W L. Human breast cancer: correlation of relapse and survival with ampli¢cation of the HER-2/neu oncogene. Science 1987; 235: 177^182. Toikkanen S, Helin H, Isola J, Joensuu H. Prognostic signi¢cance of HER-2 oncoprotein expression in breast cancer: a 30 -year followup [see comments]. J Clin Oncol 1992; 10: 1044 ^1048. Paik S, Hazan R, Fisher E R, Sass R E, Fisher B, Redmond C et al. Pathologic ¢ndings from the National Surgical Adjuvant Breast and Bowel Project: prognostic signi¢cance of erbB-2 protein overexpression in primary breast cancer [Prior annotation incorrect]. J Clin Oncol 1990; 8:103^112. Tandon A K, Clark G M, Chamness G C, Ullrich A, McGuire W L. HER-2/neu oncogene protein and prognosis in breast cancer. J Clin Oncol 1989; 7: 1120 ^1128. Muss H B, Thor A D, Berry D A, KuteT, Liu E T, Koerner F et al. c-erbB-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994; 330: 1260 ^1266. Thor A D, Berry D A, Budman D R, Muss H B, KuteT, Henderson I C et al. erbB-2, p53, and e⁄cacy of adjuvant therapy in lymph node-positive breast cancer [see comments]. J Natl Cancer Inst 1998; 90: 1346 ^1360.
200
40. Paik S, Bryant J, Park C, Fisher B, Tan-Chiu E, Hyams D et al. erbB-2 and response to doxorubicin in patients with axillary lymph node-positive, hormone receptor-negative breast cancer [see comments]. J Natl Cancer Inst 1998; 90: 1361^1370. 41. Sjogren S, Inganas M, Lindgren A, Holmberg L, Bergh J. Prognostic and predictive value of c-erbB-2 overexpression in primary breast cancer, alone and in combination with other prognostic markers. J Clin Oncol 1998; 16: 462^ 469. 42. Wright C, Nicholson S, Angus B, Sainsbury J R, Farndon J, Cairns J et al. Relationship between c-erbB-2 protein product expression and response to endocrine therapy in advanced breast cancer. Br J Cancer 1992; 65: 118 ^121. 43. Carlomagno C, Perrone F, Gallo C, De Laurentiis M, Lauria R, Morabito A et al. c-erb B2 overexpression decreases the bene¢t of adjuvant tamoxifen in early-stage breast cancer without axillary lymph node metastases. J Clin Oncol 1996; 14: 2702^2708. 44. Te“tu B, Brisson J, PlanteV, Bernard P. p53 and c-erbB-2 as markers of resistance to adjuvant chemotherapy in breast cancer. Mod Pathol 1998; 11: 823^ 830. 45. Shak S. Overview of the trastuzumab (Herceptin) anti-HER2 monoclonal antibody clinical program in HER2-overexpressing metastatic breast cancer. Herceptin Multinational Investigator Study Group. Semin Oncol 1999; 26 (4 Suppl 12): 71^77. 46. Slamon D J, Leyland Jones B, Shak S, Fuchs H, PatonV, Bajamonde A et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344: 783^792. 47. JacobsT W, Gown A M,Yaziji H, Barnes M J, Schnitt S J. Speci¢city of HercepTest in determining HER-2/neu status of breast cancers using the United States Food and Drug Administration-Approved Scoring System. J Clin Oncol 1999; 17: 1983. 48. Tubbs R R, Pettay J D, Roche P C, Stoler M H, Jenkins R B,GroganT M. Discrepancies in clinical laboratory testing of eligibility for trastuzumab therapy: apparent immunohistochemical false-positives do not get the message. J Clin Oncol 2001; 19: 2714 ^2721. 49. Birner P, Oberhuber G, Stani J, Reithofer C, Samonigg H, Hausmaninger H et al. Evaluation of the United States food and drug administration-approved scoring and test system of her-2 protein expression in breast cancer. Clin Cancer Res 2001; 7: 1669^1675. 50. JacobsT W,Gown A M,Yaziji H, Barnes M J, Schnitt S J.Comparison of £uorescence in situ hybridization and immunohistochemistry for the evaluation of HER-2/neu in breast cancer. J Clin Oncol 1999; 17: 1974 ^1982. 51. Wang S, Saboorian M H, Frenkel E, Hynan L, Gokaslan S T, Ashfaq R. Laboratory assessment of the status of Her-2/neu protein and
CURRENT DIAGNOSTIC PATHOLOGY
52.
53.
54.
55.
56.
57.
58.
59. 60.
61.
62.
oncogene in breast cancer specimens: comparison of immunohistochemistry assay with £uorescence in situ hybridisation assays. J Clin Pathol 2000; 53: 374 ^381. Pauletti G, Dandekar S, Rong H, Ramos L, Peng H et al. Assessment of methods for tissue-based detection of the HER-2/neu alteration in human breast cancer: a direct comparison of £uorescence in situ hybridization and immunohistochemistry. J Clin Oncol 2000; 18: 3651^3664. Kakar S, Puangsuvan N, Stevens J M, Serenas R, Mangan G, Sahai S et al. HER-2/neu assessment in breast cancer by immunohistochemistry and £uorescence in situ hybridization: comparison of results and correlation with survival. Mol Diagn 2000; 5: 199^207. Couturier J,Vincent-Salomon A, Nicolas A, Beuzeboc P, Mouret E, Zafrani B et al. Strong correlation between results of £uoresence in situ hybridization and immunohistochemistry for the assessment of ERBB2 (HER-2/neu) gene status in breast carcinoma. Mod Pathol 2000; 13: 1238 ^1243. Hoang M P, Sahin A A, Ordonez N G, Sneige N. HER-2/neu gene ampli¢cation compared with HER-2/neu protein overexpression and interobserver reproducibility in invasive breast carcinoma. Am J Clin Pathol 2000; 113: 852^ 859. Ridol¢ R L, Jamehdor M R, Arber J M. HER-2/neu testing in breast carcinoma: a combined immunohistochemical and £uorescent in situ hybridization approach. Mod Pathol 2000; 13: 866^ 873. Lebeau A, Deimling D, Kaltz C, Sendelhofert A, I¡ A, Luthardt B et al. HER-2/neu analysis in archival tissue samples of human breast cancer: comparison of immunohistochemistry and £uorescence in situ hybridization. J Clin Oncol 2001; 19: 354 ^363. Buchdunger E, Matter A, Druker B J. Bcr-Abl inhibition as a modality of CML therapeutics. Biochim Biophys Acta 2001; 1551: M11^M18. O’Dwyer M E, Druker B J. Chronic myelogenous leukaemiaFnew therapeutic principles. J Intern Med 2001; 250: 3^9. Pietras K, Ostman A, Sjoquist M, Buchdunger E, Reed R K, Heldin C H et al. Inhibition of platelet-derived growth factor receptors reduces interstitial hypertension and increases transcapillary transport in tumours. Cancer Res 2001; 61: 2929^2934. Joensuu H, Roberts P J, Sarlomo-Rikala M, Andersson L C, Tervahartiala P, Tuveson D et al. E¡ect of the tyrosine kinase inhibitor STI571in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 2001; 344: 1052^1056. Berman J, O’Leary T J. Gastrointestinal stromal tumor workshop. Hum Pathol 2001; 32: 578 ^582.