Introduction
Endometria Intraepithelial Neoplasia (EIN) is a clonal proliferation of architecturally and cytologically altered premalignant endometrial glands which are prone to malignant transformation to endometrioid (Type I) endometrial adenocarcinoma. EIN lesions are non-invasive genetically altered neoplasms which arise focally, and may convert to malignant phenotype upon acquisition of additional genetic damage. Diagnostic criteria for EIN have been developed by histopathologic correlation with clinical outcomes, molecular changes, and objective computerized histomorphometry.

EIN should not be confused with unrelated serous intraepithelial carcinoma (serous EIC), which is an early phase of (Type II) papillary serous adenocarcinomas of the endometrium.

Management of EIN lesions follows guidelines long established for atypical endometrial hyperplasia. A high concurrent cancer rate (26%), and concern that sampling errors may miss an occult tumor, have led to a prevailing view that immediate hysterectomy is justified by its combined diagnostic and therapeutic benefits. Young patients wishing to preserve fertility, and women who are poor surgical risks, are candidates for hormonal (progestin) therapy. Systemic progestins can successfully ablate up to 90% of endometrial precancers in young women [1] , although it is not possible in advance to predict that fraction which will respond. A decision to treat hormonally must thus be made between the clinician and patient in full light of the risks, and with the precondition that regular followup surveillance can be performed.

A Combined Molecular and Histopathologic Model for EIN
Latent, premalignant, and malignant phases of EIN-mediated endometrial carcinogenesis are diagrammed in Figure 1. In almost half of apparently normal women, histologically unremarkable proliferative endometria contain a small fraction of (PTEN tumor suppressor gene) mutant endometrial glands. This phase may be construed as "latent" because not only do the mutated glands look completely normal under the microscope, but they progress to EIN and cancer at very low efficiency. This latent phase may persist for years, with continued presence of scattered and interspersed mutant glands after many menstrual cycles [2] . Mutant glands are probably represented in the reserve population of cells that regenerate a new functionalis each month. Endocrine factors act upon these "latent precancers" to modulate involution, or progression to EIN. Transition to EIN requires accumulation of additional genetic damage in at least one "latent precancer" cell, which then clonally expands from its point of origin (indicated by expanding arrows) to form a contiguous grouping of a tightly packed and cytologically altered glands recognizable as EIN. The monoclonal precancer (EIN) develops internal heterogeneity through mutation, and advantageous events selected by local conditions result in hierarchical subclones (left to right) of varying success. EIN lesions have only marginal increases in growth potential, and retain susceptibility to further growth modulation by hormonal factors. Some involute. Others, through additional mutation and selection, reach a stage where hormonal support is no longer required for survival. Malignant transformation to cancer is defined by accumulation of sufficient genetic damage to permit invasion of adjacent stromal tissues.

The Foundations of EIN
Endometrial Intraepithelial Neoplasia has its origins in molecular genetic, experimental, clinical, and histopathologic studies. Each of these approaches has provided one facet of a multidisciplinary perspective on this disease [3] .

Precancer Criteria fulfilled for EIN.
Much as the German pathologist Koch developed a series of postulates that must be fulfilled in order to scientifically prove pathogenesis of disease by a specific infectious organism, similar postulates may be formulated for premalignant disease. We here list predictions to be met for a clinically relevant and biologically distinctive category of endometrial precancers, all of which have been fulfilled in the case of EIN.

Precancer Postulate 1:
EIN differs from normal tissues . EIN are bona fide neoplasms, comprised of a monoclonal outgrowth of a single transformed cell [4, 5] from a polyclonal source field. These benign expansile clones have only a marginal advantage beyond normal endometrial tissues, and in the absence of additional genetic damage lack the ability to invade or metastasize. Lesions with microsatellite instability have marker genotypes different than normal source tissues [6] .

Precancer Postulate 2:
EIN shares some, but not all features with carcinoma. Cells in the early stages of endometrial carcinogenesis should have some features which distinguish them from normal tissues, and whose retention during progression establishes them as the physical progenitors of carcinoma. Both EIN and endometrial carcinoma are monoclonal lesions, and those markers characteristic of monoclonality (nonrandom inactivation of a particular X chromosome copy, presence of a particular altered microsatellite) are conserved between the EIN and carcinoma lesions of individual patients [4, 5, 6, 7, 8, 9] . Genetic alteration of specific genes implicated in endometrial carcinogenesis has been shown to be conserved between EIN and carcinomas which occur in individual patients. This is true for inactivation of the PTEN tumor suppressor gene [10, 11, 12] , mutation of the KRAS oncogene [13, 14, 15] , epigenetic inactivation of the DNA repair gene MLH1 [16] . 63% of EIN lesions, for example, have lost the ability to express the tumor suppressor protein from the PTEN gene, a phenotype shared with over 80% of endometrial cancers [2, 12] .

Precancer Postulate 3:
EIN can be diagnosed. Diagnostic criteria applicable to a routine pathology practice are presented below [18, 19] . Additionally, there is an objective reference standard for EIN diagnosis in computerized morphometry of H&E stained slides [7] . Cytologic and architectural characteristics of H&E stained tissues are measured to calculate a "D-Score" which indicates EIN when less than a threshold of 1 [7, 20] .

Precancer Postulate 4:
EIN increases risk for carcinoma [21, 22] . Available clinical outcome studies have applied image analysis of pathologic endometria to identify subsets of women with EIN and correlated this diagnosis with future or concurrent carcinoma [20, 23, 24, 25] . 26% of women diagnosed with EIN already have cancer at the time of diagnosis, and the remainder have a 46 fold elevated cancer risk in the ensuing years [21] .

Precancer Postulate 5:
Genetic and hormonal mechanisms of carcinogenesis converge in EIN . Endometrial expression of the tumor suppressor gene PTEN normally increases in an estrogenic environment [26] . This functional requirement for increased tumor suppression activity of PTEN under estrogen rich conditions cannot be met in PTEN-defective EIN lesions. Thus, most EIN lesions (those 63% with lost PTEN protein) will have a defective tumor-suppressor response to estrogens. Correspondingly, if the mitogenic effects of estrogens are mitigated by progestins, PTEN mutant endometrial glands undergo selective involution relative to PTEN-intact glands [27] .

Precancer Postulate 6:
Introducing EIN genotype into an animal produces premalignant lesions and heightened cancer risk. 63% of EIN lesions are comprised of cells which are defective in production of the PTEN tumor suppressor gene product [2] . Heterozygote PTEN mutant mice uniformly (100%) develop endometrial "hyperplasia," and 21% of these progress to carcinoma [28] .

PTEN Tumor Supressor: A Biomarker for Endometrial Carcinogenesis
Anti-PTEN antibodies are the first commercially available "special stain" for neoplastic endometrioid (Type I) endometrial disease. Paraffin tissue immunohistochemistry with anti-PTEN antibody 6H2.1 (Cascade Biosciences, Winchester, MA) [12, 29] shows that over half of endometrioid endometrial adenocarcinomas and their precursor EIN lesions have lost PTEN protein due to genomic mutational or deletional inactivation [2, 12] . Many commercial anti-PTEN antibodies do not work on paraffin embedded tissues. A working protocol for successful use of antibody 6h2.1 is available online at www.endometrium.org. The most common problems are inadequate antigen retrieval, failure to incubate primary antibody overnight at 4°C, too low an antibody titration (use 1:100 for older blocks), use of old sections (must be cut and used within days), lack of adequate controls to recognize when it is working, and use of dark obscuring counterstains.

Caution is advised in using PTEN immunohistochemistry to make diagnoses of individual patients. It is an insensitive (half of EIN lesions have normal PTEN expression) and nonspecific (over a third of normal proliferative and anovulatory endometria contain PTEN-null glands) marker for EIN. PTEN expression is greatly affected by the endocrine and clinical context in which it is applied. Normal endometrial PTEN expression is greatly reduced or lost in secretory and atrophic glands [26] . Absence of expression in these circumstances cannot be equated with loss of function. Correspondingly, an estrogenic environment increases endometrial stromal and normal gland PTEN expression, improving contrast with a localizing PTEN-mutant clone.

PTEN immunohistochemistry can, however, be a useful educational tool for pathologists to delineate extent and configuration of mutant clones, and relate these features to routinely stained slides. Figure 2 shows a hematoxylin and eosin stained EIN lesion with matching PTEN immunohistochemistry images to delimit clonal precancerous lesions with single cell resolution. This educational objective can be segregated from the diagnostic process by setting aside interesting cases and undertaking a batch staining after they have already been signed out. This provides the benefit of comparison of multiple examples without compromising patient care. PTEN immunohistochemistry is not a routine or requisite part of EIN diagnosis.

Diagnosis OF EIN
A framework for EIN Diagnosis is shown in Table 1. Notable is the clear separation of endometrial changes caused by unopposed estrogens, and carcinoma, from EIN.

Table 1 - EIN Diagnostic Schema

EIN Nomenclature	Topography	Functional Category	Treatment
Benign Architectural Changes of Unopposed Estrogen	Diffuse	Estrogen Effect	Hormonal therapy
EIN Endometrial Intraepithelial Neoplasia	Focal progressing to diffuse	Precancer	Hormonal or surgical
Carcinoma	Focal progressing to diffuse	Cancer	Surgical stage-based

Topography of EIN
The distribution of a lesion is useful in distinguishing between the diffuse, field-wide effects, of an abnormal hormonal environment (anovulation, or persistent estrogen effect), surface changes secondary to stromal breakdown, and more focal EIN. Clonal origin from a single cell requires EIN lesions to begin as local processes within the endometrial compartment. Early EIN lesions are easily diagnosed by their contrast in architecture and cytology with the background from which they have emerged. Over time, EIN lesions may completely overrun the background endometrium, thereby removing the convenient lesion-to-background contrast in morphology which assist in EIN diagnosis. For this reason, or because of fragmentation, many EIN lesions must be diagnosed without the benefit of comparison with companion benign tissues. Exclusion of artifact and careful evaluation of the architectural and cytologic features of EIN usually permits accurate diagnosis in these instances.

EIN Diagnostic Criteria
All of the diagnostic criteria of Table 2, listed as A-E below, must be met in order to make an EIN diagnosis. The entire slide should first be scrutinized under low magnification for localizing lesions, and if found, these areas examined under higher power to assess possible changes in cytology within the architecturally distinct focus. Widespread EIN lesions that have replaced the entire endometrial compartment tend to have a sufficiently atypical cytology that background normal endometrium is no longer required as a reference point for accurate diagnosis.

Size, architecture, and cytology features are easy EIN diagnostic criteria. Much more difficult are exclusion of benign mimics and adenocarcinoma from the differential diagnosis. There are no simple rules for benign mimic exclusion. The broad universe of competing entities can only be recognized on sight by one who has the easy familiarity that comes with experience. Consistent demarcation of the EIN-adenocarcinoma threshold remains important clinically because it provides a basis for the clinician to evaluate the risks of electing hormonal rather than surgical therapy in younger patients who wish to retain fertility.

Special diagnostic challenges, such as recognition of EIN within polyps, interpretation of subdiagnostically small or fragmented lesions, and interpretation of lesions with non-endometrioid differentiation have specific caveats presented below that should be carefully studied.

Table 2 - EIN Diagnostic Criteria. Modified after [17]

EIN Criterion	Comments
Architecture	Area of Glands greater than Stroma
Cytology	Cytology differs between architecturally crowded focus and background, or clearly abnormal.
Size >1 mm	Maximum linear dimension exceeds 1mm.
Exclude mimics	Benign conditions with overlapping criteria: Basalis, secretory, polyps, repair, etc..
Exclude Cancer	Carcinoma if mazelike glands, solid areas, polygonal "mosaic-like" glands, myoinvasion, or significant cribriforming

A. Architecture: Gland area exceeds stromal area:
A cardinal architectural feature of endometrial precancers is glandular crowding, with a threshold quantitative cutoff for EIN lesions of less than half of the tissue area occupied by stroma (Volume Percentage Stroma). Areas with large dominant cysts should always be avoided in making this assessment. Although EIN is an epithelial disease, visual assessment of the glands themselves is complicated by frequent artifactual displacement from associated stroma, pale staining of most epithelia, and visual "shimmering" between gland epithelia and lumens. These may all be avoided by focusing on the stromal compartment which has the significant advantages of a more uniform composition throughout the specimen, and superior staining qualities. By focusing on the stroma itself only intact fragments in which stroma has not been avulsed from glands will be evaluated.

Careful review of graphic and histologic examples of varying stromal densities will assist in training your eye to classify patient material as above or below the diagnostic threshold. EIN lesions tend to cluster with a median volume percentage stroma of about 40% and non-EIN (benign) lesions cluster at a median of approximately 75%. These differences are sufficiently great that visual assessment by a trained eye can be informative.

B. Cytology of architecturally crowded area is different than background, or clearly abnormal:
There is no absolute standard for cytologic features of EIN lesions, but the cytology of EIN is usually clearly demarcated as divergent from that of co-existing benign endometrial tissues in the same patient. The manner of cytologic change in EIN varies considerably from patient to patient, and can include but not be limited to, increased variation in nuclear size and contour, clumped or granular chromatin texture, change in nucleoli, change in nuclear/cytoplasmic ratio, and altered cytoplasmic differentiation. Stereotypical static descriptions of cytologic atypia, such as nuclear rounding and appearance of nucleoli are met in many but not all EIN lesions. In this sense, a fixed presentation of cytologic atypia is not a prerequisite for EIN. Attempts to define an absolute standard are confounded by the extreme morphologic plasticity of endometrial glandular cells under changing hormonal, repair, and differentiation conditions.

Cytologic changes in some EIN lesions are manifest as a change in differentiation state to a tubal, mucinous, micropapillary, or eosinophilic phenotype. These must be distinguished from the scattered random pattern of hormonally, or surface located repair-induced "metaplasias." Further details of how to interpret non-endometrioid EIN lesions are presented in the "Pitfalls" section below.

In those cases with no normal glands for internal reference, it is necessary to assess the freestanding cytology of relevant fragments in the context of their architectural features. Some EIN lesions occupy the entire tissue sample, and should not be underdiagnosed for lack of a convenient benign gland in the area.

C. Size >1mm in maximum dimension:
Accurate EIN diagnosis requires a contiguous field of glands sufficiently large to enable reliable assessment of architecture. A minimum lesion size of 1 mm maximum dimension was required in the previous clinical outcome studies [20, 23, 25, 30] for an EIN lesion to achieve elevated cancer risk. That area of an EIN lesion which meets architectural (gland area) and cytologic (changed) criteria for diagnosis must measure a minimum of 1mm in maximum dimension, a scale which usually encompasses more than 5-10 glands. Most biopsy formats produce tissue fragments in excess of 1.5-2mm. The size requirement must be met in a single tissue fragment, not added amongst multiple fragments. There is no formal evidence that once beyond the minimum 1mm, EIN lesions should be stratified by size, but if a lesion is discretely focal, it may be of interest to the clinician to know what fraction of the available curettings contain lesion.

Individual or small clusters of cytologically altered glands have an undefined natural history and are best diagnosed descriptively (See Pitfalls section below).

D. Exclusion of Benign Mimics
Patients with one of the conditions listed below may still have an EIN, but this diagnosis should be made with careful consideration into how the coexisting factor(s) may modify the criteria for EIN diagnosis. If a specimen is refractory to confident diagnosis, a comment as to the nature of the problem may be useful in directing management.

1. Reactive changes caused by infection, physical disruption, recent pregnancy, or recent instrumentation. These can cause piling up of the epithelium, and loss of nuclear polarity..
   
   
2. Artifactual gland displacement. Beware diagnosing an EIN lesion if the cytology is identical between areas with crowded compared to uncrowded glands! Many of these are artifactual disruptions where the stroma is sheared and glands pushed in apposition .
   
   
3. Persistent Estrogen Effect: Randomly scattered cysts of protracted estrogen exposure and occasional branching glands are commonly encountered in anovulatory or estrogen-exposed endometria. Gland density is uniformly irregular throughout the endometrial compartment, with occasional clusters of glands having a cytology identical to the uncrowded areas. These can be diagnosed as "proliferative endometrium with architectural changes of unopposed estrogens". With increasing duration, microthrombi form and scattered stromal breakdown may be associated with epithelial piling along the collapsed stromal surfaces.
   
   
4. Mid to late secretory endometrium displays loss of nuclear polarity, nuclear enlargement, and variation in nuclear size which if measured objectively by computerized morphometry overlaps substantially with EIN lesions. Stromal responsiveness to progesterone is not homogenous at all endometrial depths. Lack of stromal pre-decidualization in the deeper functionalis and superficial basalis makes glands appear crowded, and these same glands may display a worrisome cytology and complicated saw-toothed luminal profiles
   
   
5. Endometrial polyps contain irregularly spaced glands in which scattered glands may differ from native endometrium due to their tendency to have reduced hormonal responsiveness. Benign polyps may also have low volume percentage stroma caused by cysts (senile polyps) or random aggregations of glands. Approximately 10% of EIN lesions, however, will present within an endometrial polyp and these must be diagnosed as described below in the "Pitfalls" section.
   
   
6. Endometrial breakdown is one of the most common settings for overdiagnosis of a benign endometrium as a precancer or cancer. Breakdown may follow an ovulatory or anovulatory cycle and persist into the transitional period between late menses and early proliferative endometrium. Altered cytology is due to piling up of epithelial cells unsupported by stroma, and associated nuclear changes such as loss of polarity which may be accentuated under certain fixation conditions which exaggerate chromatin texture (Bouin's fixative).

1. Meandering or "mazelike" lumens
   
   
2. Solid epithelium
   
   
3. Cribriform architecture.
   
   
4. "Mosaic" gland pattern of distorted polygonal glands with threadlike intervening stroma

Acknowledgements and Credits
Some of the text and many of the Figures have been used, with permission, from the website www.endometrium.org [31] . All rights reserved by G. Mutter.

References

1. Randall TC, Kurman RJ. Progestin treatment of atypical hyperplasia and well-differentiated carcinoma of the endometrium in women under age 40. Obstet Gynecol 1997; 90(3):434-440.
   
   
2. Mutter GL, Ince TA, Baak JPA, Kust G, Zhou X, Eng C. Molecular identification of latent precancers in histologically normal endometrium. Cancer Res 2001; 61:4311-4314.
   
   
3. Baak JP, Mutter GL. EIN and WHO94. J Clin Pathol 2005; 58(1):1-6.
   
   
4. Jovanovic AS, Boynton KA, Mutter GL. Uteri of women with endometrial carcinoma contain a histopathologic spectrum of monoclonal putative precancers, some with microsatellite instability. Cancer Res 1996; 56:1917-1921.
   
   
5. Mutter GL, Chaponot M, Fletcher J. A PCR assay for non-random X chromosome inactivation identifies monoclonal endometrial cancers and precancers. Am J Pathol 1995; 146:501-508.
   
   
6. Mutter GL, Boynton KA, Faquin WC, Ruiz RE, Jovanovic AS. Allelotype mapping of unstable microsatellites establishes direct lineage continuity between endometrial precancers and cancer. Cancer Res 1996; 56:4483-4486.
   
   
7. Mutter GL, Baak JPA, Crum CP, Richart RM, Ferenczy A, Faquin WC. Endometrial precancer diagnosis by histopathology, clonal analysis, and computerized morphometry. J Pathol 2000; 190:462-469.
   
   
8. Esteller M, Garcia A, Martinez-Palones JM, Xercavins J, Reventos J. Detection of clonality and genetic alterations in endometrial pipelle biopsy and its surgical specimen counterpart. Lab Invest 1997; 76:109-116.
   
   
9. Faquin WC, Fitzgerald JT, Boynton KA, Mutter GL. Intratumoral genetic heterogeneity and progression of endometrioid type endometrial adenocarcinomas. Gynecol Oncol 2000; 78:152-157.
   
   
10. Levine RL, Cargile CB, Blazes MS, Van Rees B, Kurman RJ, Ellenson LH. PTEN mutations and microsatellite instability in complex atypical hyperplasia, a precursor lesion to uterine endometrioid carcinoma. Cancer Res 1998; 58:3254-3258.
    
    
11. Maxwell G, Risinger J, Gumbs C, Shaw H, Bentley R, Barrett J et al. Mutation of the PTEN tumor supressor gene in endometrial hyperplasias. Cancer Res 1998; 58:2500-2503.
    
    
12. Mutter GL, Lin MC, Fitzgerald JT, Kum JB, Baak JPA, Lees J et al. Altered PTEN expression as a diagnostic marker for the earliest endometrial precancers. J Natl Cancer Inst 2000; 92:924-930.
    
    
13. Duggan BD, Felix JC, Muderspach LI, Tsao J-L, Shibata DK. Early mutational activation of the c-Ki-ras oncogene in endometrial carcinoma. Cancer Res 1994; 54:1604-1607.
    
    
14. Sasaki H, Nishii H, Takahashi H, Tada A, Furusato M, Terashima Y et al. Mutation of the Ki-ras protooncogene in human endometrial hyperplasia and carcinoma. Cancer Res 1993; 53:1906-1910.
    
    
15. Mutter GL, Wada H, Faquin W, Enomoto T. K-ras mutations appear in the premalignant phase of both microsatellite stable and unstable endometrial carcinogenesis. Mol Pathol 1999; 52:257-262.
    
    
16. Esteller M, Catasus L, Matias-Guiu X, Mutter GL, Prat J, Baylin SB et al. hMLH1 Promoter Hypermethylation Is an Early Event in Human Endometrial Tumorigenesis. Am J Pathol 1999; 155(5):1767-1772.
    
    
17. Silverberg SG, Mutter GL, Kurman RJ, Kubik-Huch RA, Nogales F, Tavassoli FA. Tumors of the uterine corpus: epithelial tumors and related lesions. In: Tavassoli FA, Stratton MR, editors. WHO Classification of Tumors: Pathology and Genetics of Tumors of the Breast and Female Genital Organs. Lyon, France: IARC Press, 2003: 221-232.
    
    
18. Mutter GL. Diagnosis of premalignant endometrial disease. J Clin Pathol 2002; 55(5):326-331.
    
    
19. Mutter GL, The Endometrial Collaborative Group. Endometrial intraepithelial neoplasia (EIN): Will it bring order to chaos? Gynecol Oncol 2000; 76:287-290.
    
    
20. Baak JPA, Nauta J, Wisse-Brekelmans E, Bezemer P. Architectural and nuclear morphometrical features together are more important prognosticators in endometrial hyperplasias than nuclear morphometrical features alone. J Pathol 1988; 154:335-341.
    
    
21. Baak JPA, Mutter GL, Robboy S, van Diest PJ, Uyterlinde AM, Orbo A et al. In endometrial hyperplasias, the molecular-genetics and morphometry-based EIN classification more accurately predicts cancer-progression than the WHO94. Cancer 2004; In Press.
    
    
22. Hecht JL, Ince TA, Baak JP, Baker HE, Ogden MW, Mutter GL. Prediction of endometrial carcinoma by subjective endometrial intraepithelial neoplasia diagnosis. Mod Pathol 2004; In Press.
    
    
23. Dunton C, Baak J, Palazzo J, van Diest P, McHugh M, Widra E. Use of computerized morphometric analyses of endometrial hyperplasias in the prediction of coexistent cancer. Am J Obstet Gynecol 1996; 174:1518-1521.
    
    
24. Orbo A, Baak JP, Kleivan I, Lysne S, Prytz PS, Broeckaert MA et al. Computerised morphometrical analysis in endometrial hyperplasia for the prediction of cancer development. A long-term retrospective study from northern Norway. J Clin Pathol 2000; 53(9):697-703.
    
    
25. Baak JP, Orbo A, van Diest PJ, Jiwa M, de Bruin P, Broeckaert M et al. Prospective multicenter evaluation of the morphometric D-score for prediction of the outcome of endometrial hyperplasias. Am J Surg Pathol 2001; 25(7):930-935.
    
    
26. Mutter GL, Lin MC, Fitzgerald JT, Kum JB, Ziebold U, Eng C. Changes in endometrial PTEN expression throughout the human menstrual cycle. J Clin Endocrinol Metab 2000; 85:2334-2338.
    
    
27. Zheng W, Baker HE, Mutter GL. Involution of PTEN-Null Endometrial Glands with Progestin Therapy. Gynecol Oncol 2004; 92:1008-1013.
    
    
28. Stambolic V, Tsao MS, Macpherson D, Suzuki A, Chapman WB, Mak TW. High incidence of breast and endometrial neoplasia resembling human Cowden syndrome in pten+/- mice. Cancer Res 2000; 60(13):3605-3611.
    
    
29. Perren A, Weng L, Boag A, Ziebold U, Thakore K, Dahia P et al. Immunocytochemical evidence of loss of PTEN expression in primary ductal adenocarcinomas of the breast. Am J Pathol 1999; 155:1253-1260.
    
    
30. Baak JPA, Mutter GL, Robboy S, van Diest PJ, Uyterlinde AM, Orbo A et al. The molecular genetics and morphometry-based Endometrial Intraepithelial Neoplasia (EIN) classification scheme predicts the risk of progression to cancer more effectively than the WHO94 "Endometrial hyperplasia" system. Am J Obstet Gynecol 2004; Submitted.
    
    
31. Mutter GL. www.endometrium.org : A focus on premalignant lesions of the endometrium. www.endometrium.org . 2003.