A comparative morphometric and cytophotometric study of endometrial hyperplasia, atypical hyperplasia, and endometrial carcinoma

On the concept of objectivity in digital image analysis in pathology

AIMS

The term 'objective' connotes a method that is based on facts and not influenced by personal opinions, perception or emotion. One often reads in the biomedical literature claims of objectivity for methods that use digital image analysis applied to histology. Since objective assessment of histology would represent a huge leap forward in scientific measurement and clinical diagnosis, such claims should be substantiated by strong evidence. This paper takes a selective look at the literature on image analysis to assess the definition of objectivity in image analysis and asks whether such a claim is ever justified.

METHODS

First, a brief background on the basic science of image analysis in histology details some of the controversies and opinions in the field. Then, a literature review of a subset of papers pertaining to image analysis in histology (with claims of objectivity) is conducted to determine what evidence exists for objectivity in these methods.

RESULTS

It was found that image analysis may have many benefits (speed, indefatigability, standardisation, etc.). However, algorithms are devised and implemented by human beings who make subjective decisions at each stage of the algorithm design and implementation process. Thus, image analysis methods can be seen as deterministic processes which 'objectively' implement the subjective decisions of the programmer. This indicates that 'inter-observer' variation in image analysis is equivalent to 'inter-algorithm' variation (which is rarely studied) and that a single computer algorithm's repeatability is of lesser importance than the repeatability of the image analysis method as a whole (including the block, slide and field selection and the method of tissue processing).

CONCLUSION

Repeatability and automaticity must not be confused with objectivity, but a lack of objectivity does not imply a lack of utility. Unless specific evidence of objectivity is provided, editors should insist that claims of objectivity in image analysis papers be either removed or justified prior to publication.

Description of a Novel System for Grading of Endometrial Carcinoma and Comparison With Existing Grading Systems

The most widely used system for grading of endometrial carcinoma is the International Federation of Gynecology and Obstetrics (FIGO) grading system. This grading system requires evaluation of histologic features that are difficult to assess reproducibly. Two hundred and two cases of endometrial carcinoma, treated by hysterectomy, were retrieved from the archives of Vancouver General Hospital (1983-1998). For each tumor, the architectural pattern, nuclear grade, and mitotic index were assessed. The tumor architectural pattern, nuclear grade, and mitotic index were significant predictors of patient outcome (P < 0.0001 for each, by univariate analysis). There were no prognostic differences between patients having predominantly solid versus papillary tumors, or tumors with mild versus moderate nuclear atypia. The tumors were then classified into high and low grade based on assessment of these three features. The presence of at least two criteria of these three: 1) predominantly papillary or solid growth pattern, 2) mitotic index > or =6/10 high power fields, or 3) severe nuclear atypia, resulted in a tumor being considered high grade. Low-grade tumors satisfied at most one of those criteria. The proposed grading system was found to be an independent predictor of patient outcome when patient survival was adjusted for FIGO stage, patient age, and tumor cell type. It also had more prognostic power than other grading systems tested when it was applied to all tumors, regardless of their cell type; however, the FIGO grading system was superior for prognostication when only carcinomas of endometrioid type were considered. With the FIGO grading system, no significant difference in survival was observed between patients with grade 1 and grade 2 tumors. Combining FIGO grades 1 and 2 results in a binary system (grades 1 and 2 vs. grade 3) that was the most prognostically significant grading system tested, with the additional advantages of being highly reproducible and familiar to practicing pathologists.

Nuclear size, shape, and density in endometrial carcinoma: relationship to survival at over 5 years of follow-up. Does analyzing only cells occupying the G0-G1 peak add useful information?

OBJECTIVE

The authors, using image analysis, previously demonstrated nuclear size and summed optical density to be independent prognostic indicators of recurrence in patients with endometrial carcinoma. The same tumors were analyzed by studying the optical features in the G0-G1 peak to see if this changed the values found as well as their importance as prognostic features at greater than 5 years of follow-up.

METHODS

Tumors from 74 consecutive patients, surgically treated, with endometrial cancer, were evaluated. Survival, depth of invasion, lymphvascular space invasion, FIGO stage, grade, histology were analyzed. DNA index, progesterone receptor status, as well as nuclear size (NUSZ), shape (NUSH), and summed optical density (NUSD) were evaluated. NUSZ, NUSH, and NUSD were quantified using image analysis.

RESULTS

Fifteen patients died from disease during the observation period of the study. Mean follow-up was 82 months with a median of 84 months. Forty-nine patients had stage I cancers, five stage II, 17 stage III, and three stage IV. NUSZ and NUSD were all significantly different between the original (entire cell cycle) and the re-measured (G0G1 only) values (both P < 0.001). Multivariate analysis showed both the original (P = 0.0001) and G0G1-only (P = 0.046) NUSZ and the original (P = 0.0002) and G0G1-only (P = 0.018) NUSD to be independent prognosticators of survival.

CONCLUSION

Image analysis is able to quantify cellular and nuclear parameters not otherwise quantifiable. NUSD and NUSZ correlated with traditional prognostic indicators, were demonstrated independent predictors of survival at over 5 years of follow-up. Although the re-measured NUSZ and NUSD from only the G0-G1 peak were significantly different from the original NUSZ and NUSD, they were not as valuable as prognostic factors. Nuclear size and summed optical density measured from the entire cell cycle are independent prognostic indicators of survival at greater than 5 years of follow-up. Measuring nuclear morphometric features in the G0-G1 peak only does not add any new prognostic information.

DNA ploidy and S-phase fraction analyses of hyperplastic, atypical and cancerous endometrium using flow cytometry from paraffin-embedded tissues

Atypical hyperplasia of the endometrium is an entity that needs to be clearly distinguished from other florid hyperplastic states on the one hand and from well-differentiated adenocarcinomas on the other. This may at times be difficult on pure morphological grounds. In this study, DNA ploidy and S-phase fraction of hyperplastic, atypical and cancerous endometrium were evaluated using flow cytometry from paraffin-embedded tissues. In total, 72 cases (24 hyperplastic, 24 atypical and 24 adenocarcinomas) were selected. All hyperplastic endometria showed a diploid stemline, while 2/24 atypical hyperplasias showed aneuploid (Near-diploid) peaks. Both of these cases were severely atypical and one of these, on hysterectomy, showed early invasive carcinoma. There was no significant difference in mean S-phase fractions of hyperplastic vs. atypical endometria. DNA aneuploidy was significantly more common with much higher S-phase fractions in poorly and moderately differentiated carcinomas than in well-differentiated ones. It was concluded that aneuploid (near-diploid) peaks, if ever present in atypical hyperplasias, may indicate an aggressive disease/neoplastic transformation.

Correlation of survivin mRNA detection with histologic diagnosis in normal endometrium and endometrial carcinoma

BACKGROUND

The objective of this study was to determine if survivin mRNA expression is a marker of endometrioid adenocarcinoma and if survivin mRNA levels correlate with tumor grade and stage.

METHODS

Twenty-six samples of endometrioid adenocarcinoma and 18 cases of benign endometrium were obtained at surgery. RNA was extracted from tissues and was used for quantitative real-time RT-PCR, targeting a 91-bp sequence of survivin mRNA and the levels were standardized to the levels of ribosomal RNA. Statistical analysis of the correlation between histologic diagnosis and the corrected survivin mRNA levels was performed by the Fisher Exact test and the Kruskal-Wallis test.

RESULTS

Survivin mRNA was detected in all specimens. Survivin mRNA levels were increased in proliferative endometrium (P = 0.0509) and was increased in correlation with ascending grade in endometrioid adenocarcinoma (P = 0.01).

CONCLUSION

Survivin mRNA is not a specific marker of endometrial cancer, but may reflect an important mechanism in tumor progression of the endometrial mucosa.

Computerised morphometrical analysis in endometrial hyperplasia for the prediction of cancer development. A long-term retrospective study from northern Norway

AIMS

To evaluate and compare the long term prognostic value of the WHO classification and the computerised multivariate morphometrical D score in endometrial hyperplasia. To test the reproducibility of the D score in two different centres.

METHODS

Histopathological WHO classification and computerised morphometrical analysis using the D score (< 0, high risk; > 1, low risk; 0-1, uncertain) in a population based study from northern Norway of archival dilatation and curettage material from 68 women with 10-20 years of follow up.

RESULTS

Of the 68 patients included in the study, 18 developed cancer. The sensitivity and specificity of the D score (< 0 v > 1) were 100% and 78%, respectively, which was better than the WHO classification (89% and 60%, respectively). The negative and positive predictive values for the D score were 100% and 58% and of the WHO classification 94% and 44%, respectively. This study found a slightly higher specificity for the D score than former retrospective studies, but otherwise the results were comparable. The D score results were reproducible between the two centres (R = 0.91; slope = 0.98; intercept = 0.3).

CONCLUSIONS

D score assessment is a reproducible and more accurate predictor of outcome of endometrial hyperplasia than the WHO classification assessed by an experienced gynaecological pathologist. Routine application of the D score might reduce over and undertreatment of endometrial hyperplasia.

The prognostic value of semiquantitative nuclear grading in endometrial carcinomas

The objective of this study was to evaluate the prognostic value of a convenient nuclear grading system based on only the proportion of nuclei measuring more than 10 microm in length at the shortest axis in endometrial carcinomas. Of the 235 cases reviewed, 9 serous and 5 clear cell adenocarcinomas and 2 small cell carcinomas were eliminated, resulting in a study population of 219 cases of endometrial adenocarcinoma. The architectural grade was determined by the FIGO system. The criteria for nuclear grade were as follows: grade 1, no nucleus measuring more than 10 microm in length at the shortest axis; grade 2, nuclei measuring more than 10 microm in length at the shortest axis seen in percentages ranging between 0 and 10% of tumor cells in active areas; and grade 3, more than 10% of tumor cells in active areas with nuclei measuring more than 10 microm in length at the shortest axis. The criteria for combined grades were as follows: the tumors were graded according to the architectural grade, but high-grade nuclear abnormalities increased the grade by one for architectural grade 1 and 2 tumors. The cumulative 10-year survival rates for architectural grades 1, 2, and 3 were 92.4, 82.6, and 65.2%, respectively (chi2 = 17.9, P = 0.001). The survival rates for nuclear grades 1, 2, and 3 were 96.2, 76.1, and 70.1%, respectively (chi2 = 21.6, P < 0.001), while for combined grades 1, 2, and 3 the survival rates were 98.3, 83.2, and 65.2%, respectively (chi2 = 26.9, P < 0.001). Even when the cases were limited to the 147 stage I endometrial carcinomas examined, the prognostic value of the combined grading system was still found to be superior to that of the architectural grading system. Our observations therefore supported the FIGO recommendation for nuclear grade not only in stage I endometrial carcinomas, but also in all stages of endometrial carcinomas.

Potential criteria for cohort selection in chemoprevention trials of uterine adenocarcinoma

Women at risk of uterine cancer include those with one or more of the following characteristics: obesity, nulliparity, late menopause, diabetes mellitus, prolonged unopposed estrogen use, and tamoxifen therapy. Risk is additionally increased by the presence of endometrial hyperplasia. The incorporation of biomarkers into the selection criteria of cohort groups at risk for developing endometrial cancer offers an innovative approach to the clinical design of chemoprevention trials of endometrial adenocarcinoma. Biomarkers that may be useful in cohort selection include nuclear morphometry, specific genetic abnormalities, and markers of proliferation and differentiation.

Application of the image analysis system and Imago program to biological samples

Recent developments in computer systems and new stereological tools have made estimation of quantitative data in biological material simple, easy and fast. The purpose of the present study is to show the advantages and disadvantages of a new program developed by the University of Córdoba for the determination of stereological measurements. As an example, endometrial material was used for demonstrating the technical characteristics. Thus, the nuclear area and volume-weighted mean nuclear volume-Vv(nucl)-of 30 routinely processed, paraffin-embedded endometrial specimens from 12 atypical hyperplasias (AH) and 18 well-differentiated adenocarcinomas (WDA) were investigated. We show the excellent visualization of the material and rapid measurement of these quantitative parameters using the Imago program. In addition, the mean values of Vv(nucl) are more efficient and useful in separating AH and WDA than the mean nuclear area. It is concluded that Vv(nucl) is a superior and sensitive estimator for discrimination.

Image cytometry: current applications and future trends

Image cytometry has numerous clinical and research applications and is particularly useful in anatomic pathology for the study of malignant lesions. Modern image systems encompass morphometry, densitometry, neural networks, and expert systems. Rapid advances in technology and the development of user-friendly systems have provided pathologists with an alternative to flow cytometry, particularly useful in the evaluation of small or hypocellular specimens. The most common current application of image cytometry is for DNA analysis, followed by quantitation of immunohistochemical staining. Newer uses under active investigation include development of expert systems that may act as diagnostic consultants in the future. Beyond DNA analysis, image cytometry holds great promise for improved tumor classification, for screening and surveillance in high-risk populations, and as a tool to improve diagnostic ability. This article discusses types of image analysis systems, specimen preparation, data acquisition, current applications in specific organ sites, and possible future applications.

Stereological estimate of nuclear volume in endometrial adenocarcinoma of endometrioid type: reproducibility and intra-tumour variation

Stereological volume weighted mean nuclear volume estimate (Vv) is reported to be highly reproducible and to provide excellent prognostic information for some tumours. The aim of the present study was to investigate the reproducibility and the intra-tumour variation of nuclear Vv and compare it with a morphometric nuclear estimate, i.e. the mean shortest nuclear axis, and with conventional histopathological parameters used in the grading of endometrial adenocarcinomas. Sixty-three endometrioid adenocarcinomas were included in the study. Both Vv and mean shortest nuclear axis showed an acceptable reproducibility and the correlation between them was moderate (Spearman test; rs = 0.8). One-third of the tumours showed a marked intra-tumour variation. A considerable discrepancy between Vv and/or mean shortest nuclear axis and nuclear and architectural grade was found.

Immunodetection of sialyl-Tn antigen in normal, hyperplastic and cancerous tissues of the uterine endometrium

The expression of sialyl-Tn antigen (STn) in normal, hyperplastic and neoplastic tissues of the uterine endometrium was examined by immunoperoxidase staining of formalin-fixed, paraffin-embedded samples using the monoclonal antibody TKH-2, directed toward the STn structure (NeuAc 2-6GalNac 1-O-serine or threonine). STn was expressed in 13 of 18 normal postovulatory endometria with an increasing staining intensity and incidence in the late secretory phase. It was consistently absent in 10 proliferative endometria. None of 5 cystic, 4 adenomatous or 12 atypical hyperplasias expressed STn, but areas of severe cytological atypia in 3 atypical hyperplasias showed faint expression. STn expression was detected in 36 of 43 adenocarcinomas. Although the extent of staining varied from a few to most of the cancer cells, general staining was observed throughout the cytoplasm of cancer cells with increased staining of the luminal surface and frequent positive staining of intraluminal mucin. Thus, it is clear that STn is selectively expressed in cancer cells and shows restricted expression in normal and hyperplastic endometrial tissues. STn may be an early marker of malignant transformation and has potential for use as a diagnostic aid in the surgical pathology of the uterine endometrium.