Characterization of human colorectal mucosa, polyps, and cancers by means of computerized morphonuclear image analyses

CLASSIFICATION OF TUMOR HISTOPATHOLOGY VIA SPARSE FEATURE LEARNING

Our goal is to decompose whole slide images (WSI) of histology sections into distinct patches (e.g., viable tumor, necrosis) so that statistics of distinct histopathology can be linked with the outcome. Such an analysis requires a large cohort of histology sections that may originate from different laboratories, which may not use the same protocol in sample preparation. We have evaluated a method based on a variation of the restricted Boltzmann machine (RBM) that learns intrinsic features of the image signature in an unsupervised fashion. Computed code, from the learned representation, is then utilized to classify patches from a curated library of images. The system has been evaluated against a dataset of small image blocks of 1k-by-1k that have been extracted from glioblastoma multiforme (GBM) and clear cell kidney carcinoma (KIRC) from the cancer genome atlas (TCGA) archive. The learned model is then projected on each whole slide image (e.g., of size 20k-by-20k pixels or larger) for characterizing and visualizing tumor architecture. In the case of GBM, each WSI is decomposed into necrotic, transition into necrosis, and viable. In the case of the KIRC, each WSI is decomposed into tumor types, stroma, normal, and others. Evaluation of 1400 and 2500 samples of GBM and KIRC indicates a performance of 84% and 81%, respectively.

Advances in biophotonics detection of field carcinogenesis for colon cancer risk stratification

The process of neoplastic transformation of the colon involves a progression through hyperproliferative epithelium through the aberrant crypt foci→small adenoma→large adenoma→invasive cancer→metastatic disease. These are orchestrated by sequential genetic and epigenetic events which provide the underpinnings of cellular alterations such as early induction in proliferation/suppression of apoptosis, along with the late stage increase in invasiveness. Colorectal cancer (CRC) averages 49-111 mutations per tumor encompassing 10-15 critical signaling pathways[1]. Accumulating such a high number of mutations requires a fertile mutational field, which is the hallmark of colon carcinogenesis.

While genetic susceptibility to colorectal cancer is well-known, at least half of the risk is believed to be due to exogeneous factors (e.g., obesity, diet, exercise). Understanding these risk factors represents a promising mode of tailoring screening modality and intensity. However, previous attempts using these factors (i.e., NCI risk calculator) have only been modestly successful with an area under receiver operating characteristics (ROC) curve (AUC) of just 0.61. One of the most important concepts is that risk is the interaction between these genetic and environmental components and is driven by the variety of polymorphisms. Thus, predicting risk is difficult given the complexity. On the other hand, the colonic mucosa represents the end product of the complex interplay between these multiple factors. The power of field carcinogenesis is that it reflects this interplay between genetics and environment.

Nanocytology of rectal colonocytes to assess risk of colon cancer based on field cancerization

Developing a minimally invasive and cost-effective prescreening strategy for colon cancer is critical because of the impossibility of conducting colonoscopy on the entire at-risk population. The concept of field carcinogenesis, in which normal-appearing tissue away from a tumor has molecular and, consequently, nano-architectural abnormalities, offers one attractive approach to identify high-risk patients. In this study, we investigated whether the novel imaging technique partial wave spectroscopic (PWS) microscopy could risk-stratify patients harboring precancerous lesions of the colon, using an optically measured biomarker (L(d)) obtained from microscopically normal but nanoscopically altered cells. Rectal epithelial cells were examined from 146 patients, including 72 control patients, 14 patients with diminutive adenomas, 20 patients with nondiminutive/nonadvanced adenomas, 15 patients with advanced adenomas/high-grade dysplasia, 12 patients with genetic mutation leading to Lynch syndrome, and 13 patients with cancer. We found that the L(d) obtained from rectal colonocytes was well correlated with colon tumorigenicity in our patient cohort and in an independent validation set of 39 additional patients. Therefore, our findings suggest that PWS-measured L(d) is an accurate marker of field carcinogenesis. This approach provides a potential prescreening strategy for risk stratification before colonoscopy.

Morphometic analysis of TCGA glioblastoma multiforme

Background

Our goals are to develop a computational histopathology pipeline for characterizing tumor types that are being generated by The Cancer Genome Atlas (TCGA) for genomic association. TCGA is a national collaborative program where different tumor types are being collected, and each tumor is being characterized using a variety of genome-wide platforms. Here, we have developed a tumor-centric analytical pipeline to process tissue sections stained with hematoxylin and eosin (H&E) for visualization and cell-by-cell quantitative analysis. Thus far, analysis is limited to Glioblastoma Multiforme (GBM) and kidney renal clear cell carcinoma tissue sections. The final results are being distributed for subtyping and linking the histology sections to the genomic data.

Results

A computational pipeline has been designed to continuously update a local image database, with limited clinical information, from an NIH repository. Each image is partitioned into blocks, where each cell in the block is characterized through a multidimensional representation (e.g., nuclear size, cellularity). A subset of morphometric indices, representing potential underlying biological processes, can then be selected for subtyping and genomic association. Simultaneously, these subtypes can also be predictive of the outcome as a result of clinical treatments. Using the cellularity index and nuclear size, the computational pipeline has revealed five subtypes, and one subtype, corresponding to the extreme high cellularity, has shown to be a predictor of survival as a result of a more aggressive therapeutic regime. Further association of this subtype with the corresponding gene expression data has identified enrichment of (i) the immune response and AP-1 signaling pathways, and (ii) IFNG, TGFB1, PKC, Cytokine, and MAPK14 hubs.

Conclusion

While subtyping is often performed with genome-wide molecular data, we have shown that it can also be applied to categorizing histology sections. Accordingly, we have identified a subtype that is a predictor of the outcome as a result of a therapeutic regime. Computed representation has become publicly available through our Web site.

Absence of prognostic value of nuclear shape factor analysis in colorectal carcinoma: relevance of interobserver and intraobserver variability

PURPOSE

Several retrospective studies, including our previous investigation, have shown a prognostic value of nuclear shape factor in colorectal carcinomas. This prospective study was designed to assess the reliability of nuclear shape factor determined by nuclear morphometry and to confirm its prognostic value.

METHODS

Ninety-eight patients who underwent colorectal carcinoma resection were prospectively enrolled. Measurement of nuclear shape factor was performed by using a computer-based image analysis system. Nuclear shape factor was defined as the degree of circularity of the nucleus (1.0 for a perfect circle and <1.0 for any other elliptical shape). The prognostic impact of nuclear shape factor on ten-year survival and the intraobserver and interobserver agreement were assessed.

RESULTS

The nuclear shape factor mean values by American Joint Committee on Cancer stage were: 0.73 (0.07) in Stage I, 0.74 (0.06) in Stage II, and 0.75 (0.05) in Stage III carcinomas (P = 0.78, ANOVA). The intraobserver agreement was poor for observer A (r = 0.28) and practically nonexistent for observer B (r = -0.004, Pearson correlation). The intraclass coefficient for interobserver agreement was practically nonexistent. No significant association between nuclear shape factor and ten-year survival was found.

CONCLUSIONS

Our prospective results, as opposed to our previous retrospective results, suggest that the reliability for nuclear shape factor morphometric analysis is very poor. We failed to confirm a prognostic value for nuclear shape factor in colorectal carcinoma.

Ductal carcinoma in situ of the breast (DCIS) with heterogeneity of nuclear grade: prognostic effects of quantitative nuclear assessment

BACKGROUND

Previously, 50% of patients with breast ductal carcinoma in situ (DCIS) had more than one nuclear grade, and neither worst nor predominant nuclear grade was significantly associated with development of invasive carcinoma. Here, we used image analysis in addition to histologic evaluation to determine if quantification of nuclear features could provide additional prognostic information and hence impact prognostic assessments.

METHODS

Nuclear image features were extracted from about 200 nuclei of each of 80 patients with DCIS who underwent lumpectomy alone, and received no adjuvant systemic therapy. Nuclear images were obtained from 20 representative nuclei per duct, from each of a group of 5 ducts, in two separate fields, for 10 ducts. Reproducibility of image analysis features was determined, as was the ability of features to discriminate between nuclear grades. Patient information was available about clinical factors (age and method of DCIS detection), pathologic factors (DCIS size, nuclear grade, margin size, and amount of parenchymal involvement), and 39 image features (morphology, densitometry, and texture). The prognostic effects of these factors and features on the development of invasive breast cancer were examined with Cox step-wise multivariate regression.

RESULTS

Duplicate measurements were similar for 89.7% to 97.4% of assessed image features. For the pooled assessment with approximately 200 nuclei per patient, a discriminant function with one densitometric and two texture features was significantly (p < 0.001) associated with nuclear grading, and provided 78.8% correct jackknifed classification of a patient's nuclear grade. In multivariate assessments, image analysis nuclear features had significant prognostic associations (p <or= 0.05) with the development of invasive breast cancer. Texture (difference entropy, p < 0.001; contrast, p < 0.001; peak transition probability, p = 0.01), densitometry (range density, p = 0.004), and measured margin (p = 0.05) were associated with development of invasive disease for the pooled data across all ducts.

CONCLUSION

Image analysis provided reproducible assessments of nuclear features which quantitated differences in nuclear grading for patients. Quantitative nuclear image features indicated prognostically significant differences in DCIS, and may contribute additional information to prognostic assessments of which patients are likely to develop invasive disease.

Relationship between genetic alterations and prognosis in sporadic colorectal cancer

Because chromosomal chromosomal instability (CIN) and microsatellite instability (MSI) are important genetic alterations in colorectal cancers, we classified the sporadic colorectal cancers (CRC) on the status of the CIN and MSI and explored their molecular profiles. A total of 213 colorectal tumors were collected for analysis of DNA ploidy, MSI, loss of heterozygosity (LOH), mutation of p53 (exons 5 to 9), Ki-ras (exons 1 and 2) and BRAF (V599E). Relationships between clinicopathological variables and molecular analyses were analyzed with the chi(2) test (Yates' correction). Kaplan-Meier survival curves were compared using log-rank test. Variables with p < 0.1 were entered into the Cox regression hazard model for multivariate analysis. High microsatellite instability (MSI-H) existed in 19 tumors (8.9%), which were more likely to be right-sided (31.6%) with poor differentiation (26.3%). Seventy-one (33.3%) tumors were diploid and 142 (66.7%) were aneuploid. Mutations in p53, Ki-ras and BRAF were found in 45.1%, 41.8% and 4.2% of tumors, respectively. Based on MSI, and CIN, 3 classes were defined: (i) High microsatellite instability MSI-H tumors: young age, high carcinoembryonic antigen (CEA) level, right colon, poorly differentiated, mucin production, high BRAF mutation, lower allelic loss and relatively good prognosis; (ii) Microsatellite stability (MSS) diploid tumors: right colon, poorly differentiated, less infiltrative tumor, mucin production, lower allelic loss and low p53, BRAF mutation; (iii) MSS aneuploid tumors: more infiltrative invasion, greater allelic loss and high p53 mutation. According to multivariate analysis, tumor stage and p53 mutation were significantly associated with disease progression. The MSS diploid and MSS aneuploid CRCs could be subtyped with p53 mutation and had different prognostic outcome and molecular profiles. The 4-year disease-free survival (DFS) of patients with MSS-diploid, wild-type p53 tumors was 67% and significantly higher than those of patients with MSS-diploid, mutant p53 CRC (30%, p = 0.003). The same trend was found in patients with MSS-aneuploid CRC(wild p53 vs. mutant p53, 64% vs. 41%, p = 0.009). We concluded that CIN, MSI and p53 mutation status might be used as a multiple parameter profile for the prognosis of sporadic colorectal cancer.

Loss of heterozygosity and DNA aneuploidy in colorectal adenocarcinoma

BACKGROUND

This study evaluated the relationship between DNA aneuploidy and loss of heterozygosity (LOH) at different genetic loci in colorectal adenocarcinoma.

METHODS

A total of 112 patients with surgically removed colorectal adenocarcinoma in Taipei Veterans General Hospital from January 1999 to July 2001 were included in this study. The pattern of DNA ploidy was determined with DNA flow cytometry, and the LOH of various genetic loci was determined with fluorescence polymerase chain reaction and denaturing gradient gel electrophoresis. The relationship between DNA ploidy, LOH of various genetic loci, and clinicopathologic variables was analyzed with the chi(2) test with Yates' correction as well as by multivariate binary logistic regression analysis.

RESULTS

Seventy-one (63.4%) of the 112 carcinomas had DNA aneuploidy. The DNA aneuploidy was not associated with any clinicopathologic variable. Ninety-one tumors (81.3%) exhibited LOH in at least one genetic locus. In the univariate analysis, the DNA aneuploidy was associated with LOH of Tp53-penta, D8S254, D5S346, and high-frequency LOH (P =.001, P =.016, P =.041, and P <.001, respectively). In the multivariate analysis, the most significant factor influencing DNA aneuploidy was D8S254, followed by Tp53-penta, high-frequency LOH, and D5S346.

CONCLUSIONS

DNA aneuploidy is strongly associated with LOH at specific genetic loci.

Image cytometry as a discriminatory tool for cytologic specimens obtained by endoscopic retrograde cholangiopancreatography

BACKGROUND

Routine brush cytology is relatively insensitive for the diagnosis of biliary and pancreatic malignancy. Sensitivity can be improved by measuring DNA and proliferation. The goal of this study was to assess the discriminatory capacity of image cytometry using pancreaticobiliary brush cytology specimens obtained during endoscopic retrograde cholangiopancreatography (ERCP). Analysis included morphometry, DNA quantification, and characterization of nuclear chromatin distribution and condensation.

METHODS

Brush cytology specimens were obtained during ERCP from 22 chronic pancreatitis specimens, 11 pancreatic adenocarcinoma specimens, 13 primary sclerosing cholangitis specimens, and 11 cholangiocarcinoma specimens and contrasted with 25 normal epithelia specimens. A SAMBA 2005 image processor was used to analyze Feulgen stained chromatin density and distribution. Discriminant analysis of 37 morphonuclear variables was performed to characterize differences between: 1) chronic pancreatitis and pancreatic adenocarcinoma, and 2) primary sclerosing cholangitis and cholangiocarcinoma.

RESULTS

Chronic pancreatitis was distinguished from pancreatic adenocarcinoma (P < or = 0.001); sensitivity and specificity were both estimated to be 82%. Primary sclerosing cholangitis was distinguished from cholangiocarcinoma (P < or = 0.01); sensitivity and specificity were estimated to be 82% and 85%, respectively.

CONCLUSIONS

Multiparameter image cytometry has potential as an adjuvant diagnostic technique in patients with pancreaticobiliary malignancy.

Strong decrease in biotin content may correlate with metabolic alterations in colorectal adenocarcinoma

Short-chain fatty acids are an important source of energy for colonocytes. One of these is propionate, which is metabolized through carboxylation by propionyl-CoA carboxylase (PCC), an enzyme encoded by 2 genes, PCCA and PCCB. The co-factor of this reaction is biotin, a product of intestinal bacterial metabolism, as is propionate. Despite detailed knowledge about the metabolic effects and physiology of biotin, the relative amounts of this vitamin in normal colonic mucosae and in tumour tissue remains quite unknown. The biotin content in normal and cancerous cells from the distal digestive tract was examined on 10 pairs of tissue specimens of colorectal cancer and adjacent normal mucosae using reflectance in situ hybridization (RISH). Having observed a high biotin content in colon mucosae and a low content in colorectal-cancer cells, we then studied the transcription levels of PCCA and PCCB genes in 9 colorectal cancers and the corresponding mucosae. In all cases, the levels of mRNA were lower in colorectal cancers than in normal mucosae, the decrease being always more marked for PCCB than for PCCA. In normal mucosae and in adenocarcinoma cancer cells, PCCA and PCCB transcription levels were strongly related to the amount of biotin detected, but not to the number of chromosomes 13 (which carries PCCA) or 3 (which carries PCCB).

Genetic heterogeneity in sporadic colorectal adenomas

The majority of colorectal cancers develop from adenomatous polyps under the influence of factors that are still poorly understood. Tumourigenesis is generally considered a multistep process in which multiple genetic alterations occur, eventually reflected in abnormalities of the cellular DNA content. Macroscopical features such as tumour size and tumour architecture (tubular, tubulovillous, or villous) are correlated wit the chance of malignancy in the lesion. Grade of dysplasia can be considered an indicator for the level of progression of the adenoma towards invasive carcinoma. These characteristics were correlated with the presence or absence of K-ras mutations and the DNA ploidy in a prospective study performed on 46 large sporadic colorectal adenomas resected by endoscopy. DNA ploidy and K-ras mutations were analysed in two samples taken at distant sites in the adenomas. Aneuploidy was present in 12 adenomas (26 per cent) and K-ras mutations occurred in 26 (57 per cent). A highly significant correlation was found between aneuploidy and adenoma size, architecture, and grade of dysplasia. The presence of K-ras mutations was significantly correlated only with the size of the adenomas. The proportion of adenomas with aneuploidy and/or a K-ras mutation increased when two samples were analysed instead of one. This observation suggests that the prevalence of genetic mutations and of aneuploidy is probably underestimated, as generally only one sample is investigated. No correlation was observed between K-ras mutations and ploidy. This study demonstrates the presence of genetic heterogeneity in colorectal adenomas and supports the notion that K-ras mutation is an early event, while aneuploidy is a late event in the adenoma-carcinoma sequence.

Multiparametric image cytometry in mycosis fungoides

Eighteen cases of early mycosis fungoides were compared with 18 cases of eczematous dermatitis by multiparametric image cytometry. A minimum of 100 lymphocytes was measured in each case. A large number of measurements was acquired for each lymphocyte, characterizing nuclear DNA content, area, shape, and texture. There were significant differences between the two groups, especially in nuclear DNA content and texture. These differences allowed the two groups of nuclei to be distinguished with 78% accuracy. The two groups of lesions were distinguished with 94% accuracy, using neural network analysis.

Image analysis in dermatopathology

BACKGROUND/AIMS

Image analysis in dermatopathology has been used for DNA ploidy analysis, morphometry, stereology, and quantitative immunohistochemistry. The object is to review image analysis in dermatopathology and evaluate these modalities and their application in pigmented lesion pathology, for elucidation of tumor behaviour and architecture and as an aid in tumor identification and prognostication.

CONCLUSION

Image analysis in dermapathology has a huge potential. The techniques are difficult and at present mainly used in specialized centres.

Chemotherapy-induced nuclear alterations of morphologic and genomic characteristics in a human colon cancer grafted onto nude mice

PURPOSE

A human Dukes B colonic adenocarcinoma was grafted onto 40 nude mice. The mice were divided into four groups, one control and three representing experimental conditions. Animals in the three experimental groups received either adriamycin (ADR), 5-fluorouracil (5-FU), or camptothecin (CPT) over a 25-day period beginning 34 days after grafting. Control animals received saline on an identical schedule. Animals were killed 105 days after grafting.

METHODS

The effect of therapy was assessed by three techniques: 1) tumor size was periodically measured during the life of the animals, 2) modifications of APC, Ki-ras, and p53 genes were studied by polymerase chain reaction, dot-blot analysis, restriction analysis, and DNA sequencing, and 3) image cytometry of Feulgen-stained material was used to characterize 15 parameters describing morphometric, densitometric, and textural features of tumor nuclei.

RESULTS

When compared with controls, tumor growth (size) was maximally suppressed by treatment with CPT (P < or = 0.001). Growth was inhibited significantly by treatment with 5-FU (P < or = 0.01); no statistical difference in tumor size was observed between controls and animals treated with ADR. Modifications of APC, Ki-ras, and p53 genes were not observed; however, treatment did inhibit amplification of APC and p53 genes.

CONCLUSIONS

The 15 morphonuclear parameters were assessed to define populations of cell nuclei altered by chemotherapy. Although CPT maximally suppressed growth, it did not alter nuclear morphology when compared with controls. Treatment with either 5-FU or ADR resulted in nuclear morphologic alterations defined as distinct populations using multivariate analysis. Nonsupervised linear discriminant analysis was used to quantify the relative proportions of these populations. Four morphonuclear parameters were identified, which discriminated nuclei exposed to either ADR or 5-FU from controls.

Variability of neoplastic parameters in colon tumours, and its significance in diagnostic practice

We have reviewed the value of individual variability in the reaction of tissues to treatment with carcinogens, and the manifestation of this variability in different morphological (histological, morphometric, and ultrastructural), histochemical and immunohistochemical parameters generated in tumorous tissues. Moreover, we also reviewed data in the literature on individual variability in the manifestation of some biochemical and immunochemical markers which are accumulated in the serum of tumour-bearing patients and which are characteristic for the different phases of tumourigenesis. The high variability of different tumorous parameters suggests that none can be utilized alone as a conclusive marker of neoplasia and that only their combined use can give objective information. We also reviewed the impact of this variability in the evaluation of various pathological methods to detect different stages of neoplastic transformation in the colon. It has been concluded that none of the present approaches can be absolutely conclusive and without false results, and that objective information regarding early cancerous changes may be obtained only by use of combined analyses utilizing several laboratory methods. The diagnostic procedures should be based on the complex utilization of all appropriate methods using the quantitative interpretation of the obtained data. Multivariate analysis of many parameters should be very effective for the prediction of therapeutic results.

A morphometrical analysis of minute depressed adenomas in familial polyposis coli

In a morphometrical study of minute adenomas in familial polyposis coli (FPC), ranging from 0.8 mm to 1.5 mm in diameter, five parameters of depressed adenomas were compared with those of ordinary polypoid adenomas. The density of glands (DG), an index of structural atypia, and the nuclear-cytoplasmic (N/C) ratio, an index of cellular atypia, were higher in depressed adenomas than in polypoid adenomas (74.8% vs 66.8% and 37.8% vs 32.7%, respectively). The cell area (C-area) of depressed adenomas was less than that of polypoid adenomas (107.3 microns 2 vs 121.1 microns 2), but there was no significant difference in values for the nuclear area (N-area) in the two types (40.0 microns 2 vs 39.2 microns 2). The higher N/C ratio of depressed adenomas was, therefore, not caused by enlargement of the nuclei but by reduction of the cell size. In addition, values of the sphericity of nuclei (SN) (3.14 vs 3.65) indicated rounder nuclei in depressed adenomas than in ordinary ones. The higher DG and N/C ratio of the depressed adenomas indicates their higher grade of atypia than that of ordinary polypoid adenomas.

DNA densitometry of colorectal cancer

DNA analysis was assessed by densitometry for 281 cases of colorectal adenocarcinoma. Detection of aneuploidy in a single case rose from 65% if one, to 92.5% when three or more sections, were analysed. Although aneuploid tumours had significantly larger nuclear areas than near diploid tumours (p = 0.009), densitometric measurements showed no association with clinicopathological variables. DNA content determined by densitometry was compared with that from flow cytometry on 465 tissue sections from 241 cases. Aneuploidy assessed by flow cytometry was significantly associated with that determined by densitometry (p < 0.01 for all comparisons), ploidy state being similar in 381 sections (82%, kappa = 0.63, p < 0.001), and 187 cases (77.6%, kappa = 0.57, p < 0.001). Univariate survival analysis showed that DNA densitometric variables had no significant association with survival in (a) all cases, (b) cases without lymph node metastases, or (c) cases without distant metastases. Multivariate regression analysis of densitometric and clinicopathological variables identified Dukes's stage, patient age, and tumour differentiation as the combination of variables most closely related to survival. Densitometric measurement of DNA content could not significantly improve on the prognostic model containing these three variables. It is concluded that, although the assessment of DNA content by densitometry is comparable with that of flow cytometry, conventional histological variables remain the best predictors of prognosis in colorectal cancer.

Computerized morphonuclear characteristics and DNA content of adenocarcinoma of the pancreas, chronic pancreatitis, and normal tissues: relationship with histopathologic grading

We report the morphonuclear characteristics of normal (13 cases), benign (ie, chronic) pancreatitis (six cases), and neoplastic (ie, ductal) adenocarcinoma (22 cases) tissues of the pancreas. This description is based on computerized cell image analysis, which permits the determination of parameters related to the morphometric (nuclear area), densitometric (nuclear DNA content), and chromatin texture features of Feulgen-stained nuclei from paraffin-embedded archival material. We observed that nuclear area discriminates between normal and benign (ie, chronic pancreatitis) as opposed to neoplastic cell nuclei. Morphonuclear parameters describing chromatin pattern characteristics made it possible to discriminate between grade I pancreatic carcinoma and normal and benign cell nuclei on the one hand, and grades I and III carcinoma on the other hand. The nuclear DNA content increased in a continuous manner from normal and benign through low-grade to high-grade neoplastic tissues of the pancreas. Combining the morphometric, densitometric, and textural parameters into one equation, we were able to calculate a score (ie, the malignancy level index) that showed a close relationship to conventional histopathologic grading. Thus, the computer-aided diagnosis of cytologic specimens from pancreatic lesions offers information of the same significance as that obtained by conventional histopathologic grading.

Histopathologic grading and DNA ploidy in relation to survival among 206 adult astrocytic tumor patients

BACKGROUND

The authors studied the benefit of performing histopathologic grading and DNA ploidy characterization with respect to patient survival in a series of 206 astrocytomas (AST) for which they obtained 134 complete clinical follow-ups.

METHODS

The material analyzed came from archival material, i.e., formalin-fixed paraffin-embedded tissues. DNA ploidy was assessed by means of a cell image processor computing the integrated optical density (IOD) on Feulgen-stained nuclei.

RESULTS

Results showed that histopathologic diagnosis in three grades, i.e., AST, anaplastic astrocytoma (ANA), and glioblastoma multiforme (GBM), had a significant prognostic value. Patients with AST showed a mean survival time (between histopathologic diagnosis and death) of more than 36 +/- 6 months (AST versus ANA or GBM) (P less than 0.001). Patients with ANA and GBM showed a mean survival time of 15 +/- 2 and 10 +/- 1 months, respectively, (ANA versus GBM) (P less than 0.05). Patient age strongly correlated with survival. Patients younger than 40 years of age had a mean survival time of 20 +/- 4 months. Patients between 41 and 60 years of age had a mean survival time of 12 +/- 2 months, and patients older than 60 years of age had a mean survival time of 11 +/- 1 months.

CONCLUSIONS

Considering DNA ploidy characterization, the authors noticed that aneuploid ANA (DNA index [DI] more than 1.30) were associated with a significantly higher mean patient survival time compared with that associated with euploid ANA. In contrast, the authors did not find this in either of the groups with AST and GBM. Recognizing six DNA histogram types (diploid, triploid, tetraploid, hyperdiploid, hypertriploid, and polymorphic), the authors observed that hypertriploid tumors were associated with greater patient survival compared with what happened in the cases of the five other DNA histogram types. This was true with respect to the three AST histopathologic types. Thus, DNA ploidy determination seemed helpful in characterizing aggressiveness in adult AST.

The relationship of DNA aneuploidy to molecular genetic alterations in colorectal carcinoma

Altered total nuclear DNA content is frequent in colorectal carcinomas, but the mechanisms producing aneuploidy are unknown. Therefore, DNA ploidy by flow cytometry was correlated with molecular genetic alterations and tumor characteristics in 50 colorectal carcinomas. The prognostic value of these alterations was also evaluated because aneuploidy has been associated with poor prognosis. Thirty-nine of the carcinomas (78%) were DNA aneuploid. When compared with diploid carcinomas, aneuploid tumors had greater mean fractional allelic loss, defined as the fraction of evaluable nonacrocentric autosomal arms with deletion (0.25 +/- 0.15, range 0-0.667, vs. 0.12 +/- 0.10, range 0-0.345; P = 0.006). DNA index by flow cytometry correlated with fractional allelic loss (r = 0.38, P = 0.006). Aneuploid tumors also had more frequent allelic loss on chromosome 17p (87% vs. 55%; P = 0.017), but less frequent ras gene mutation (44% vs. 82%; P = 0.025). Among the 25 right-sided and 25 left-sided tumors, DNA diploid tumors were more frequent on the right side (P = 0.002), whereas deletion of 17p was found predominantly on the left side. Aneuploidy was associated with moderate and poor differentiation of the carcinomas but not with distant metastasis. By contrast, high fractional allelic loss, deletion of 17p, and deletion of 18q were associated with distant metastasis. In survival analysis of patients with Dukes' B or C carcinoma, DNA aneuploidy was not a significant discriminator, but patients whose tumor had deletion of 17p or deletion of both 17p and 18q had poorer survival (P = 0.045 and 0.022, respectively). The results suggest that DNA aneuploidy is associated with some of the molecular genetic alterations and phenotypic characteristics of colorectal carcinomas but is not a reliable indicator of metastatic potential.

Monitoring of radiotherapy-induced morphonuclear modifications in the MXT mouse mammary carcinoma by means of digital cell image analysis

The present work deals with the characterization of the morphological changes induced by ionizing radiation on MXT mouse mammary cancer cell nuclei. The monitoring of the radiotherapy-induced morphonuclear modifications was carried out by means of digital cell image analysis, which made it possible to compute morphometric (nuclear area), densitometric (nuclear DNA content) and textural (chromatin pattern characteristics) parameters assessed on Feulgen-stained nuclei. The biological material was obtained from serial fine needle-aspirations performed on control tumors and 2Gy and 8Gy irradiated tumors. Digital cell image analyses were assessed at the 7th, 18th, and 28th days-post irradiation. We showed that radiotherapy induced specific morphonuclear modifications that appeared in a dose-dependent and time-dependent manner. The 2Gy and 8Gy radiotherapy doses led to an increase in the mean nuclear area value, this feature being observed at both the 1st and the 4th week post-irradiation. The resultant effect on the chromatin pattern characteristics corresponded to a decrease in overall condensation level as compared to the control cell nuclei. Finally, the mean nuclear DNA content increased at the 1st week post-irradiation, but decreased at the 4th week post-irradiation as compared to the MXT control tumor. Computerized cell image analysis, therefore, appears to be a useful tool in helping to monitor the radiotherapy-induced effects that occur in neoplastic cell nuclei and which can be observed by means of optical microscopy.

Quantitative nuclear cell image analyses of thyroid tumors from archival material

Sixty-three sections of Feulgen-stained thyroid cell nuclei from paraffin-embedded material, including five multinodular goiters, 10 adenomas, 36 papillary carcinomas, seven follicular carcinomas, and five medullary carcinomas were analyzed by means of the SAMBA 200 (TITN, Grenoble, France) cell image processor. This was done in order to obtain nuclear characteristics of papillary versus follicular carcinomas. The nuclear features were assessed by morphometric, densitometric, and textural parameters. Our preliminary results indicate that the cell nuclei from typical histopathologic specimens of follicular thyroid cancers belong to a larger thyroid cell nuclei population corresponding to the histopathologic family of papillary thyroid cancers. This follicular neoplastic cell nuclei population appears to be quite distinct from the typical medullary neoplastic cell nuclei population which also belongs to the papillary neoplastic cell nuclei population. It appears that there is a specific papillary cell nuclei subpopulation containing typical hypochromatic cell nuclei. We also observed a dramatic increase in nuclear size and hyperchromatism between normal (multinodular goiters) and neoplastic (carcinomas) thyroid tissues, with the benign tissues (adenomas) showing intermediate nuclear characteristics.

Digital morphonuclear analyses of sensitive versus resistant neoplastic cells to vinca-alkaloid, alkylating, and intercalating drugs

We tested 12 resistant cell lines in vitro in order to evaluate common morphonuclear characteristics induced by various cytotoxic drugs on cell lines of different origins. We used the MXT mouse mammary cancer and the neoplastic J82 and T24 human bladder cell lines, whose variants are either sensitive or resistant to a vinca alkaloid derivative (Navelbine, NVB), to an investigational alkylating agent (PE1001), and to Adriamycin (ADR). We tested cell population variants resistant to NVB + PE1001 + ADR. The level of chemoresistance was evaluated by a colorimetric assay assessing the 50% concentration-induced inhibition of cellular growth (IC50) brought about by each drug on the growth of each cell variant under study. We show that resistant neoplastic cell nuclei present common morphonuclear characteristics, independent of cell origin (neoplastic mouse mammary versus human bladder cells) and the drug used (vinca alkaloid, alkylating, and intercalating derivatives). Our results further indicate that the phenotype of resistant versus sensitive cells corresponds to cell nuclei populations with smaller nuclei and less nuclear DNA content and, as a consequence, a chromatin texture showing large pale areas with some hyperchromatic clumps.