Prognostic relevance of tumour-associated macrophages and von Willebrand factor-positive microvessels in colorectal cancer.

Tumour-associated macrophages (TAM) are involved in tumour angiogenesis and anti-tumour immune response. In colorectal cancer (CRC), an association of high microvascular density (MVD) and unfavourable prognosis has been reported by some investigators. However, heterogeneous patient groups were studied. We, therefore, analysed the correlation between TAM and MVD and the prognostic relevance of MVD, TAM and T lymphocyte infiltration for long-term survival in a homogeneous group of 70 patients with moderately differentiated cancers of the International Union Against Cancer (UICC) stages II and III, who did not receive chemotherapy. MVD was evaluated using immunohistochemistry with antibodies against CD34 and von Willebrand factor (vWF). TAM and T lymphocytes were visualised with antibodies against CD68 and CD3, respectively. Statistical analysis did not reveal a significant correlation between TAM and T lymphocyte numbers and MVD. Multivariate analysis of immunohistochemical data from all CRC patients and the subgroup of patients with UICC stage-II CRC identified TAM- and vWF-positive microvessel numbers as prognostically relevant markers. Low numbers of TAM- and high numbers of vWF-positive microvessels were associated with an unfavourable prognosis. In conclusion, TAM- and vWF-positive microvessel numbers may serve as independent prognostic markers for patients with UICC stage-II and -III CRC and may help to identify patients with an unfavourable prognosis.

Application of multilevel models to morphometric data. Part 1. Linear models and hypothesis testing.

Morphometric data usually have a hierarchical structure (i.e., cells are nested within patients), which should be taken into consideration in the analysis. In the recent years, special methods of handling hierarchical data, called multilevel models (MM), as well as corresponding software have received considerable development. However, there has been no application of these methods to morphometric data yet. In this paper we report our first experience of analyzing karyometric data by means of MLwiN - a dedicated program for multilevel modeling. Our data were obtained from 34 follicular adenomas and 44 follicular carcinomas of the thyroid. We show examples of fitting and interpreting MM of different complexity, and draw a number of interesting conclusions about the differences in nuclear morphology between follicular thyroid adenomas and carcinomas. We also demonstrate substantial advantages of multilevel models over conventional, single-level statistics, which have been adopted previously to analyze karyometric data. In addition, some theoretical issues related to MM as well as major statistical software for MM are briefly reviewed.

Application of multilevel models to morphometric data. Part 2. Correlations.

Multilevel organization of morphometric data (cells are "nested" within patients) requires special methods for studying correlations between karyometric features. The most distinct feature of these methods is that separate correlation (covariance) matrices are produced for every level in the hierarchy. In karyometric research, the cell-level (i.e., within-tumor) correlations seem to be of major interest. Beside their biological importance, these correlation coefficients (CC) are compulsory when dimensionality reduction is required. Using MLwiN, a dedicated program for multilevel modeling, we show how to use multivariate multilevel models (MMM) to obtain and interpret CC in each of the levels. A comparison with two usual, "single-level" statistics shows that MMM represent the only way to obtain correct cell-level correlation coefficients. The summary statistics method (take average values across each patient) produces patient-level CC only, and the "pooling" method (merge all cells together and ignore patients as units of analysis) yields incorrect CC at all. We conclude that multilevel modeling is an indispensable tool for studying correlations between morphometric variables.

How to improve microscopic images obtained with consumer-type digital cameras.

AIM: Digital imaging is useful in conventional photography because it immediately provides images, and the image quality can be improved afterwards by the use of computer programs. The major disadvantages of consumer-type digital cameras mounted on microscopes are (i) unequal illumination through the image, and (ii) a coloured background. A computer program was specifically adapted and refined to improve images obtained with consumer-type digital cameras mounted on microscopes. METHODS AND RESULTS: An approach using a division operation between the specimen image and a background image leads to homogeneous illumination throughout the image, with automatically corrected brightness and white background. The correct colour spectrum is preserved by correction of the histogram. This approach was obtained from the freeware computer program 'Image Arithmetic'. In a test, three different consumer-type digital cameras (Sony, Nikon, Olympus) on different microscopes were used to obtain images of different types of histological specimens (cervical smear, bone marrow biopsy, and colonic biopsy). The computer program dramatically improved the quality of images obtained with all tested cameras. CONCLUSION: Using this approach, even low-cost digital cameras mounted on microscopes produce brilliant images with homogeneous illumination and a white background, the image quality being comparable with expensive cameras especially designed for microscopes.

Primary unit for statistical analysis in morphometry: patient or cell?

In a series of 16 oxyphilic follicular neoplasms of the thyroid (8 adenomas and 8 carcinomas), three different approaches for the analysis of morphometric data were evaluated. It was shown that the statistical design of morphometric studies is by nature nested due to subsampling of cells within each patient. Therefore, the most appropriate analysis would be to account for this hierarchical structure. However, related statistical methods are not at present well established, especially as far as classification rules are concerned. Therefore, the nested design is converted into the simple factorial one by considering only one kind of statistical unit - either patients or cells. The results of the study presented indicate that ignoring the patient as unit of analysis leads to a substantial error in statistical output, regardless of the particular procedure applied. Moreover, the size of the error can be neither diminished nor controlled. Choosing patients as primary units assures accurate results and also has an advantage of gaining some additional information by calculating several distributional estimates in each patient. However, this approach often requires a reduction of dimensions and, furthermore, is not encouraged in certain fields of quantitative cytology. Advantages and disadvantages of all approaches have been summarized and practical recommendations for their use have been worked out.

Multivariate karyometric approach in differential diagnosis of follicular thyroid neoplasms. A study of 31 cases.

A retrospective analysis of 19 follicular adenomas, 12 minimally invasive follicular carcinomas and 3 widely invasive follicular carcinomas of the thyroid was performed on 5-microm-thick Feulgen-stained paraffin sections by means of a semiautomatic system for picture analysis. The major aim was to assess the potential of multiparameter karyometry for separation of the first two tumour types. Sixteen planimetric and densitometric features were defined in each case on 200-300 randomly selected nuclei and processed by a number of uni- and multivariate statistical methods. Despite predominantly significant ANOVA results a substantial overlap between tumour groups limited the practical usefulness of any karyometric feature alone. Factor and cluster analyses indicated independence of planimetric and densitometric parameters from each other, which was of crucial importance in finding an optimal subset of variables for discriminant analysis. The classification rule derived from the latter procedure was checked by the "jack-knife" method, by classification of 3 widely invasive cancers and by hierarchical tumour clustering. Sensitivity and specificity of the model for detection of malignancy were 100% and 94.7%, respectively. A multivariate karyometric approach, when applied correctly, can be a useful tool for differentiation between follicular adenomas and minimally invasive follicular carcinomas of the thyroid.