gms | German Medical Science

MAINZ//2011: 56. GMDS-Jahrestagung und 6. DGEpi-Jahrestagung

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V.
Deutsche Gesellschaft für Epidemiologie e. V.

26. - 29.09.2011 in Mainz

Automated microscopic cell co-localization analyses in physiological germinal centers and follicular lymphoma

Meeting Abstract

  • Ulf-Dietrich Braumann - Institut für Medizinische Informatik, Statistik und Epidemiologie, Universität Leipzig, Leipzig
  • Karoline Koch - Institut für Pathologie, Sektion Hämatopathologie und Lymphknotenregister, Christian-Albrechts-Universität Kiel, Kiel
  • Markus Löffler - Institut für Medizinische Informatik, Statistik und Epidemiologie, Universität Leipzig, Leipzig
  • Wolfram Klapper - Institut für Pathologie, Sektion Hämatopathologie und Lymphknotenregister, Christian-Albrechts-Universität Kiel, Kiel

Mainz//2011. 56. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds), 6. Jahrestagung der Deutschen Gesellschaft für Epidemiologie (DGEpi). Mainz, 26.-29.09.2011. Düsseldorf: German Medical Science GMS Publishing House; 2011. Doc11gmds459

DOI: 10.3205/11gmds459, URN: urn:nbn:de:0183-11gmds4595

Published: September 20, 2011

© 2011 Braumann et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en). You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.


Outline

Text

Introduction/background: Contrasting to aggressive B-cell lymphoma (BCL), low-grade, indolent BCL display a complex tissue organization w.r.t. lymphoma cells and various reactive bystander cells, like T-cells and dendritic cells. This lymphoma tissue organization is most evident in follicular lymphoma (FL). The presumed physiological counterpart and origin of FL is the lymph follicle with the germinal center (GC), exhibiting a similar cellular composition but differing in tissue organization. The presence of bystander cells in FL has been shown to be associated with the clinical course [1]. However, most analyses mainly have addressed the content of bystander cells. A comprehensive analysis of local spatial adjacencies of B-cells vs. bystander cells could provide new insight into tissue organization and the role of bystander cells in lymphoma pathogenesis, maintenance and evolution. Here we report on our development of a tailored fluorescence microscopy-related image processing chain for precise and reproducible analyses of spatial cell distributions in lymph follicles.

Material and methods: Fluorescence multi-staining was applied on FL specimen and reactive tonsils as physiological counterpart, combining four markers of bystander cells with B-cell transcription factor Pax5, and Ki67 to evaluate proliferation. Three respective four-channel digital images were obtained from each staining using conventional fluorescence microscopy. As initial image processing step automatic follicle and follicle zone segmentation using a kernel-density estimation approach was applied. Cell nuclei and cell segmentation was done using a sequence of high-pass filtering, constant thresholding and morphological opening. Co-localization analyses referring to B-cell vs. bystander cell distances were accomplished utilizing bystander cell-related distance maps obtained by Euclidean distance transform [2].

Results: A systematic comparative cell co-localization study was performed on physiological GC and FL follicles. Comparison of dark vs. light zone measurements in the physiological GC indicated clear differences in distance histograms for all analyzed bystander cells except macrophages. Within dark zone obtained distances are decisively enlarged, while in light zone short distances appear dominant. For the FL exhibiting no functional zones, measurements appear as average between the light and dark zones of the physiological GC. Cross-comparisons using histometric numbers like B-cell portion with immediate adjacency, spatial cell densities, and cell area coverage provide further insights into the follicular architectural details.

Discussion/conclusions: The present work exemplifies the potential of advanced histometric methods. Our approach might provide insights into the complex tumor vs. bystander cell interactions in FL, a tissue that is not easily accessible for live cell imaging.


References

1.
Dave SS, Wright G, et al. Prediction of Survival in Follicular Lymphoma based on Molecular Features of Tumor-infiltrating Immune Cells. The New England Journal of Medicine. 2004;351:2159-69.
2.
Toriwaki J, Mori K. Distance Transformation and Skeletonization of 3D Pictures and Their Applications to Medical Images. In: Bertrand G, Imiya A and Klette R, eds. Digital and Image Geometry. Lecture Notes in Computer Science series. 2001;2243:412-29. Springer-Verlag.