Conference name
Exhibitors & Sponsors

Torsten Nielsen

Ki67 as a biomarker of proliferation


(Second abstract and handout, see below)

Increased proliferation is a fundamental feature of most cancers, and its measurement has clinical validity for prognosis, prediction and as an intermediate endpoint in neoadjuvant treatment. Thousands of genes and proteins are associated with the process of cell proliferation. Among these, Ki67 has been selected as particularly useful for immunohistochemistry, despite uncertainty as to its specific biological role. It is valuable because of its expression throughout the cell cycle, and its particular structure with epitopes that allow for highly sensitive and specific immunohistochemical detection. Despite this, preanalytical and analytical factors in Ki67 staining have not been rigorously assessed for clinical use in many settings, at least not in a manner that would meet EGAPP criteria for level 1 proof of analytical validity. In breast cancer, efforts have been made to standardize tissue handling and interpretation to achieve this level of evidence for ER and HER2. Several gene expression profile-based tests are in active use to assess breast cancer risk and to help make critical decisions about the need for chemotherapy, but these tests are expensive and not as widely accessible as immunohistochemistry. There is therefore considerable clinical interest in using Ki67 IHC stains for this purpose (among others). The Breast International Group and the North American Breast Cancer Group have struck a working group to assess Ki67 as a clinical biomarker in breast cancer, and have promulgated standards for preanalytical handling and for staining methodology. Scoring methods, on the other hand, have lacked consensus in the scientific community. A QC study among participating labs in the working group revealed good intra-observer scoring consistency but worrisome differences in inter-observer scores on the same cases, to a degree that appears to preclude direct transposition of clinical decision-making cutpoints between labs. As a result, a web-based calibration tool was developed in an effort to standardize scoring and provide a proficiency assessment test. Results will be presented from an ongoing study into the ability of this approach to improve inter-observer consistency of Ki67 assessment in breast cancer, with an update on the current status of efforts to demonstrate high level clinical utility for this biomarker.

Analytically robust expression profiles from FFPE sections using Nanostring technology


High throughput genomic technologies are leading to the identification of complex patterns of changes in cancer DNA and RNA, which have clear potential for improving diagnosis, prognosis and prediction in cancer care. Gene expression profiles, originally derived from microarray-based research, are the furthest along in clinical development. In breast cancer, several signatures have come into clinical use, most of which were developed as supervised signatures for risk profiling in particular clinical settings. Unsupervised analyses identify breast cancer intrinsic subtypes as the major biological gene signatures consistently found across patient datasets, and these molecular subtypes have become part of the St. Gallen consensus about how to interpret breast cancer, although methods for clinical detection are not yet standardized. The Strategic Partnering to Evaluate Cancer Signatures initiative of the NIH funded work to translate intrinsic subtypes into clinical tests, and the most widely applicable forms of testing are those which can be applied to formalin-fixed, paraffin embedded tissue. Genomic tests that work on FFPE material can not only be easily integrated into standard laboratory work flow, but also can be applied to archival cohort and clinical trials material to demonstrate clinical validity in prospective-retrospective research designs that can give a high level of evidence without waiting a decade for results. For this purpose, the intrinsic subtypes were distilled into a 50 gene signature (PAM50), originally designed to be detected by qRT-PCR, which can be made to work on FFPE specimens with careful attention to primer design issues. The recent development of Nanostring technology, based on multiplexed color-coded probe pairs, has provided a new platform for RNA quantification on FFPE tissue, with particular advantages: excellent performance on FFPE, no need for enzymatic amplifications, direct digital counting of RNA species, robustness to low template quality and quantity, a wide range of linear detection capacity, ease of use and consistency of results. PAM50 has been adapted to the Nanostring platform, a technology which has recently become available in Denmark and which can be run in research or clinical modes. Crucially, this methodology is designed to be installed and applied in local hospital laboratories, rather than requiring specimens to be sent out-of-country. Recent results demonstrating the analytical reproducibility and clinical validity of Nanostring-based PAM50 testing in multiple settings will be discussed.


Torsten O. Nielsen is a clinician-scientist and Professor of Pathology and Laboratory Medicine at the University of British Columbia. Of Danish descent, Dr. Nielsen was born in Vancouver, Canada and completed the combined MD/PhD program at McGill University, with his thesis work concerning human DNA replication biology. During his residency training in Anatomical Pathology, he undertook elective research at Stanford University that led into his subsequent research programs translating gene expression profiles and microarray work into clinical care, for breast cancers and sarcoma. In sarcomas he helped develop DOG1 and TLE1 as novel diagnostic biomarkers and ran basic science projects elucidating the biology of synovial sarcoma and tenosynovial giant cell tumor, work which has led to clinical trials of targeted therapies. In breast cancer, Dr. Nielsen pioneered the development of immunohistochemical surrogate markers for basal-like breast cancer and has worked to develop refined panels incorporating ER, PR, HER2 and Ki67 that identify the intrinsic molecular subtypes. With US-based colleagues, he also developed the PAM50 gene expression panel for breast cancer intrinsic subtype, and worked to convert this into clinical tests applicable to paraffin-embedded blocks using qRT-PCR and more recently Nanostring technology. He has won several awards for his research, including young investigator of the year from the Canadian Association of Pathologists and the Canadian Society for Clinical Investigation, and the basic science award from the Canadian Breast Cancer Symposium. Dr. Nielsen is an executive member of correlative science and pathology committees at the NCIC-Clinical Trials Group and the Alliance for Clinical Trials In ONcology. Currently, he is directing a joint project of the Breast International Group and the North American Breast Cancer Group concerning the standardization of Ki67 testing and scoring. Dr. Nielsen will be presenting two talks: one on Ki67 as a breast cancer biomarker, and the other on the intrinsic subtyping using the Nanostring platform.

Exhibitors & Sponsors



Ultimate Web A/S