NSBCC

Project 5

Nanotechnology Tools for a Pathologic Analysis of Glioblastoma (brain cancer) and Other Cancers
Browse subjects in this section:

NANOTECHNOLOGY TOOLS FOR A PATHOLOGIC ANALYSIS OF GLIOBLASTOMA (brain cancer) AND OTHER CANCERS

Consider the images below.  These are positron emission tomography (PET) scans of two patients that were identically diagnosed by the ‘gold standard’ of medicine, pathology, to have identical disease – gastrointestinal cancer – or GIST.  PET, which was invented by NSBCC co-Director Michael Phelps, is an FDA-approved diagnostic tool for most cancers.  PET is an imaging technique that doesn’t look at the static structure of the body, such as MRI or CT imaging, but instead images metabolism.  PET-based technologies are a key part of the NSBCC, and they are discussed in Project 3.  Briefly, cancerous tumors exhibit a rapid rate of growth, and so certain metabolic processes are greatly accelerated in tumors – in this case, it is the amplified metabolism of glucose by the tumors that makes them stand out as dark patches against the light grey background of non-cancerous tissues.  Both patients clearly have cancer.

Based upon the pathologic diagnosis of GIST, both patients were given a drug called Gleevec.  Most cancer drugs are chemotherapies, which means that they are poisons, which means that they kill all tissues, but hopefully the kill the cancer tissues fastest.  Gleevec is different.  It is designed to attack a particular cellular processes that are critical to the cancer, and thus ‘shut down’ the tumor.  As can be seen the images, Gleevec performed spectacularly the patient represented by the top image, but had no effect on the patient represented by the bottom image.  Related drugs, that are just now emerging into the world of cancer therapy, are discussed in more detail in Project 4.

The inevitable conclusion is that these patients had different diseases, and the current practice of pathology could not discriminate between these diseases.  This is not unique to GIST.  In fact, just about all cancers, such as cancers of the breast, prostate, brain, etc., are, in fact, multiple diseases – each arising from different genetic and environmental mutations, each requiring a different treatment, and each with different expected outcome.  The goal of this NSBCC Project is to develop tools that can be utilized by pathologists for discriminating between these different diseases for a variety of cancers, and to therefore provide guidance in the administration of proper therapies.  A further goal is to develop these tools so that they can yield such a diagnosis with minimal invasion to the patient (i.e. through analysis of materials extracted from a skinny needle biopsy rather than through analysis of a surgically removed section of a tumor). A final goal is to utilize these tools to understand the heterogeneity of tumors by analyzing microscopic quantities of the tumor.



The current model of pathology practice for cancer diagnosis involves microscopic examination of diseased tissue, typically removed using surgery. Based on tissue architecture patterns and the morphological appearance of the cells, as well as on the presence or absence of a limited set of protein markers, the pathologist renders a broad diagnosis typically conveying tumor type and grade (a measure of the biological aggressiveness of the tumor). Risk stratification is then based on this morphological diagnosis and the patient is typically treated with relatively toxic, non-specific chemotherapies and radiation. For most cancer patients, the results are not satisfactory.  Cancer is currently the #1 killer of adults in the U.S.

Cancer – a disease of complexity: No two individual tumors are identical. In contrast, a systems biology approach to cancer aims to define the protein and gene biomodules and networks that are responsible for the emergent properties of cancer (i.e. their proliferative capacity, their metastatic potential, their resistance to therapies, etc.) (see Project 1 discussion).  By capturing information about relationships between key elements of the system, commonalities between highly individual cancers can be defined and targeted. A major challenge in improving such pathology involves redefining the relatively non-specific nature of the pathologic diagnosis itself. Just as each cancer stratifies into a set of diseases, each of those diseases is characterized by its own molecular signature. Traditional pathological examination cannot distinguish these relevant tumor subsets, because they are usually microscopically identical.

A NEW APPROACH TO CANCER PATHOLOGY  (see figure below)                The application of high-throughput genomic and proteomic approaches to cancer has begun to resolve the diversity of cancers at the molecular scale. New approaches for specifically targeting proteins or pathways altered in cancer cells are now available as diagnostic markers and therapeutic targets, as are the glioblastoma-specific transcripts that will emerge from Project 1. Their success as diagnostic markers and clinical agents depends largely on identifying the most informative molecular subsets of the most effective markers and subsets.  Currently, patient inclusion in clinical trials is highly reliant on histological classification, which provides only limited insight into the molecular heterogeneity of cancer.  Thus, potentially effective treatments that may be of benefit to specific patient subsets (which are not detectable by histology), will not be recognized. With the integration of molecular analysis, heterogeneous groups can be identified to allow patient stratification.



Glioblastoma – the cancer targeted by this project: Glioblastoma is the most common malignant brain tumor of adults, and is among the most lethal of all cancers. Despite aggressive surgical approaches, optimized radiation therapy regimens and a wide variety of cytotoxic chemotherapies, the median survival of glioblastoma patients is one year from the time of diagnosis.

The current model of pathology diagnosis for brain tumors is based on the hypothesis of Bailey and Cushing (rendered in 1928) that brain cancers could be classified by their microscopic resemblance to a presumed CNS cell of origin or its developmental precursor. This model remains the guiding principal for brain tumor classification. This classification and grading system has proven to be quite useful for predicting the overall survival for groups of brain tumor patients. However, it provides relatively limited insight into the underlying molecular networks that are perturbed by the disease. Further, clinically relevant subsets that may differ significantly in their clinical course and response to therapy cannot be identified by the current classification system.  Clearly, new approaches to the understanding and treatment of glioblastoma are needed.


 

For more information on Glioblastoma, see:

Mischel, P. S., Nelson, S. F. & Cloughesy, T. F. Molecular analysis of glioblastoma: pathway profiling and its implications for patient therapy. Cancer Biol Ther 2, 242-7 (2003).

Mischel, P. S. & Cloughesy, T. F. Targeted molecular therapy of GBM. Brain Pathol 13, 52-61 (2003).

http://www.abcsquared.org/   (ABC2 for Accelerate Brain Cancer Cure)

 

Related Categories: Project 5 Publications

Grab the RSS newsfeed for this category:
XML feed
HomeAbout NSBCCPeopleProjectsContact UsLogin
© 2005, NSBCC: NanoSystems Biology Cancer Center
Caltech Chemistry 127-72 1200 East California Blvd.
Pasadena, CA 91125