Exosomes as potential biomarkers of radiation exposure

Exosomes = potential biomarkers of radiation in the Read more

Before the cardiologist goes nuclear w/ stress #AHA17

Measuring troponin in CAD patients before embarking on stress testing may provide Read more

Virus hunting season open

Previously unknown viruses, identified by Winship + UCSF scientists, come from a patient with a melanoma that had metastasized to the Read more

Cancer

Questions only a network of pathologists can answer

When a patient is fighting a brain tumor, pathologists usually obtain a tiny bit of the tumor, either through a biopsy or after surgery, and prepare a microscope slide. Looking at the slide, they can sometimes (but not always) tell what type of tumor it is. That allows them to have an answer, however tentative, for that critical question from the patient: “How long have do I have?” as well as give guidance on what kind of treatment will be best.

Dan Brat, a pathologist specializing in brain tumors at Emory Winship Cancer Institute, gave a presentation this week explaining how he has been asking more complicated questions, ones only a network of pathologists armed with sophisticated computers can answer:

  • What genes tend to be turned on or off in the various types of brain tumors?
  • What does the pattern look like when a tumor is running out of oxygen?
  • What if we get a “robot pathologist” to look at hundreds of thousands of brain tumor slides?
Under the microscope, the shapes of cell nuclei in brain tumors look different depending on the type of tumor.

Under the microscope, the shapes of cell nuclei in brain tumors look different depending on the type of tumor.

Brat was speaking at a caBIG (cancer Biomedical Informatics Grid) conference, taking place at the Emory Conference Center this week. caBIG is a computer network sponsored by the National Cancer Institute that allows doctors to share experimental data on cancers. Brat explained that low-grade brain tumors come in two varieties: oligodendrogliomas and astrocytomas. Under the microscope, cell nuclei in the first tend to look round and smooth, but the second look elongated and rough. Kind of like the differences between an orange and a potato, he said.  He and colleague Jun Kong designed a computer program that could tell one from the other. They had the program look through almost 400,000 slides, using resources compiled through caBIG (Rembrandt and Cancer Genome Atlas databases). Sifting through the data, they could find that certain genes are turned on in each kind of tumor.

Imagine a "robot pathologist" that can sift through thousands of images from brain tumor samples.

Imagine a "robot pathologist" that can sift through thousands of images from brain tumor samples.

Daniel Brat, MD, PhD, principal investigator for the In Silico Brain Tumor Research Center

Daniel Brat, MD, PhD, principal investigator for the In Silico Brain Tumor Research Center

Eventually, this kind of information could help a patient with a brain tumor get good responses to those “How long?” and “How am I going to get through this?” questions.

Joel Saltz, who leads Emory’s Center for Comprehensive Informatics, has been a central figure in developing tools for centers such as Emory’s In Silico Brain Tumor Research Center. In September 2009, Emory was selected to host one of five “In Silico Research Centers of Excellence” by the National Cancer Institute.

Posted on by Quinn Eastman in Cancer Leave a comment

Strategies to target cancer stem cells

A story in last Friday’s New York Times highlights research on “cancer stem cells”: a fraction of cells in a tumor that are especially resistant to chemotherapy and resemble the body’s non-cancerous stem cells in their ability to renew themselves.

The story describes work by a team at the Broad Institute, who reported in the journal Cell that they had identified compounds that specifically kill cancer stem cells. The hope is that compounds such as these could be combined with conventional treatments to more effectively eliminate cancers.

However, scientists disagree on whether the phenomenon of cancer stem cells extends to different kinds of cancer and what is the best way to target them. Previously not much was known about how to attack these cells.

Work at Emory’s Winship Cancer Institute has been tracking how some biomarkers in cancer cells resemble or differ from those found in stem cells. These markers may help researchers home in on the cancer stem cells.

 

Anticancer therapy must target more than one type of cell. TIC means tumor initiating cell, DTC means differentiated tumor cell, and CPG means cancer progenitor

If "cancer stem cells" play the critical roles some scientists think they do, anticancer therapy must target more than one type of cell. In this figure from Van Meir + Hadjipanayis' review, TIC means tumor initiating cell, DTC means differentiated tumor cell, and CPG means cancer progenitor cells.Â

 

 

In a recent review, Emory brain cancer specialists Erwin Van Meir and Costas Hadjipanayis write:

The “cancer stem cell” hypothesis has invigorated the neuro-oncology field with a breath of fresh thinking that may end up shaking the foundation of old dogmas, such as the widely held belief that glioblastoma tumors are incurable because of infiltrative disease. If the infiltrated cells are in fact differentiated tumor cells, their dissemination beyond the surgical boundary may not be the primary cause of tumor recurrence.

Van Meir, the editor of a new book on brain cancer, adds this comment:

Clearly a lot more work needs to be done to understand the precise cause of glioblastoma recurrence after surgery and chemotherapy and how to prevent it.  The possibility of developing therapeutics that can specifically target the brain cancer stem cells is an exciting new development but will have to proceed with caution to spare normal stem cells in the brain. Developing new imaging tools that can track cancer stem cells in the brain of treated patients is also an important objective and some of the Emory investigators are evaluating the use of nanoparticles to this purpose.

A new faculty member at Winship, Tracy-Ann Read, recently published her research on a molecule that could be used to identify “tumor-propagating cells” in medulloblastoma, a form of brain cancer. She says:

Although cancer stem cells have been identified in many different types of cancer, it is becoming increasingly clear that the properties of these cells may vary greatly among the different tumor types. It is unlikely that one  therapeutic agent will be able to target the cancer stem cells in for example all types brain tumors. Hence  much work still needs to be done in terms of analyzing the properties of these cells in each tumor type and identifying the genes that are responsible for their unique ability to propagate the tumors. 

Winship’s director Brian Leyland-Jones has also reported at the San Antonio Breast Cancer Symposium that molecules that distinguish a hard-to-treat form of breast cancer resemble those that maintain stem cells.

Nice round-up from Nature’s stem cell blog editor Monya Baker

Posted on by Quinn Eastman in Cancer Leave a comment
« Previous   1 2 3 4 5 6 7 8 9