A pathology slide showing samples from a DCIS tumor (left) and an invasive breast cancer.
The way tumors are genetically analyzed in the lab today could not have been anticipated even a few years ago. In addition to providing information about the genes that are altered in cancers, these cutting-edge gene sequencing techniques have allowed scientists to study the development and evolution of cancers at the level of individual cells. To perform some of these methods, however — including the sequencing of individual cancer cells — investigators need high-quality tissue samples in which the DNA is well preserved.
Memorial Sloan Kettering has an incredible storehouse of patient samples going back decades. Because they are tied to the long-term clinical records of individual patients, they have the potential to be extremely valuable for research. Specifically, they could help to uncover the genetic changes that occurred early in a cancer’s history, when it was first discovered, and determine the influence those genes had on how the patient ultimately fared.
But the majority of these older samples were preserved not for genetic analysis, but so that pathologists could view the cells under a microscope to make a diagnosis. They’ve been treated with a chemical preservative called formalin and fixed in paraffin wax. As it turns out, this technique degrades the DNA and makes the analysis of individual cells from tumors incredibly challenging.
MSK experimental pathologists Jorge Reis-Filho and Britta Weigelt have found a way to extract the DNA from the cells in these samples and repair it.
“To be able to study individual cancer cells from these damaged samples, we’re borrowing techniques from the field of forensics,” says Dr. Reis-Filho, the senior author of a new study published February 6 in Nature Medicine. “Once the DNA is restored, we can use the same sequencing methods that we use for the analysis of cells from fresh or frozen samples.”
Looking for Genetic Clues to Breast Cancer Behavior
Dr. Reis-Filho is an expert in the genetics of breast cancer. Because breast cancers tend to be heterogeneous — meaning that within a tumor, there may be different subtypes of cells, each with their own set of faulty genes — he uses a technique called single-cell analysis. Such research enables his group to identify genes in individual breast cancer cells that trigger them to progress from noninvasive to invasive, and to be particularly aggressive.
One goal of his work is to find ways to more accurately characterize a form of breast cancer called ductal carcinoma in situ (DCIS). DCIS is a very early-stage, noninvasive cancer, but people who develop it have about a 25% chance that their cancer will come back after surgery and become invasive.
Finding ways to predict which DCIS tumors are the most dangerous could help patients with low-risk disease avoid unnecessary therapies. To do this, though, researchers need to learn more about the genetic changes that occur early in the development of these tumors and how these tumors evolve. Because DCIS takes a long time to progress, they need to study samples old enough that the ultimate outcome is known.
A Daring Proposal
Dr. Reis-Filho believes that single-cell analysis may be the best way to understand the development and progression of DCIS. This knowledge is essential to predict the genes connected to the aggressiveness of the cancer, but this technology hasn’t been possible to use with samples that have degraded DNA.
Luciano Martelotto, a former postdoctoral research fellow in Dr. Reis-Filho’s lab, decided to look for a way around this problem. Dr. Martelotto was one of the lead authors of the new study, which also included contributors from several other institutions.
“About three years ago, Luciano came to me and said that he wanted to develop a method for doing single-cell analysis of samples that had been fixed in formalin,” Dr. Reis-Filho says. “I thought his efforts had very little chance of success, but today I’m humbled because together we have created the first-ever method to be able to do just that.”
This research is opening up massive archives of stored tissue samples with known long-term disease and treatment information, says Larry Norton, MSK Deputy Physician-in-Chief for Breast Cancer and one of the study’s coauthors.
“This science eventually might help us identify cases of DCIS that would never turn into aggressive cancer even if left alone, so that we do not have to operate on or use radiation therapy for these patients,” Dr. Norton says. “In addition, by giving us more information about how precancer may turn into cancer, it reveals opportunities for prevention.”
Advancing a Cutting-Edge Technology
Although it’s an important laboratory technique, single-cell analysis is still very expensive, and it’s not yet part of clinical practice. The researchers are currently working on new methods that would bring the cost down enough to eventually make the analysis possible in the clinical setting. At the same time, they also plan to continue their research to learn as much as they can about the genes that prompt DCIS to progress to invasive cancer.
Dr. Reis-Filho says the new technology has applications for studying pathology materials from other early-stage forms of cancer and precancer as well, including lobular carcinoma in situ (another form of noninvasive breast cancer precursor) and Barrett’s esophagus (a condition that often progresses to esophageal cancer).
