Hot new technologies and methods are making leaps forward in studying individual cancer cells—which will help prevent drug resistance to therapy and identify new drug targets.

May 14, 2014 — At the 2014 American Association for Cancer Research (AACR) Annual Meeting in San Diego held in April, a session titled “Single Cell Analysis of the Tumor” was dedicated to advances in technologies and understandings of tumor cell biology, on a single cell level. Why would researchers look to fight cancer at a single cell level rather than taking on the whole tumor? Think of it this way: imagine gazing at a picture of huge crowd of people, shunted into a tight space, parts overlapping, faces blurred. You will be able to say: this is where the people were standing at the time the shot was taken—in a parking lot perhaps. And you might be able to know the size of the crowd– generally. But what could you learn about each individual in the snapshot, in such a grouping? What car do they drive? How many kids do they have? What did they eat for breakfast? Instead, what if you zoom your camera in on each person in the picture, and then observe them, say in the check-out line at Trader Joe’s. Obviously only the second method would tell you important details about each person. For example, someone buying up applesauce and juice boxes most likely has young kids at home. And that fella with lemon-iced cake, ice cream, and chocolate-covered everything, has a serious sweet tooth.

The same principles apply to studying prostate cancer biology. The group shot might tell you what the entire population of cells are doing at the 30,000 foot level, but what does it say about the specific nature, make-up and appetites of individual cells? Certainly, not enough. Populations of cells, especially cancer cells, differ greatly. This heterogeneity—both genetically (as cancer cells divide, each can acquire new and unique mutations), and functionally (even cells that are identical in genetics or appearance may not be synchronous in their activities at every moment in time)—make curing many cancers, especially certain prostate cancers, especially difficult if doctors have only the view from 30,000 feet to use, to prescribe treatments.

The future of cancer care is hurdling towards precision medicine, where a cancer is treated based upon the unique attributes of each person’s tumor, rather than prescribing the same treatment to everyone, just because it works for some. Further refinement and efficacy of precision medicine requires developing and implementing new technologies and methods that enable a deeper understanding of individual cancer cells to identify those most likely to cause cancers to grow and spread. Some of the most exciting new ideas were discussed in this session at the AACR.

Single cell analysis of tumor biopsies

Dr. James Hicks, of the Cold Spring Harbor Laboratory in New York, studies the genomes of individual prostate cancer cells, in order to understand how tumor populations evolve during disease progression and become resistant to therapies. He uses a technique called “single nucleus sequencing” (SNS), in which the nuclei– the genetic center– of tumor cells are isolated from biopsy or radical prostatectomy tissues and individually sequenced.

In his presentation, Hicks emphasized that knowing where in the prostate the tumor samples were taken from is important. Understanding the evolution, or progression, of cancer cells, can lead to new treatments to halt that process. Tumor evolution can be envisioned anatomically by linking tumor cell genetics with pathology and location. In studying tumors from patients with different stages of disease, Hicks and colleagues found that Gleason grade was linked with clonality – meaning more aggressive tumors (higher Gleason grade) were made up of cells that had expanded from a common tumor cell ancestor or clone (and thereby share their parent’s key oncogenic mutations), while lower Gleason grade tumors (≤ 6) had no or very low clonality, i.e., the cells that make up the tumor are of a more divergent ancestry. However, men with low Gleason scores did have a large number of chromosomal rearrangements. This indicates that during early disease stages, different cancer cells continue acquiring unique mutations, effectively “testing out the waters,” until a mutant clone with highly aggressive potential is born and becomes the “founding seed” of an aggressive tumor.

Studying circulating tumor cells to understand drug resistance

Circulating tumor cells (CTCs) are tumor cells that have been shed into the circulation from primary and metastatic tumor sites and can be detected by blood tests. Recent clinical trials have found that CTC numbers are predictive of patient outcome and response to therapies, and a movement by the scientific community is underway to establish “liquid biopsies” (where CTC numbers are evaluated from blood draws), as a standard non-invasive method of patient monitoring. CTCs can also be studied as avatars, representing the tumors that are elsewhere in the body, to gain insight into various characteristics including genomics and drug sensitivities.

Dr. Peter Kuhn of the Scripps Research Institute in La Jolla, CA, has been analyzing the appearance, gene expression, and genomic mutations of CTCs from prostate cancer patients over the course of treatments, to study the evolution of drug-resistance mechanisms. He presented several patient case studies as examples of his findings.

In one patient, CTCs were assessed from blood draws taken before, during, and after chemotherapy and subsequent treatment with the androgen-deprivation therapy Zytiga (abiraterone acetate), which inhibits production of the androgens that fuel prostate cancer. The CTCs were examined for expression of the androgen receptor (AR), the primary driver of prostate cancer. This patient initially responded to treatment with Zytiga,–and AR expression was subdued– but then, rapidly developed drug-resistance and tumor recurrence. In recurrent Zytiga-resistant CTCs, AR expression had been regained in the entire CTC population. When the genomes of 42 individual CTCs were sequenced, it became apparent that these Zytiga -resistant CTCs with high AR expression had expanded from a single tumor cell, or clone, with a specific amplification of the AR gene. Identifying and studying the tumor cells that have acquired drug-resistant mutations and initiate recurrent disease, will allow researchers to better target such lethal tumor cells for destruction.

A new technology to capture circulating tumor cells –even those that have shed an important cancer cell marker, known as EpCAM

Current CTC capture technologies, such as the method used by Kuhn, detect and isolate pure CTCs from the blood, based on expression of an epithelial cell marker, EpCAM (expressed on cancer cells), and by excluding immune cells which express an immune cell marker, CD45. Features of individual tumor cells can then be studied to understand tumor cell biology and treatment issues such as drug sensitivity and resistance. However, expression of EpCAM can be lost by tumor cell types, including tumor initiating (“stem”) cells and tumor cells that have changed their phenotypes, or ways in which they appear and act, to become more metastatic. So, developing methods to detect CTCs that have lost EpCAM expression is critical to capture a fuller breadth of CTCs.

Dr. Daniel Haber, of Harvard Medical School, and colleagues, developed a new microfluidic device that uses a novel methodology to detect CTCs whether or not they express EpCAM. Immune cells are tagged using tiny magnetic beads stuck to antibodies that recognize proteins expressed by immune but not cancer cells. Termed “CTC-iCHIP,” this microfluidic device first passes cells through a chamber that uses hydrodynamics to remove cells based on size (red blood cells and platelets get taken out). The cells that remain (CTCs and immune cells) then travel through a “wiggly channel” — a series of curves in the device. The physics of moving through wiggly channels arranges the cells into a single-file line. Finally, cells are separated by a “magnetic moment,” where the magnetically-labeled immune cells are deflected from the path traveled by the CTCs, allowing a pure population of CTCs to be collected.

Dr. Daniel Ting, of Massachusetts General Hospital and Harvard Medical School, presented after Haber. He demonstrated the use of wiggly channel technology to isolate and analyze CTCs from mice with pancreatic cancer. He identified three unique populations of CTCs with different functions and genes expressed, supporting the necessity for this method of CTC isolation, where different and possibly more tumorigenic CTC types can be assessed.

New “high-dimensional” technologies that are revolutionizing tumor biology

Dr. Gary Nolan, of Stanford University, discussed advances in two new technologies that allow a large number of factors to be examined for single cancer cells. Time of Flight Mass Cytometry (CyTOF) is a new technology that can evaluate multiple factors, or parameters, in a single cell, simultaneously. The previous theoretical limitation for CyTOF was 100 parameters per cell, but Nolan announced that he and colleagues have pioneered new methodologies that will allow measurement of up to 500 parameters at a time, including the amounts of mRNA molecules. (Previously only proteins could be evaluated.) His group has also developed several supporting bioinformatics applications that process and visualize the “high-dimensional” volumes of data the new technology captures, helping scientists to make sense out of chaos. Understanding how multiple factors work together inside a cancer cell (and as compared with regular cells) will help scientists to identify which molecules can be targeted by therapies to kill the cancer cells.

Nolan presented a study on acute myeloid leukemia (AML) cells from pediatric patients, that were treated with stimulatory proteins called cytokines, and then examined using CyTOF for 45 parameters which included: 1) expression of cell surface proteins that identify a cell’s specific function and type (e.g. different types of white blood cells), and 2) activation levels of intracellular signaling proteins that measure how responsive the cells are to each of the cytokines, indicating how the cells would function under different conditions or in different environments.

This information matters because, Nolan found that patterns of intracellular signaling, but not of cell surface proteins, predicted the presence of certain mutations in the tumor cells as well as patient outcome. AML cells functioned, via intracellular signaling, like hematopoietic (blood-cell forming) stem and progenitor cells, but do not “look” like them—that is, they do not express the same cell surface proteins as stem /progenitor cells. So if scientists only used the surface proteins to guess at what the cells do— which works fine for figuring out many normal cells— their guesses would be all wrong for cancer cells.

Nolan went on to present a ground-breaking new technology that his lab created, which is analogous to performing CyTOF on tissue specimens, called multiplexed ion beam imaging (MIBI). Nolan is working to perfect this technology, so that MIBI can be performed in clinical laboratories across the country. This would enable high-dimensional analyses of gene expression and protein functions to be rapidly performed on cells from cancer patients’ tissues in a standardized fashion. Like CyTOF, over 100 features could be simultaneously analyzed in a single cell, an undoubtable revolution, considering that current (immunohistochemistry) technologies that examine the levels of molecules in tissues can simultaneously detect only up to a half-dozen parameters. “Pay attention to anomalies,” Nolan said, referring to the significance of studying the outlier biology of individual cells, as even a single abnormal cell can lead to total havoc. “They [outlier cells] know more about reality than you do,” said Nolan.

Overall, these exciting new studies and technologies will add significant insight into how tumor cells behave on an individual level. Such single cell studies are critical to understanding how disease pathways work together to give rise to tumors, and how some tumors evolve to become aggressive and, eventually drug-resistant. A detailed view of how individual cancer cells look and behave is revealing a mosaic of cancer cell colors and personalities. Scientists are moving toward studying cancer cells as a collection of rainbows and personality types instead of blending them together into murky shades of grey– cells that all punch into the same job. Knowing the unique things that make each cancer cell tick may allow researchers to shut down their time clocks.

Dr. Andrea Miyahira has a PhD in cancer immunology, and is Manager of Scientific Programs at the Prostate Cancer Foundation.