The NCI-60: Assessing drug effectiveness
by Richard Folkers
For centuries, venoms, toxins, and other medicinals, which are chemicals extracted from plants and animals, as well as seemingly countless other natural products, have been reputed to have therapeutic properties. In some cases, those reputations have proved true. For example, the cancer drug paclitaxel, commonly known as Taxol, which has been mostly commonly used in the treatment of lung, ovarian, and breast cancers is a medicinal derived from the bark of the Pacific yew tree.
But how does science figure out where legend ends and treatment effectiveness begins? How do you determine whether a drug—from a natural product or engineered as a synthetic chemical, for that matter—is useful as a cancer treatment? It's just common sense that biomedical researchers can't recruit a bunch of cancer patients, give them some tree bark to munch on, and see if they get better. Science is about investigation, about testing, about rigid standards of patient protection, and about building a foundation of evidence. The key question for scientists is: Before starting tests in humans, how do you measure whether a drug has the potential in cancer patients to inhibit tumor growth or prevent cancer?
It's all about chemistry. Natural products can vary widely from sample to sample. Just as the color of blossoms can vary from azalea to azalea in a garden, the strength of a plant substance suspected of anti-cancer properties can be different from plant to plant—or tree trunk to tree trunk. In most cases, the chemical in question, known as the lead compound, must be isolated and purified, ensuring that the strength tested remains consistent from sample to sample. Natural compounds are often extremely complex in their chemical makeup, so if enough chemical can't be extracted naturally, synthesizing lead compounds can be a tall order. But even after that time-consuming and expensive process is complete, the question often remains: Does it work?
For decades, lead compounds were tested principally in mice. The downsides were time, expense, and limited accuracy. Enter NCI's In Vitro Cell Line Screening Project, better known as the NCI-60, a protocol that makes it possible to analyze the anti-cancer properties of a compound in human tumor samples from 60 different cell cultures, sometimes referred to as lines, representing leukemia, melanoma, and cancers of the lung, colon, brain, ovary, breast, prostate, and kidney. The NCI-60 project, which has been testing lines since 1990 in the Developmental Therapeutics Program of NCI's Division of Cancer Treatment and Diagnosis, screened 17,200 compounds in 2011, roughly evenly divided between natural and synthetic agents.
For natural products, the process begins with specimen collection, often in far-away countries or oceans, and then transitions to an NCI laboratory in Frederick, Md., where technicians take a ground-up sample and add solvents, creating two extracts: one that can be dissolved in water and one that is not water-soluble. The extracts, which may well contain 50 to 100 or more chemical compounds, are then tested, one chemical at a time, for cancer cell-killing or cell-inhibiting potential in the NCI-60 cell lines. The most promising—the "hits"—move on to further testing.
Those hits, which account for about 2 percent of samples tested, are sent to an isolation laboratory, in a drive to identify precisely which chemical compounds are responsible for the biological activity of the sample. Researchers use a process known as bioactivity-driven chromatography to separate individual chemical compounds from the extract; they then employ state-of-the-art instrumentation to define chemical structures; ultimately mass spectrometry determines the type of the compound based on its molecular weight. Finally, nuclear magnetic spectroscopy shows how atoms within a molecule are linked together.
To further that effort, NCI recently announced the rollout of CellMiner, a web-based suite of genomic and pharmacologic tools to analyze data and explore patterns from NCI-60.
Since 1990, more than 100,000 natural products have gone through the NCI screening process, driving the number of drugs in NCI's repository that have had some kind of screening process to over 400,000. While only a tiny fraction move into further testing, the library of tested compounds remains an important source of information, allowing scientists to compare a potential cancer-associated molecular target in the human body against many thousands of compounds.
In August 2012, one prime study example, published in PNAS, showed that a single gene, Schlafen-11 or SLFN11, sensitizes cells to substances known to cause irreparable damage to DNA. The scientists analyzed nearly 1,500 compounds commonly tested in the NCI-60 and identified three common chemotherapy drugs that might be effective against tumors expressing the SLFN11 gene.
Compounds can enter NCI's testing process at many points in their purification and development process. In some cases, those compounds could even be drugs that have previously been rejected in a different form, due to toxicity issues or other problems the first time they were analyzed.
Importantly, researchers from outside of government can submit compounds—both synthetic compounds and purified natural products—to NCI for screening and for assistance down the road in testing that may lead to clinical trials in cancer patients. There is no cost to the researcher for NCI-60 screening.