Livermore program accelerates cancer diagnostics

A former Lawrence Livermore National Laboratory (LLNL) chemist is attempting to commercialize an accelerator-based measurement technology, developed at the lab, that predicts whether some cancerous tumors will respond to a widely used chemotherapy agent. If the test is approved for use, patients who won’t respond to the drug would be spared from having to endure the toxic side effects and could pursue other therapies.

The technology employs radiocarbon tracing using accelerator mass spectrometry (AMS) to determine if the chemotherapy agent carboplatin will be effective in treating individuals with bladder and lung cancer. Carboplatin is effective in nearly two-thirds of patients with those cancers, but there is no currently approved method for determining a priori who will and will not respond.

In current clinical trials, lung and bladder cancer patients are given a microdose of carboplatin tagged with carbon-14, says Paul Henderson, CEO of the LLNL spinoff Accelerated Medical Diagnostics. Biopsies of the tumors are taken 24 hours later, and the samples are analyzed at LLNL’s accelerator mass spectrometer to determine the extent to which the drug with the ¹⁴C has bound with cancer-cell DNA. Trial patients then undergo a course of treatment with carboplatin. So far, the patients who showed the highest level of drug–DNA binding from the microdose responded positively to the carboplatin, Henderson says.

To finance the trials, the company has pulled together grants from the National Institutes of Health, the Susan and Gerry Knapp Family Fund, NSF, the Department of Energy, and the University of California, Davis, Comprehensive Cancer Center. All told, about $3 million has been spent on developing the technology. Henderson expects that between $30 million and $50 million will be needed to gain Food and Drug Administration approval. In addition to completing a second, larger round of clinical trials, the company must build and obtain quality certification for a facility to manufacture the radiocarbon-labeled microdoses. That facility will include a purpose-built accelerator mass spectrometer.

Financing for the second set of trials and the manufacturing facility will be raised from venture capital and other private equity sources, says Henderson. With 300 000 cancer patients annually undergoing treatment with drugs similar to carboplatin, he says he doesn’t expect trouble attracting investors. The company recently received FDA approval for trials to assess radiocarbon microdosing as a diagnostic for another chemotherapy drug, oxaliplatin, in the treatment of metastatic breast cancer. About 20% of patients with that disease will respond favorably to the drug, he says.

From brains to toxins

There are other biomedical applications for AMS. It has been used to conduct metabolite analysis at the picomole to attomole level and to identify macromolecular targets such as proteins for drugs. Using a different ¹⁴C methodology, researchers at LLNL and Sweden’s Karolinska Institute reported in June they had found evidence that new neuron formation occurs in the hippocampus region of the brain. Previously, using the same ¹⁴C technique, scientists at the two institutions had determined that neurogenesis does not occur in either the cortex or cerebellum regions of the brain. Other applications of AMS in the biological sciences include the investigation of tissue turnover rates and the pharmacokinetic parameters of drugs, toxicants, and nutrients, including their absorption, distribution, metabolism, and excretion.

The neurogenesis research relies on the spike in ¹⁴C levels that resulted from atmospheric testing of nuclear weapons (see the Physics Update on page 18 ). Since aboveground tests were banned by treaty in 1963, ambient levels of ¹⁴C have steadily declined. When a cell divides, the newly synthesized DNA incorporates a trace amount of ¹⁴C that is proportional to the environmental level at the time of mitosis. The radioactive carbon in a cell nucleus can thus be used as a time stamp of when the cell was born.

Several NIH institutes provide a total of about $3.5 million annually to LLNL’s AMS program, including a core grant of $2 million a year to support a user facility. But Graham Bench, the AMS program director, says that lasers may replace AMS for biomedical applications by the end of the decade. “It’s a fundamentally different way to quantify ¹⁴C,” Bench says. The sample is combusted in excess oxygen to oxidize the carbon to carbon dioxide, he explains. The CO₂ can be separated from other gases cryogenically. “You shine a laser beam having a specific wavelength that gets preferentially absorbed by ¹⁴C. The amount of that absorption can tell you the amount of ¹⁴C content in the CO₂ sample.”

A laser-based device could lower the cost of equipment needed for analyzing radiocarbon tracers by a factor of 10, to as little as a few hundred thousand dollars. And the instrument might be small enough to fit on a tabletop, Bench says. He adds that the expertise and infrastructure required to operate an accelerator and associated equipment would be eliminated. Researchers from LLNL are working with Picarro, a company that manufactures spectroscopic instruments for methane leak detection, to develop a tabletop version of the laser device. Data from a prototype instrument have been promising, he says. Other entities pursuing the technology include Sweden’s Lund University, Rutgers University, Italy’s National Institute of Optics, and the European Laboratory for Non-Linear Spectroscopy, also in Italy.

“The overarching vision would be one in every clinic, one in every research lab,” says Bench. “Once you make the instrumentation cheaper and easier to operate, the number of people who want to use it will rapidly increase.” There are hundreds, if not thousands, of scintillation counters in the US measuring radiocarbon decay, he notes. “Think about something that size and that easy to use, but with the sensitivity of AMS.”